HomeMy WebLinkAboutBradley Lake Final Supporting Design Report Vol 3 1988Alaska Power Authority
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION
CONTRACT
BRADLEY LAKE
HYDROELECTRIC PROJECT
FEDERAL ENERGY REGULATORY COMMISSION
PROJECT NO. P-8221-000
VOLUME 3
DAM AND SPILLWAY
STABILITY ANALYSIS
Prepared By
STONE & WEBSTER ENGINEERING CORPORATION
MARCH 1988
TABLE OF CONTENTS
TABLE OF CONTENTS
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 1 -REPORT
VOLUME 2 -DESIGN CRITERIA
VOLUME 3 -DAM AND SPILLWAY STABILITY ANALYSIS
VOLUME 4 -CALCULATIONS
VOLUME 5 -CALCULATIONS
VOLUME 6 -CALCULATIONS
VOLUME 7 -CALCULATIONS
VOLUME 8 -CALCULATIONS
VOLUME 9 -CALCULATIONS
0216R-4460R/CG 1
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 1
REPORT
1.0 INTRODUCTION
2.0 DESIGN AND GENERAL TECHNICAL DATA
2.1 DESIGN
2.2 DESIGN LOADS
2.3 STABILITY CRITERIA
2.4 MATERIAL PROPERTIES
2.5 GENERAL TECHNICAL DATA
3.0 SUITABILITY ASSESSMENT
3.1 SPECIFIC ASSESSMENTS
4.0 GEOTECHNICAL INVESTIGATIONS
4.1 CHRONOLOGY OF INVESTIGATIONS
4.2 BORING LOGS, GEOLOGICAL REPORTS AND LABORATORY TEST
RESULTS
5.0 BORROW AREAS AND QUARRY SITES
5.1 BORROW AND QUARRY AREAS
5. 2 OTHER MATERIAL SOURCES
6.0 STABILITY AND STRESS ANALYSIS
6.1 GENERAL
6.2 DIVERSION TUNNEL INCLUDING INTAKE STRUCTURE
6. 3 MAIN DAM
6.4 SPILLWAY
6.5 POWER TUNNEL AND PENSTOCKS
6.6 POWERHOUSE/SUBSTATION EXCAVATION, COFFERDAM
AND TAILRACE CHANNEL
6.7 POWERHOUSE
6.8 REFERENCES
7.0 BASIS FOR SEISMIC LOADING
7.1 GENERAL
7.2 SEISMOTECTONIC SETTING
7.3 SEISMIC DESIGN
0216R-4460R/CG 11
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 1
REPORT
8.0 SPILLWAY DESIGN FLOOD BASIS
8.1 STUDY METHODOLOGY
8.2 WATERSHED MODEL CALIBRATION
8.3 PROBABLE MAXIMUM FLOOD
8.4 SPILLWAY DESIGN FLOOD
8.5 MODEL TEST
9.0 BOARD OF CONSULTANTS
9.1 INDEPENDENT BOARD OF CONSULTANTS
9.2 FERC BOARD OF CONSULTANTS
APPENDIX A
Plates
Exhibit F
1
2
3
4
5
6
7
8
9
10
13
14
15
16
17
18
19
20
Figures
F.6.2-5
F.6.2-6
DRAWINGS
Title
General Plan
General Arrangement -Dam, Spillway and Flow Structures
Concrete Faced Rockfill Dam -Sections and Details
Spillway -Plan, Elevations and Sections
Power Conduit Profile and Details
Intake Channel and Power Tunnel Gate Shaft -Sections and
Details
Civil Construction Excavation at Powerhouse -Plan
Civil Construction Excavation at Powerhouse -Elevations
90 MW Pelton Powerhouse
Construction Diversion -Sections and Details
Main Dam Diversion -Channel Improvements
General Arrangement -Permanent Camp and Powerhouse
Barge Dock
Powerhouse Substation and Bradley Junction
Main One Line Diagram
Martin River Borrow Area
Waterfowl Nesting Area
Powerhouse Access Roads
Mean Horizontal Response Spectrum
Design Accelerogram
0216R-4460R/CG 111
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 1
REPORT
APPENDIX B ATTACHMENTS
B.l Construction Schedule Contract Dates
B.2 Meetings of the Independent Board of Consultants
Meeting No. 1
Meeting No. 2
Meeting No. 3
Meeting No. 4
Meeting No. 5
Meeting No. 6
Meeting No. 7
Meeting No. 8
Meeting No. 9
Meeting No. 10
May 12 and 13, 1983
July 11 to 15, 1983
September 25 to 27, 1984
November 4 and 5, 1985 with response of
November 25, 1985
January 28, 1986
May 6 to 8, 1986 with response dated May 21, 1986
August 12 to 14, 1986 with response dated
October 20, 1986
December 8 to 10, 1986
Site Visit by Mr. A. Merritt on December 11, 1986
May 5 to 1, 1987
December 17 and 18, 1987
B.3 Meetings of the FERC Board of Consultants
Meeting No. 1
Meeting No. 2
March 6 and 7, 1986
May 28 and 29, 1986 with response dated July 11,
1986
Hydraulic Model Test of Spillway July 9, 1986
Meeting No. 3 August 18 to 20, 1986 with response dated
October 28, 1986
Meeting No. 4
Meeting No. 5
Meeting No. 6
0216R-4460R/CG
Hydraulic Model Test Spillway and Diversion Tunnel
August 29 and September 25, 1986
January 27, 1987 with response dated January 29,
1987
May 26 to 28, 1987 with response
December 7 and 8, 1987 with response
lV
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 2
DESIGN CRITERIA
1.0 Civil Design Criteria
2.0 Geotechnical Design Criteria
3.0 Structural Design Criteria
Part A General Design Criteria
Part B Special Requirements for Major Structures
Section 1.
Section 2.
Section 3.
Section 4.
Section 5.
Section 7.
Main Dam Diversion
Main Dam
Spillway
Power Tunnel Lining, Intake and Gate Shaft
Steel Liner and Penstock
Tailrace
4.0 Hydraulic Design Criteria
1. Main Dam Diversion
2. Tailrace
3. Hydraulic Turbines, Governors and Spherical Valves
4. Spillway
5. Power Intake, Tunnel and Penstock
5.0 Architectural Design Criteria
0216R-4460/CG v
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 3
DAM AND SPILLWAY STABILITY ANALYSIS
DAM STABILITY REPORT
Section Section Title
1.0 INTRODUCTION
1.1 PURPOSE
1.2 SCOPE
1.3 DAM SAFETY CRITERIA
2.0 DESCRIPTION OF PROJECT FEATURES
2.1 GENERAL
2.2 MAIN DAM
2.3 UPSTREAM COFFERDAM
3.0 DESIGN EARTHQUAKE REGIME
3.1 SEISMOTECTONIC SETTING
3.2 DESIGN RESPONSE SPECTRA
3.3 ACCELEROGRAM DEVELOPMENT
4.0 ALTERNATIVE METHODS OF ANALYSIS
4.1 GENERAL STABILITY CRITERIA
4.2 PSEUDOSTATIC METHOD
4.3 SARMA/NEWMARK METHOD
4.4 FINITE ELEMENT METHOD
4.5 SELECTION OF SARMA METHOD
5.0 SARMA ANALYSIS METHODOLOGY
5.1 MATERIALS PROPERTIES AND EARTHQUAKE SELECTION
5.2 LEASE II ANALYSIS
5.2.1 Static Analysis
5.2.2 Critical Circles and Accelerations
5.3 SARMA ANALYSIS
5.3.1 Data Requirements
5.3.2 Processing
5.3.3 Analytical Output
5.3.4 Significance of Results
6.0 BRADLEY LAKE EMBANKMENT ANALYSES
6.1 EARTHQUAKE RECORDS
6.2 INPUT PARAMETERS
6.3 DESIGN CASES
6.4 LEASE II ANALYSES
6.5 SARMA ANALYSES
6.6 INTERPRETATION OF RESULTS
0216R-4460R/CG Vl
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 3
DAM AND SPILLWAY STABILITY ANALYSIS
Section Section Title
7.0
8.0
6.7
6.7.1
6.7.2
6.7.3
6.7.4
6.7.5
6.7.6
6.7.7
6.7.8
6.8
7.1
7.2
7.3
7.4
SPECIAL STUDIES
Megathrust (a = .55g)
DBE (ah = .375g)
Influence of Downstream Berm
Failed Concrete Face
Varying Embankment Height
Planar Slip Surfaces
La Union Accelerogram
Parametric Analyses
COFFERDAM
CONCLUSION
CRITICAL CASES
SUMMARY OF CRITICAL FAILURE SURFACES
PREDICTED DISPLACEMENTS
RESPONSE TO VARIOUS EVENTS
BIBLIOGRAPHY
LIST OF FIGURES
Figure Title
1 Project Location Map
2 Main Dam Area -General Arrangment
3 Main Dam Sections
4 (Not Used)
5 MCE Response Spectra -Mean and Chosen
6 Rockfill Friction Angles
7 Intermediate av/ah Ratio
8 Selected Sliding Surfaces -Main Dam
9 Critical Acceleration Plots
10 Permanent Deformation Plots
11 MCE Response/Displacement Plots
12 Megathrust Response/Displacement Plots
13 DBE Response/Displacement Plots
14 Flow Through Dam Without Face
15 Dam Height vs. Acceleration and Displacement
16 Wedge Stability: Sloped Sliding Planes
17 Wedge Stability: Horizontal Sliding Planes
18 La Union Response/Displacement Plots
19 Response Spectrum -La Union E-W Record
20 Response Spectrum -Taft Record
21 Arias Intensity
22 Taft Response/Displacement Plot
0216R-4460R/CG Vll
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 3
DAM AND SPILLWAY STABILITY ANALYSIS
SPILLWAY STABILITY REPORT
Section
1.0 INTRODUCTION
1.1 PURPOSE
1.2 SCOPE
1.3 SPILLWAY SAFETY CRITERIA
2.0 DESCRIPTION OF PROJECT FEATURES
2.1 GENERAL
2.2 OGEE SECTION
2.3 NON-OVERFLOW SECTIONS
2.4 GEOLOGIC CONDITIONS
3.0 DESIGN EARTHQUAKE REGIME
3.1 SEISMOTECTONIC SETTING
3.2 DESIGN RESPONSE SPECTRA
3.3 ACCELEROGRAM DEVELOPMENT
4.0 STABILITY CRITERIA
4.1 GENERAL
4.2 LOADS
4.2.1 Deadweight
4.2.2 Ice
4.2.3 Hydrostatic
4.2.4 Earthquake
4.2.5 Wind
4.2.6 Up1 ift
4.2.7 Temperature
4.3 LOADING CONDITIONS
4.4 ACCEPTANCE CRITERIA
4.4.1 Stability Requirements
4.4.2 Minimum Allowable Stress
4.4.3 Shear-Friction Factor of Safety
5.0 METHODS OF ANALYSIS
5.1 STATIC METHOD
5.2 FINITE ELEMENT METHOD
5.3 SARMA METHOD
6.0 STATIC ANALYSIS
6.1 STABILITY ANALYSIS
6.2 RESULTS
7.0 FINITE ELEMENT ANALYSIS
7.1 STRESS ANALYSIS
7.2 RESULTS
0216R-4460R/CG Vlll
Section
8.0
8.1
8.2
9.0
9.1
9.2
10.0
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
0216R-4460R/CG
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 3
DAM AND SPILLWAY STABILITY ANALYSIS
Section Title
SARMA ANALYSIS
STABILITY ANALYSIS
RESULTS
CONCLUSIONS
CRITICAL CASES
SUMMARY OF STABILITY CONDITIONS
BIBLIOGRAPHY
LIST OF FIGURES
Title
Project Layout Map
General Arrangement -Main Dam Area
General Arrangement -Spillway
Project Response Spectra
Hybrid Accelerogram
Static Spillway Model
Case I -Static Analysis-Base El 1124
Case II -Static Analysis-Base El 1124
Case IV -Static Analysis-Base El 1124
Finite Element Model -Base El 1160
Finite Element Model -Base El 1150
Finite Element Model -Base El 1124
Finite Element Analysis: Case III -Max. Tensile Stresses
-Base El 1160
Finite Element Analysis: Case III -Max. Compressive
Stresses -Base El 1160
Finite Element Analysis: Case V -Max. Tensile Stresses -
Base El 1160
Finite Element Analysis:
Stresses -Base El 1160
Finite Element Analysis:
-Base El 1150
Finite Element Analysis:
Stresses -Base El 1150
Finite Element Analysis:
Base El 1150
Finite Element
Stresses -Base
Finite Element
-Base El 1124
Analysis:
El 1150
Analysis:
lX
Case V Max. Compressive
Case III -Max. Tensile Stresses
Case III -Max. Compressive
Case V -Max. Tensile Stresses -
Case V -Max. Compressive
Case III -Max. Tensile Stresses
Figure
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
0216R-4460R/CG
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 3
DAM AND SPILLWAY STABILITY ANALYSIS
LIST OF FIGURES
Title
Finite Element Analysis: Case III Max. Compressive
Stresses -Base El 1124
Finite Element Analysis: Case V -Max. Tensile Stresses -
Base El 1124
Finite Element Analysis: Case V Max. Compressive
Stresses -Base El 1124
SARMA Analysis Model, Ogee Sections-Sheet 1
SARMA Analysis Model, Ogee Sections -Sheet 2
SARMA Analysis Model, Non-Overflow Sections
SARMA Analysis: Base El 1160 -Ogee
SARMA Analysis: Base El 1150 -Ogee
SARMA Analysis: Base El 1130 -Ogee
SARMA Analysis: Base El 1124 -Ogee
SARMA Analysis: Base El 1160 -Left Abutment
SARMA Analysis: Base El 1124 -Right Abutment
Spillway Stability Analysis Summary -Sheet 1
Spillway Stability Analysis Summary -Sheet 2
Spillway Stability Analysis Summary -Sheet 3
X
HYDRAULIC
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 4
CALCULATIONS
Calculation
Title No.
SPILLWAY CREST SHAPE H-027
FLOOD ROUTING -P.M.F. THROUGH SPILLWAY H-028
FLOOD ROUTING -FLOOD OF RECORD THROUGH H-033
BRADLEY LAKE & DIVERSION TUNNEL
DESIGN THRUSTS -POWER PENSTOCK NEAR H-036
MANIFOLD
SIMPLIFIED DAM BREAK ANALYSES AND WATER H-046
SURFACES PROFILES
WAVE RUNUP AND FORCE ON DAM PARAPET H-048
TAILRACE CHANNEL SLOPE PROTECTION H-050
PROTECTION AGAINST WAVES FOR THE H-066
UPSTREAM COFFERDAM & POWER TUNNEL
INTAKE ROCK PLUG
ICE FORCE ON DAM PARAPET H-068
INVESTIGATION OF NEED FOR AERATION OF
SPILLWAY FLOW H-077
RIPRAP DESIGN H-079
FILLING BRADLEY LAKE RESERVOIR H-081
0216R-4460R/CG Xl
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 5
CALCULATIONS
GEOTECHNICAL
Title
ROCK STRESS IN CIRCULAR TUNNEL LININGS
AND SELECTION OF EXTERNAL WATER
PRESSURE CRITERIA
GROUND WATER SEEPAGE LOADS ON DIVERSION
TUNNEL LINER
VERIFICATION OF INTAKE GEOMETRY FOR
THE POWER AND DIVERSION INTAKES AT THE
BRADLEY LAKE RESERVOIR
EXTERNAL ROCK & GROUND WATER LOADS ON
POWER INTAKE AND GATE SHAFT STRUCTURES
FINAL STABILITY ANALYSIS: BRADLEY LAKE
MAIN DAM
PENSTOCK -MANIFOLD THRUST BLOCK
EMBEDMENT LENGTH AND STABILITY ANALYSIS
ROCK MODULI FOR POWER TUNNEL TRANSIENT
STUDY
DESIGN OF ROCK SUPPORT FOR THE MAIN
POWER INTAKE STRUCTURE
PLINTH AND TOE SLAB GEOMETRY -MAIN DAM
0216R-4460R/CG Xll
Calculation
No.
G(Ak)-04
G(Ak)-08
G(Ak)-10
G(Ak)-22
G(D)-24
G(Ak)-29
G(Ak)-31
G(Ak)-35
G(D)-38
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 6
CALCULATIONS
GEOTECHNICAL
Calculation
Title No.
GROUNDWATER INFLOW & LEAKAGE INTO POWER
TUNNEL G(Ak)-41
EVALUATION OF SHEAR STRENGTH OF ROCK
MASSES AT THE BRADLEY LAKE SITE G(Ak)-47
EVALUATION OF EXTERNAL LOADS ON POWER
TUNNEL LINER G(Ak)-48
VERIFICATION OF CONFINEMENT TO PREVENT
HYDRAULIC JACKING OF THE POWER TUNNEL G(Ak)-49
TAILRACE SLOPE STABILITY & PROTECTION G(A)-50
DESIGN OF ROCK BOLTS FOR DIVERSION G(A)-58
TUNNEL & GATE SHAFTS
DAM TOE PLINTH LOADS G(A)-60
POWER TUNNEL INTAKE EXCAVATION
DESIGN
MANIFOLD & PENSTOCK THRUSTBLOCK
STABILITY CONSIDERING SHEAR ZONE
FEATURE
POWERHOUSE CELLULAR SHEETPILE
COFFERDAM STABILITY ANALYSIS
EVALUATION OF CONCRETE LINER
REQUIREMENTS FOR THE MAIN
POWER TUNNEL
MAIN DAM FACE SLAB DESIGN
SPILLWAY: SARMA DISPLACEMENT
ANALYSIS
SPILLWAY OF THE UPSTREAM COFFERDAM
TOE AND ABUTMENT PLINTH DOWEL EMBED.
LENGTHS AND QUANTITIES
0216R-4460R/CG Xlll
G -70
G -86
G(Ak)-89
G(Ak)-90
G(Ak)-93
G -98
G -104
G -106
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 7
CALCULATIONS
STRUCTURAL
Title
WIND LOADS FOR DESIGN CRITERIA
SNOW & ICE LOADS FOR DESIGN CRITERIA
SEISMIC DESIGN DATA
MAIN DAM DIVERSION TUNNEL LINING AND
GATE CHAMBER ANALYSIS
POWER TUNNEL INTAKE
POWER TUNNEL GATE CHAMBER AND LINING
DESIGN AND ANALYSIS
GATEHOUSE CONCRETE STRUCTURE
0216R-4460R/CG XlV
Calculation
No.
SDC.l
SDC.2
SDC.3
SC-133-3
SC-151-16
SC-152-21
SC-152-32
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 8
CALCULATIONS
STRUCTURAL
Title
DAM PARAPET
MAIN DAM TOE PLINTH DESIGN
SEGMENTS A, B, C, D
ABUTMENT DESIGN
SPILLWAY STABILITY ANALYSIS -
STATIC ANALYSIS
FINITE ELEMENT ANALYSIS OF SPILLWAY
FOR SEISMIC LOAD COMBINED WITH
DEAD WEIGHT, ICE THRUST, AND
WATER LOADS
SPILLWAY TRAINING WALLS
0216R-4460R/CG XV
Calculation
SC-191-26
SC-191-27
SC-191-29
SC-201-8A
SC-201-34
SC-205-23
TABLE OF CONTENTS (Continued)
FINAL SUPPORTING DESIGN REPORT
GENERAL CIVIL CONSTRUCTION CONTRACT
VOLUME 9
CALCULATIONS
STRUCTURAL
PENSTOCK AND MANIFOLD ANCHOR BLOCKS
MAIN DIVERSION & MAIN INTAKE BULKHEADS
MAIN DAM DIVERSION PENSTOCK DESIGN
POWER TUNNEL INTAKE TRASH RACKS
POWER PENSTOCK THRUST RINGS AND
MISC. COMPONENTS
REQUIRED THICKNESS OF STEEL LINER
UNDER INTERNAL AND EXTERNAL PRESSURE
STRESS ANALYSIS OF FLANGE WITH
108" INSIDE DIAMETER
LOCAL STRESSES DUE TO GEOMETRY
DISCONTINUITY AT REDUCERS AND MITERED
ELBOWS
REQUIRED THICKNESS OF ELLIPSOIDAL
HEADS FOR PENSTOCK
STRESS ANALYSIS OF POWER
PENSTOCK WYE BRANCH
PENSTOCK ACCESS FLANGE BOLTS
0216R-4460R/CG xvi
Calculation
No.
SC-261-25
SS-132-2
SS-134-12
SS-153-10
SS-261-16A
SS-261-17A
SS-261-178
SS-261-17C
SS-261-17D
SS-261-17F
SS-261-18
DAM
STABILITY REPORT
DAM STABILITY REPORT
BRADLEY LAKE HYDROELECTRIC PROJECT
Prepared for
ALASKA POWER AUTHORITY
March 1988
STONE & WEBSTER ENGINEERING CORPORATION
DENVER, COLORADO 80111
PROJECT NO. P-8221-000
BRADLEY LAKE HYDROELECTRIC PROJECT
ALASKA POWER AUTHORITY
DAM STABILITY REPORT
TABLE OF CONTENTS
Section Section Title
1.0 INTRODUCTION
1.1 PURPOSE
1.2 SCOPE
1.3 DAM SAFETY CRITERIA
2.0 DESCRIPTION OF PROJECT FEATURES
2.1 GENERAL
2.2 MAIN DAM
2.3 UPSTREAM COFFERDAM
3.0 DESIGN EARTHQUAKE REGIME
3.1 SEISMOTECTONIC SETTING
3.2 DESIGN RESPONSE SPECTRA
3.3 ACCELEROGRAM DEVELOPMENT
4.0 ALTERNATIVE METHODS OF ANALYSIS
4.1 GENERAL STABILITY CRITERIA
4.2 PSEUDOSTATIC METHOD
4.3 SARMA/NEWMARK METHOD
4.4 FINITE ELEMENT METHOD
4.5 SELECTION OF SARMA METHOD
5.0 SARMA ANALYSIS METHODOLOGY
5.1 MATERIALS PROPERTIES AND EARTHQUAKE
SELECTION
5.2 LEASE II ANALYSIS
5.2.1 Static Analysis
5.2.2 Critical Circles and Accelerations
5.3 SARMA ANALYSIS
5.3.1 Data Requirements
5.3.2 Processing
5.3.3 Analytical Output
5.3.4 Significance of Results
4329R/CG
Page
1-1
1-1
1-1
1-2
2-1
2-1
2-1
2-2
3-1
3-1
3-3
3-5
4-1
4-1
4-2
4-3
4-4
4-5
5-1
5-1
S-5
S-5
S-6
5-7
S-7
5-9
5-10
5-11
TABLE OF CONTENTS (Cont'd)
Section Section Title Page
6.0 BRADLEY LAKE EMBANKMENT ANALYSES 6-1
6.1 EARTHQUAKE RECORDS 6-1
6.2 INPUT PARAMETERS 6-1
6.3 DESIGN CASES 6-3
6.4 LEASE II ANALYSES 6-6
6.5 SARMA ANALYSES 6-8
6.6 INTERPRETATION OF RESULTS 6-8
6.7 SPECIAL STUDIES 6-10
6.7.1 Mega thrust (ah = .SSg) 6-10
6.7.2 DBE (ah = .375g) 6-11
6.7.3 Influence of Downstream Berm 6-12
6.7.4 Failed Concrete Face 6-12
6.7.5 Varying Embankment Height 6-13
6.7.6 Planar Slip Surfaces 6-14
6.7.7 La Union Accelerogram 6-16
6.7.8 Parametric Analyses 6-18
6.8 COFFERDAM 6-20
7.0 CONCLUSIONS 7-1
7.1 CRITICAL CASES 7-1
7.2 SUMMARY OF CRITICAL FAILURE SURFACES 7-2
7.3 PREDICTED DISPLACEMENTS 7-3
7.4 RESPONSE TO VARIOUS EVENTS 7-6
8.0 BIBLIOGRAPHY 8-1
4329R/CG
Table
1
2
3
4
5
6
7
8
4329R/CG
LIST OF TABLES
Title
Main Dam Characteristics
Main Dam Cofferdam Characteristics
Historic Seismic Events Summary
Design Factors of Safety
Input Parameters
Main Dam Static Stability Summary
Main Dam SARMA Results
SARMA Results-Old Geometry
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
4329R/CG
LIST OF FIGURES
Title
Project Location Map
Main Dam Area -General Arrangement
Main Dam Sections
(Not Used)
MCE Response Spectra -Mean and Chosen
Rockfill Friction Angles
Intermediate av/ah Ratio
Selected Sliding Surfaces -Main Dam
Critical Acceleration Plots
Permanent Deformation Plots
MCE Response/Displacement Plots
Megathrust Response/Displacement Plots
DBE Response/Displacement Plots
Flow Through Dam Without Face
Dam Height vs. Acceleration and Displacement
Wedge Stability: Sloped Sliding Planes
Wedge Stability: Horizontal Sliding Planes
La Union Response/Displacement Plots
Response Spectrum -La Union E-W Record
Response Spectrum -Taft Record
Arias Intensity
Taft Response/Displacement Plot
1.0 INTRODUCTION
1.1 PURPOSE
The purpose of this stability analysis is to document that the designs of
the Bradley Lake Project main darn and cofferdam meet stability criteria
established for the site. Figure 1 shows the site location and
relationship of the darnsite to other project features.
1.2 SCOPE
This report presents results of stability analyses of the Bradley Lake main
embankment and the associated upstream cofferdam. The darnsite general
arrangement is shown on Figure 2.
The Bradley Lake darn embankment was analyzed to determine its factor of
safety under various static loading conditions, and to predict its
potential deformation under seismic loading conditions. Static cases
analyzed include maximum and minimum normal headwater levels, Probable
Maximum Flood (PMF), and end of construction conditions.
The seismic analyses include estimates of maximum anticipated permanent
deformation expected to result from the Maximum Credible Earthquake (MCE),
the Design Basis Earthquake (DBE), and an intermediate sized mega thrust
event. In addition, parametric studies were performed to determine the
sensitivity of the analysis to the relevant variables.
4329R/CG 1-1
1 • 3 DAM SAFETY CRITERIA
The basic requirement which must be met by the embankment design is that
the reservoir must be retained under all conditions evaluated. Dam safety
criteria were established for this analysis to aid in evaluating
embankment performance.
For static loading conditions, factors of safety were calculated and
compared with design criteria minimum safety factors. These minimum safety
factors were based on current industry standard practice. The calculated
factors of safety were greater than or equal to the recommended minima.
For dynamic loading conditions, the safety criteria were in the form of
deformation limits since a safety factor is often not relevant to fill
structures under dynamic loading. The limit on vertical deformation was
loss of one half of the freeboard for permanent structures. The normal
maximum headwater level on the main dam leaves 10 ft. of freeboard (not
including wave parapet), so the greatest allowable vertical deformation was
5 ft. The limit on horizontal deformation was based on the need to keep
the bedding layer beneath the concrete face sufficiently intact in order to
limit seepage through the embankment after a major earthquake.
4329R/CG 1-2
2.0 DESCRIPTION OF PROJECT FEATURES
2.1 GENERAL
The main dam wi 11 be built at the outlet of Bradley Lake in a rock gorge
and will be constructed on an excavated bedrock foundation. The upstream
cofferdam will be permanent, providing for future low reservoir level
access to the diversion tunnel inlet and providing for potential dam face
dewatering.
2.2 MAIN DAM
The main dam included in the Bradley Lake Project design is a concrete
faced rockfill dam with a design crest length of 602.5 ft. and a maxi-mum
height of approximately 120 ft. (Fig. 3). The slopes of the embankment
will be 1.6H:lV. The embankment crest width is 18ft plus the wave parapet
wall.
The concrete face will rest on a bedding layer at least 12 ft. thick
(horizontally) composed of angular gravel. The concrete face is topped by
a parapet wall four feet high at the upstream edge of the dam crest.
The location chosen for the main dam has a relatively steep (near vertical
in places) right abutment and a generally more gently sloped left
abutment. Table 1 provides a summary of main dam characteristics.
4329R/CG 2-1
2.3 UPSTREAM COFFERDAM
The upstream cofferdam (Fig. 4) will not be incorporated into the main dam,
but will remain in place to facilitate dewatering of the full concrete face
and toe plinth of the main dam. Its slopes will be 2H: lV. It will be
approximately 30 ft. high and 200 ft. long. See Table 2 for a sunnnary of
cofferdam characteristics.
4329R/CG 2-2
3.0 DESIGN EARTHQUAKE REGIME
3.1 SEISMOTECTONIC SETTING
The detailed project seismic design studies and parameters are provided in
two reports by Woodward-Clyde Consultants (Ref. 1 and 2).
Southern Alaska is one of the world's most seismically active regions. The
primary cause of seismic activity in southern Alaska, including the site
area, is the stress imposed on the region by the relative motion of the
Pacific and the North American tectonic plates at their common boundary.
The Pacific plate is moving northward relative to the North American plate
at a rate of about 2-1/2 in/yr., causing the underthrusting of the Pacific
plate. This underthrusting results primarily in compressional deformation,
which causes folds, high-angle reverse faults, and thrust faults to develop
in the overlying crust. A counterclockwise rotational effect also induces
strike-slip faulting parallel to the plate boundary.
The boundary between the plates where the underthrusting occurs is a
northwestward-dipping megathrust fault or subduction· zone. The Aleutian
Trench marks the surface expression of this subduction zone and is located
on the ocean floor approximately 185 miles southeast of Bradley Lake. The
orientation of the subduction zone is inferred along a broad inclined band
of seismicity, referred to as the Benioff Zone, that dips northwest from
4329R/CG 3-1
the Aleutian Trench, and is approximately 30 miles beneath the surface at
the Bradley Lake site. Great earthquakes (Richter magnitude M =8 or s
greater) and large earthquakes (M =7 or greater) s have occurred
historically throughout the region and can be expected to occur in the
future. Historically (1899 to date), eight earthquakes ranging between
M =7.4 and M =8.5 have occurred within 500 miles of the site (Ref. 2). s s
Table 3 provides a representative summary of significant historic seismic
events in the project area.
Bradley Lake is situated on the overriding crustal block above the
subduction zone and between the Castle Mountain fault to the northwest and
the Patton Bay-Hanning faults to the southeast. All of these faults have
documented Holocene or historic surface ruptures. Because of the active
tectonic environment, activity is conceivable on other faults, such as
those found near or on the project site between the known active faults
mentioned above.
Two faults of regional extent exist at or near the site. The Border Ranges
Fault trends southwest beneath Kachemak Bay to the west of the project, and
the Eagle River fault crosses the southeastern end of Bradley Lake at about
the same trend. While no evidence of recent activity along these faults
has been found in the site area, recently defined data indicates recent
activity on the Eagle River Fault near Eklutna (125 mi NE of the site).
Given the tectonic setting, it is reasonable to consider these faults as
potentially active.
4329R/CG 3-2
In addition to the nearby regional faults, the site is crossed by two large
local faults, called the Bradley River Fault and the Bull Moose Fault, and
a number of probable smaller faults. The dominant trend is northeasterly,
paralleling the regional trend. The larger local faults, particularly the
Bradley River, are considered as potentially capable of independent
earthquake generation, while any of the local faults could possibly move in
sympathetic response to earthquakes occurring on the regional faults.
It is therefore concluded that the site will probably experience at least
one moderate to large earthquake during the life of the proposed project.
The possibility of significant ground rupture exists but is much less
subject to prediction and is considered to have a much lower probability.
3.2 DESIGN RESPONSE SPECTRA
The response spectra considered for this analysis were taken from a report
prepared by Woodward-Clyde Consultants (WCC) for the Army Corps of
Engineers (Ref. 1). The report documents the work performed by WCC to
develop parameters for what the Corps terms the "design maximum earthquake"
and the "ope.rational base earthquake", henceforth called Maximum Credible
Earthquake and Design Basis Earthquake, respectively. The Maximum Credible
Earthquake (MCE) is defined as the most severe earthquake believed to be
probable that could affect the site. The Design Basis Earthquake (DBE) is
less severe, and it is defined as the seismic level that is considered
likely to occur during the life of the project. Maximum Credible
Earthquakes are normally used as a basis for determining whether or not
4329R/CG 3-3
certain structures can withstand extreme events having remote probabilities
of occurring, regardless of damage level. Design Basis Earthquakes are
used as a basis for estimating the maintenance and other costs resulting
from events expected to occur, and for design of non-critical structures
where severe damage and loss of function in a seismic event is considered
an acceptable risk. The response spectra for both the DBE AND MeE will be
used in the seismic stability analysis to estimate vertical and horizontal
displacements of the dam.
Based on their work on the seismicity of the site, wee proposed two
possible response spectra for the "design maximum earthquake", the
equivalent of the MeE. The one that was expected to control was based on
rupture of one of the faults nearest the site. The resulting earthquake
would have a magnitude of M =7.5, peak ground acceleration of 0.75g at s
the site, and a significant ground motion duration of 25 seconds. This
event would have a response spectrum roughly corresponding to the upper
smooth curve on Figure 5.
The other possible MeE was an event tied to the Benioff Zone roughly 30
miles beneath the site. This event would have a magnitude of M =8. 5, s
peak ground acceleration of 0.55g at the site, and a significant ground
motion duration of 45 seconds. It was not expected to be the controlling
event unless the faults in the immediate vicinity of the site could be
shown to be inactive.
4329R/eG 3-4
A third response spectrum proposed by WCC was an event with a peak ground
acceleration approximately one half that of the MCE. This was used for the
DBE with a peak ground acceleration of 0.35g (Fig. 5).
3.3 ACCELEROGRAM DEVELOPMENT
Because the near field crustal M =7.5 s event is more severe than the
megathrust M =8.5 event, in terms of both peak parameters and spectral s
accelerations throughout the frequency range of interest, an accelerogram
for the crustal event is of primary interest. The megathrust event is
considered in detail in the WCC reports, but was utilized in design only
for parametric comparative purposes.
Since all critical structures of the Bradley Lake Project are founded on
bedrock, it is desirable to perform seismic analyses based on actual
accelerograms recorded on rock from large magnitude earthquakes having
similar parameters to those for the crustal event. More importantly, the
response spectra of any accelerograms used should match, in an average
sense, the curve shown in Figure 5. At the time the analysis was being
performed, no accelerograms recorded on rock in the near field of large
magnitude earthquakes (M 7.5+) were available anywhere in the United s
States, including Alaska. Furthermore, no such comparable record was known
to be available from elsewhere in the world. Consequently, available
accelerograms from historical earthquakes having appropriate peak and
spectral characteristics over a broad period range, even when scaled, were
not available for use.
4329R/CG 3-5
To provide the appropriate accelerograms to the COE, WCC developed
synthetic accelerograms (Ref. 1). The synthetic accelerograms were
developed by taking an existing "real event" accelerogram and modifying its
spectral ordinates using a trial-and-error frequency-domain technique. This
approach and the resulting accelerograms are valid when the response of the
structure analyzed is essentially linear and elastic; as is usually the
case for concrete dams, spillways, intake towers, power plant structures,
etc. However, for the rockfill dam designed for Bradley Lake by Stone &
Webster, WCC' s accelerograms may not be appropriate. The reason is that
synthetic accelerograms developed from interactive frequency domain
techniques are usually "frequency-rich". That is, this type of synthetic
accelerogram is normally characterized by well-ordered frequency components
which exist uniformly throughout the record. This is never the case with
accelerograms obtained from actual earthquakes, and it may lead to an
over-estimation of nonlinear phenomena, such as accumulated displacements
in a rockfill dam.
To avoid this over-estimation, a composite hybrid accelerogram consisting
of historical accelerograms from two earthquakes, having appropriate
characteristics, was developed and used for the Bradley Lake Main Dam
analyses in lieu of wee's synthetic accelerogram. This approach has been
previously used for other studies including those performed by the
California Department of Water Resources for Oroville Dam and is considered
an appropriate state-of-the-art method for simulation of strong motion
events.
4329R/CG 3-6
After examining the response spectra for recorded accelerograms from a
number of earthquakes in the United States and abroad, it was cone! uded
that a suitable accelerogram for the M =7.5 crustal event could be s
obtained by combining the S69°E component of the Taft record from the
1952 Kern County, California earthquake and the East-West component of the
San Rocco record from the September 15, 1976 Friuli, Italy earthquake. The
Taft record was scaled by a factor of 3. 5 and was used to represent the
hybrid earthquake from time 0. 00 to 2. 32 seconds and from 4. 32 seconds to
the end. The portion of the Friuli record from time 2.14 to 4.10 seconds
was scaled by a factor of 3.2 and inserted into the scaled Taft record,
replacing the portion of the Taft record from time 2.34 through 4.30
seconds in the hybrid record. In effect, the portion of the Friuli record
with the highest accelerations was spliced into the high-acceleration
portion of the Taft record, resulting in a record with greater duration and
a greater proportion of relatively high acceleration peaks. The resulting
accelerogram, called the Hybrid record, is shown on the displacement plots
starting with Figure 11, and its response spectrum is compared to the
spectrum recommended by wee in Figure 5. The significant duration of the
Hybrid record, defined as the time to reach 95% of the Arias Intensity
(Ref. 16), is 28.8 seconds. This is slightly longer than the 25 second MCE
proposed by wee. This longer event duration, when combined with the
greater density of high acceleration peaks from the combined records,
results in a design record which is felt to be conservatively intense and
definitely on the "safe" side when used to simulate the project MeE. As
4329R/CG 3-7
will be explained later in the discussion of results, it was not necessary
to develop a separate record for the mega thrust event since the magnitude
7.5 crustal earthquake has been demonstrated to be more severe to project
structures.
4329R/CG 3-8
4.0 ALTERNATIVE METHODS OF ANALYSIS
4.1 GENERAL STABILITY CRITERIA
Stability analysis of an embankment dam is intended to allow prediction of
how the embankment will behave under anticipated loading conditions.
Generally, the behavior of interest is any movement of the embankment
material that might lead to failure of the dam, or secondarily, any change
that might cause excessive leakage through the dam. An analysis commonly
results either in factors of safety or in estimates of permanent
deformation.
For static and pseudostatic analyses, stability criteria are most
appropriately given in terms of minimum factors of safety acceptable under
various loading conditions. The criteria for minimum factors of safety
used in the static portion of this analysis are found in Table 4, as
modified for the project from Wilson and Marsal, 1979 (Ref. 3).
For dynamic analyses, the stability criteria are most appropriately stated
as maximum allowable deformations. In the operating case, vertical
deformations were limited to one half the normal freeboard, which would be
5 ft. on the main dam. Horizontal deformations that would produce movement
in the bedding layer were limited to roughly one half the width of the
layer.
4329R/CG 4-1
4.2 PSEUDOSTATIC METHOD
For purposes of simplicity, dynamic stability analyses are sometimes done
by the pseudostatic method. Dynamic loads are modeled as static forces and
a normal static analysis is performed. The dynamic load due to an
earthquake. is included as the static force which would result from a steady
horizontal acceleration equal to the peak ground acceleration expected from
the earthquake. This method is useful in situations where only relatively
small peak ground accelerations are expected. In other cases it has some
limitations as described below.
Perhaps the most obvious limitation of 'the pseudostatic method is that it
gives results 1n terms of factors of safety. Thus, its results are readily
useable only if no permanent deformation is expected; that is, when the
factor of safety is one or greater. Factors of safety less than one are
meaningless in this method, and give an indication of total failure. This
interpretation is overly conservative for massive structures which can be
designed to accommodate movement.
The other limitation of the method is that it does not take into account
the frequency characteristics of the earthquake or the natural frequency
and response of the darn. This is particularly important for earth
embankments since their natural frequencies are generally significantly
lower than the frequencies associated with the highest accelerations in
earthquake accelerograrns. This causes the pseudostatic method to be overly
4329R/CG 4-2
conservative for sites where large peak accelerations are expected. For
this reason, the pseudostatic method is not recommended for use with peak
ground accelerations greater than 0.2g (Ref. 4).
4.3 SARMA/NEWMARK METHOD
This is one of the methods often used to model the response of darns to
earthquakes when the pseudostatic method is inappropriate. It is commonly
used by the Army Corps of Engineers in modeling for darn design or analysis
of existing darns.
The Sarma method starts with the calculation of resonant frequencies and
response shapes of the embankment for each frequency. The next step is
calculation of participation factors for a given potential failure wedge or
block. These factors describe how much effect each of the modes of
oscillation will have on the potential failure wedge. Once this is
accomplished, the accelerations of the wedge in each mode in response to
the earthquake accelerograrn are calculated, and the modes are combined.
The result is a time-history of the accelerations the wedge would
experience as a result of the chosen earthquake.
Once the time-history of acceleration of the individual wedge is known, the
cumulative displacement is calculated by Newmark's sliding block procedure
(Ref. 5). In this procedure, the wedge is assumed to remain fully attached
to the rest of the dam as long as the average acceleration of the wedge is
less than a specified critical (or break-free) acceleration. When the
4329R/CG 4-3
acceleration exceeds the critical acceleration, the wedge slides relative
to the darn until it comes to rest during a subsequent reversal of the
acceleration. The total movement of the wedge is the sum of all the
increments of movement that occur during a particular earthquake record.
Wedges most appropriate for analysis and their respective critical
accelerations are calculated using static and pseudostatic analyses. Aside
from basic simplifying assumptions such as homogeneity and symmetry of the
darn and elastic behavior of the material, a significant limitation of this
method is that permanent deformations are assumed to occur along a
well-defined failure surface rather than along multiple surfaces. Thus,
some care must be used in interpreting the resulting deformations. For
example, while movement along a single surface might result in predict ion
of a discrete offset of several feet in the concrete face, the same amount
of deformation distributed over a significant area of the face along
innumerable sliding surfaces would result in a curvature of the face and
limited cracking, without large discrete offsets. The latter type of
movement is more typical of granular fill structures in earthquakes.
Therefore, it is advantageous to analyze a series of different wedges to
see how the tendency to deform varies with location in the dam.
4.4 FINITE ELEMENT METHOD
The Finite Element Method (FEM) is a widely known computer modeling method
that is sometimes applied to embankment dams. It allows somewhat more
detailed modeling of the embankment configuration and variations in material
4329R/CG 4-4
properties than the Sarma method. The FEM also generally models stress
distributions and transmission of vibrations well, but it does not readily
model permanent deformations.
As with other methods, the results obtained are dependent on the values
chosen for different material properties. With the FEM, the results are
dependent to a significant degree on how the structure is modeled. For
example, whether the mass is modeled at nodes or distributed through the
elements has a significant impact on how the model responds to vibration.
For these reasons it is desirable to make numerous runs varying material
properties and some other properties of the model to improve the
reliability of, or confidence in, the resulting stresses and strains.
Since the FEM does not model permanent deformations, another step must be
taken in the analysis to compute them. This can take the form of the
Newmark sliding block analysis (Ref. 5) or one of a few other methods
(Ref. 6).
4.5 SELECTION OF SARMA METHOD
The three methods of seismic embankment analysis mentioned above were
considered for use in this analysis. The pseudostatic method was quickly
eliminated because, with a peak ground acceleration of 0.75g, it would
result in a excessively conservative design. After evaluation of the
remaining two methods, the Sarma method was chosen based on a number of
factors.
4329R/CG 4-5
While the FEM could potentially provide a better approximation of the final
shape of the embankment, the magnitude of the deformation is more important
than the approximate final shape of the embankment slopes. Permanent
deformation magnitudes calculated by the Sarma method are thought to be as
realistic as those produced by the FEM, though the Sarma method is thought
to be the more conservative. Also the ability of the FEM to model zones
within an embankment in some detail is not needed for the largely
homogeneous embankment used at Bradley Lake. Thus the FEM, which is
markedly more expensive, has little advantage over the Sarma method.
Permanent deformation analysis by the FEM requires many of the same
simplifying assumptions used by the Sarma method. Both the methods are
dependent on a number of variables, some of which are difficult to pick
with precision. For instance, the Sarma analysis requires the assumption
that the shear wave velocity is constant throughout the embankment. The
FEM is dependent to a significant degree on modeling aspects, such as how
the mass is distributed in the model, that are easy to overlook or
oversimplify.
All of the above factors make it desirable to make multiple calculations
with whatever method is used, varying the inputs through a reasonable range
and trying various potential failure surfaces or modes. In this area the
Sarma method has a clear advantage because it is significantly easier and
cheaper to use.
4329R/CG 4-6
The Sarma method is widely used and accepted, e.g., by the Army Corps of
Engineers, and is considered to be relatively conservative in the way it
models and amplifies earthquake accelerograms. Also, deformations
calculated using the Sarma method are considered to be as realistic as, and
perhaps more conservative than, those predicted by the FEM. Finally, the
Sarma method lends itself to parametric studies which would probably not be
feasible with the FEM due to constraints of time and expense. For these
reasons it was concluded that the Sarma method was the more appropriate
method for use in the Bradley Lake main dam analysis.
4320R/CG 4-7
5.0 SARMA ANALYSIS METHODOLOGY
5.1 MATERIAL PROPERTIES AND EARTHQUAKE SELECTION
Aside from the geometry of the embankment, several parameters must be
chosen to describe the mechanical properties of the fill materials. For
static and pseudostatic analyses using the Slope Stability Analysis LEASE
II program (Ref. 7), the internal friction angle (0) and the unit weight
must be picked for each material. For seismic analyses using the SARMA
program (Ref. 8), an average damping ratio (D.R.) and shear wave velocity
(V ) must be picked for the embankment. s
Values for the four parameters mentioned above can be obtained from
laboratory tests or from a review of published values. For rockfill, as
opposed to soil, tests to measure 0 are very difficult and expensive and
are rarely representative of coarse materials. However, a review of
available test results has been published (Ref. 9) which allows a
reasonable estimate of 0 to be made. The friction angle varies with
several factors. Higher values are associated with more angular, hard,
strong particles in a well graded rockfill. Higher values of 0 are also
associated with lower confining pressures. These factors and Figure 6 can
be used to estimate an appropriate friction angle.
The rockfill unit weight can be estimated directly from published values,
or based on the unit weight of the source rock and reasonable values of void
4329R/CG 5-l
ratio. Using both these approaches increases confidence in the resulting
estimate.
Shear wave velocity of rockfills can be most readily estimated based on
published values and empirical equations (Ref. 10). V is dependent on s
confining stress, degree of compaction and the level of strain associated
with the wave. Rockfill is significantly stiffer at lower strain levels
and experiences strain softening as strain levels increase and as the
shaking continues (i.e., the number of cycles of displacement increases).
The level of strain must first be estimated and used to choose V . This s
V is then used in the calculation to find the level of strain. s Thus,
the process is iterative.
The damping ratio, similar to V , is most readily obtained from s
literature. It is also strain dependent and so must be chosen by an
iterative process.
The remaining input to the analysis, 1n addition to the darn geometry, is
the earthquake record to be used. For the pseudostatic part of the
analysis performed by the LEASE II program, the required input is the ratio
between vertical and horizontal accelerations. Neglecting vertical
acceleration is considered unconservative for sites with fairly large
earthquakes, though the Army Corps of Engineers does generally omit the
vertical component in their analyses. On the other end of the spectrum,
using an av/~ of 2/3 (which is cited as a common ratio of peak
a/~) is overly conservative since it assumes a relatively large
vertical component occurring simultaneously with the peak horizontal
4329R/CG S-2
component of the earthquake. A compromise between these two extremes,
illustrated by Figure 7, was suggested by Dr. A. J. Hendron. The intent of
this approach is to find the av/~ ratio which yields the smallest
possible resultant acceleration that will be critical for a given wedge.
For the purposes of the SARMA program, an accelerogram with appropriate
characteristics must be used. Characteristics of the earthquake are site
specific and depend on such factors as whether the dam is founded on
bedrock, and proximity, length, and geologic type of capable faults. The
characteristics of the record which should be examined include significant
duration, earthquake magnitude, peak ground acceleration, response
spectrum, and source. Significant duration in the current case was
quantified as the time between reaching 5% and 95% of the Arias Intensity
(Ref. 11). Response spectrum and source are related, since a record of
bedrock acceleration generally will have a different frequency content than
one recorded on soil or on a foundation not in intimate contact with
bedrock.
The importance of the above factors is more obvious in some cases than in
others. Significant duration is loosely related to the amount of damage an
earthquake might be expected to cause. Magnitude affects the amount of
energy that an earthquake might be expected to put into a structure and
also the distribution of that energy across the range of frequencies. Peak
ground acceleration is a parameter widely used to scale the magnitude of
records to appropriate levels. The response spectrum of a record is of
4329R/CG 5-3
great significance to the current analysis because the typical earthquake
transmits most of its energy at frequencies higher than the natural
frequency of an earth or rockfill dam. The dam cannot respond to this
higher frequency portion of the record and therefore is not significantly
affected by it.
Ideally, for a dynamic stability analysis, one would be able to select one
or more recorded accelerograms that fit the above parameters. Failing that,
a record with an appropriately shaped response spectrum and duration could
be scaled up or down to achieve the right peak ground acceleration. If no
record is available with the proper duration, existing records can be
shortened by removing parts of the accelerogram, or lengthened by repeating
some parts or inserting parts of similar accelerograms. In the current
analysis, this method of combining accelerograms was used to manufacture
the Hybrid accelerogram. The resulting accelerogram fits the mean response
spectrum developed by wee significantly better than any of the available
natural records.
If no earthquake records are available that can be adjusted to be suitable,
artificial records (pseudoseismograms) can be generated by filtering white
noise to fit a specified duration and response spectrum. These may result
in an overly conservative design when inelastic response is expected, as in
the case of earth or rockfill dams.
4329R/eG S-4
To illustrate, a response spectrum from an actual earthquake record may
show a peak acceleration of 0.4g at a period of 1 second. The actual
earthquake may reach that acceleration level at that period only once while
the rest of the record does not come close to it. By contrast, a synthetic
record created to fit that response spectrum by filtering white noise would
contain a continuous component with a 1 second period and 0.4g
acceleration. This would make little difference while all motion is
elastic, but when permanent deformation is calculated, the synthetic record
would most likely result in significantly more pulses of inelastic
movement. Thus, this approach to earthquake record generation is overly
conservative for the type of analysis under consideration.
5.2 LEASE II ANALYSIS
5.2.1 Static Analysis
Once material properties (0, unit weight) have been selected, static
stability analysis can be performed using the LEASE II program. This
program calculates stability of masses defined by circular arcs using the
Simplified Bishop's method or, for non-circular failure surfaces, the
Morgenstern-Price method. Rotation centers can be tried individually or
can be selected automatically following a grid entered with the data. For
each center of rotation the program calculates factors of safety for
circles of various size, starting with the largest radius selected and
ending with the smallest radius selected.
4329R/CG 5-5
The program allows consideration to be limited to circles of significance.
For example, circles through bedrock can be excluded and circles of less
than a certain depth, such as 5 or 10 feet, can also be excluded.
Embankment geometry, water levels and phreatic surfaces can be entered and
handled by the program in as much detail as is necessary to provide the
desired results. For each static loading condition of interest, the
geometry, water levels and any other loads are entered and a grid of center
points is picked and tried. Ideally, the grid is expanded as necessary so
that the center point with the smallest minimum safety factor is
bracketed. If further refinement is needed, a finer grid can be used
around this most critical center point. At the end of this process, the
most critical circle for the given condition has been defined and its
factor of safety calculated. This procedure is repeated for each of the
relevant static loading conditions.
5.2.2 Critical Circles and Accelerations
The first step in the seismic analysis is examination of the results of the
static analysis. Seismic analyses generally assume normal operating
conditions exist, so the relevant part of the static analysis consist of
the results from the normal operating case. One or more circles from each
face of the embankment are picked for use in the SARMA program. These
selected circles have the lowest factors of safety of those analyzed, have
overall geometry which is critical to the overall stability of the dam and
4329R/CG 5-6
represent a variety of potential failure modes or configurations. As an
example of the last requirement, one circle may be picked that is entirely
within the upper half of the dam while another may reach nearly to the toe
of the embankment. Since it is not known beforehand what configuration
will be most critical in terms of permanent deformation, it is best to try
a variety of conditions.
The LEASE II program is used to determine the critical acceleration for
each of the circles chosen. This is done by inputting the same geometry
and loads used in the static analysis together with the chosen center
point, radius and horizontal and vertical accelerations based on the
a/~ ratio picked earlier. A range of seismic accelerations are used
and the resulting factors of safety are plotted versus horizontal
acceleration. The critical acceleration (~c) is the acceleration
corresponding to a factor of safety of 1.0. For each circle, the critical
acceleration must be recalculated for each new value of 0, unit weight,
or av/~.
5.3 SARMA ANALYSIS
5.3.1 Data Requirements
The SARMA program calculates the seismic response of a predefined wedge of
the dam by the Sarma method and integrates to obtain a predicted permanent
displacement of the wedge by the Newmark method, described in Section 4.3.
4329R/CG 5-7
The data required to do this include embankment geometry, wedge geometry,
fill density, a , V , damping ratio and the earthquake accelerogram. nc s
The embankment is treated as a symmetrical wedge with a pointed top and
must be composed of a single material. The data are entered as an
embankment height and a ratio between height and width. This requires some
simplification of the input, but is a very close approximation of the
Bradley Lake main dam, thereby making the use of the program viable. It
also requires that average values of the material properties be used. The
circle radius and center point are entered with the ~c from the LEASE II
analysis. The ~c includes the effects of 0, density, av/~, and
the more detailed geometry. The unit weight, in the form of density, and
the average values of V s and damping
calculation of the embankment response.
ratio are entered to allow
If a foundation layer is appropriate, its thickness (assumed uniform),
density and V are entered. The damping ratio is assumed to be the same s
as that of the embankment. In this instance, since the dam is founded on
bedrock, the intermediate foundation layer was not used.
The earthquake accelerogram is read as a series of ground acceleration
values at constant short time increments. The peak ground acceleration can
be specified, and the earthquake record scaled up or down to this point to
match the design peak acceleration.
4329R/CG 5-8
5.3.2 Processing
The SARMA program uses the embankment shape, density, and V to calculate s
the natural or resonant frequency of the embankment in various modes of
vi brat ion. It also calculates what the shapes of the various modes would
be. Since the potential failure wedge or arc is well defined, it is
possible to calculate how much the potential failure mass would be
influenced by each mode of vi brat ion. This is done by calculating an
average participation factor for the sliding mass as a whole for each mode
of vibration. Using these participation factors and the damping ratio it
is possible to determine how the wedge will respond to oscillations of a
given input frequency at the base of the dam.
Next, the ground accelerations based on the accelerogram are applied to the
base of the dam (or foundation layer if there is one). The resulting
accelerations experienced by the wedge are then calculated. This is called
the time-history of wedge acceleration. The process is repeated with the
time record reversed to obtain a second time history.
The time-history of acceleration and the critical acceleration of the wedge
are used together to calculate cumulative displacement along the sliding
surface. The wedge is assumed to move with the rest of the dam as long as
the average acceleration of the wedge is less than ~c· When the average
acceleration exceeds ~c, the wedge slides relative to the remainder of
the dam until coming to rest during a subsequent reversal of the
acceleration. The resulting increments of movement are summed to yield the
4329R/CG S-9
total anticipated movement. As mentioned above, this is performed with the
earthquake record applied in each direction consecutively. This yields a
net greatest displacement which accommodates any asymmetry of the record.
5.3.3 Analytical Output
The information available from the SARMA program includes the fundamental
resonant frequency and frequencies of other vibration modes for up to 20
modes, modal shapes, participation factors for each mode, the earthquake
accelerograrn, the wedge's time-history of acceleration, peak ground
acceleration, peak wedge acceleration, number of pulses of movement, and
total movement. The program produces a plot (such as Fig. 11) which
includes a graph of the accelerograrn and wedge acceleration response and a
parallel graph of cumulative displacement of the wedge. Much of the output
is optional and can be suppressed if not needed.
The fundamental resonant frequency can be used with the earthquake response
spectrum to see whether the embankment is responding to an outstanding peak
or valley in the spectrum. Participation factors indicate how many modes
have a significant effect on the wedge so the cost of runs can be cut by
cutting out unnecessary modes. Peak wedge acceleration can be compared to
peak ground acceleration to give an indication of how the earthquake is
amplified in the dam. The output number of greatest interest is the
predicted permanent displacement of the wedge.
4329R/CG 5-10
5.3.4 Significance of Results
As mentioned previously, bulging and settling of the embankment is a much
more likely response to an earthquake than movement along a single surface
as modeled by the SARMA program. For this reason, some thought must be
devoted to the meaning of the displacements predicted by the program.
Since the Sarma method is considered to be relatively conservative with
respect to calculated displacements, it seems reasonable that vertical
movement should not exceed movement predicted by this method, even if it
actually occurs by general settlement rather than by movement on a single
surface. Thus, the vertical component of the movement calculated by the
SARMA program is treated as the actual maximum predicted vertical
settlement .of the embankment crest.
On the other hand, if the actual mode of deformation is general settling
and bulging, the horizontal component of the predicted movement should not
be viewed as a single offset of the concrete face. Bulging would produce
curves and cracks in the face, but presumably not offsets of several feet.
Viewed as a potential discrete offset in the concrete face, the horizontal
component of the SARMA result is extremely conservative.
To make the results as representative as possible of the embankment as a
whole, it is helpful to analyze circles or other shapes of various sizes in
various parts of the dam. This provides information on where greatest
deformation might take place and provides greater assurance that the worst
case has been found.
4329R/CG 5-11
A final step that has been taken to increase confidence in the results of
the analysis is to use a range of values for the relevant variables rather
than a single number. The importance of doing this depends on the level of
precision with which the variables can be specified. If one or more
parameters cannot be specified with much precision it is helpful to vary
those parameters over a reasonable range to determine the possible range of
resulting movements.
Performing the Sarma analysis with a range of values for a parameter is
probably most significant in the case of V . This is because the shear s
wave velocity is difficult to define precisely in the first place and
because it changes as a result of "softening" of the embankment during an
earthquake. The amount and time-history of the change is also hard to
define. Another reason for varying V is to allow the unique individual s
characteristics of an accelerogram to be taken into account. If only one
V is used, the SARMA program results can be skewed by a large spike or s
trough in the response spectrum. The effect can be evaluated by varying
v . s
4329R/CG 5-12
6.0 BRADLEY LAKE EMBANKMENT ANALYSES
6.1 EARTHQUAKE RECORDS
During the conceptual design phase, an accelerogram was developed to meet
the criteria proposed by Woodward-Clyde (Ref. 1). The accelerogram used
was the Hybrid record explained in Section 3.3. This record is still
considered the most appropriate for the current analysis. Two additional
records were also tried. One was the Taft Lincoln School Tunnel S69E
component and the other was the La Union E-W record of the Michoacan,
Mexico earthquake of September 19, 1985.
6. 2 INPUT PARAMETERS
The input parameters which must be chosen for the analysis are the material
properties of the embankment and foundation. These include the friction
angle (0) and unit weight for the rockfill and semi-pervious bedding
material, as well as the shear wave velocity (V ) and the damping ratio s
for the embankment as a whole. The rockfill and the bedding material are
assumed to be cohesionless as is anticipated to generally be the case for
hard blasted rock. The rockfill material that will form the major portion
of the embankment is anticipated to be derived from blasted rock with a
moderately high compressive strength of at least 10 to 15 KSI. It will be
specified to be made up of angular particles of predominantly medium to
coarse gravel size and larger, with a specified maximum percentage passing
1 inch and #200 sieves. The particles are expected to be relatively dense
4329R/CG 6-1
and the fill should be well compacted. The semi-pervious bedding material,
used beneath the concrete face, is expected to be like the rockfill except
in its grain size distribution. Its gradation will be that of a well
graded sandy gravel, of minus 3 inch size. The embankment is founded on
bedrock, so use of a foundation layer in the analysis was not necessary.
The choice of the unit weight of the rockfill was based on laboratory test
values of a specific gravity of 2.7 for the particles. By assuming
different porosities for the rock a reasonable range of moist unit weight
of 125 to 150 pcf was determined. A moist unit weight of 135 pcf was
chosen for the analysis. The unit weight of the bedding layer might be
slightly different due to differences in gradation or compaction. The unit
weights are assumed to be the same for purposes of this analysis. This
assumption is necessary for the Sarma analysis and is considered
reasonable. The above values of the unit weight are within the range of
applicable published data.
The choice of the friction angle for the rockfill was accomplished using an
article containing collected triaxial data (Ref. 9). A rockfill composed
of material as described could have a friction angle of 48 to 53 degrees at
depths in the embankment of about 10 to SO feet (Fig. 6). Since most
failure surfaces of interest are relatively shallow, a friction angle of 48
degrees with an analytical range from 45 to SO degrees was chosen as
suitably conservative. In the case of the semi-pervious bedding material,
the same friction angles could be justified, but because of less certainty
as to the exact characteristics of the material, slightly lower values will
be used.
4329R/CG 6-2
A friction angle of 46 degrees was chosen with an evaluated range of 44 to
48.
The shear wave velocity, V , was calculated using the relationship s
between V and the shear modulus, G, and the density: s
V = (G/density)112
s
The shear modulus was calculated as shown in Seed, et. al, (Ref. 10), and
Seed & Idriss, (Ref. 12), and the density was computed for the range of
unit weights previously described. It was determined that a range of shear
wave velocities of 700 to 1000 fps was appropriate, with a most likely
value of 800 fps.
The damping ratio was determined using relationships developed for sands
(Ref. 12). Assuming a strain level of 0.1% results in a damping ratio of
15%. A range of 12 to 20% was chosen to bracket the damping ratio. The
summary of input parameters is provided in Table 5.
6.3 DESIGN CASES
Several typical loading conditions were examined to document the static
stability of the dam and to define the cases to be used in seismic
analyses. The cases of the highest and the lowest headwater elevations
(1180 feet and 1090 feet, respectively) during normal operating conditions
were designated the maximum and minimum normal operating conditions. The
minimum normal headwater elevation at the dam is assumed to be 1090 feet,
4329R/CG 6-3
rather than the usual 1080 feet, because of the effect of the upstream
cofferdam. The condition after completion of construction and prior to
filling of the reservoir was designated the end of construction case. The
probable maximum flood results in the condition of the reservoir reaching
its peak elevation, 1190.6 feet. The condition of rapidly dropping the
headwater elevation is the rapid drawdown case. Steady state seepage flow
will not be significant within the embankment due to the inherent nature of
a rockfill dam. Therefore, in all loading cases the analyses are performed
with an internally dry embankment above tailwater level. Table 6 presents
a summary of static cases and minimum calculated safety factors.
In the maximum normal operating condition (headwater elevation 1180 feet
and tail water elevation 1065 feet) the upstream slope is supported by the
water of the reservoir and the downstream slope becomes the more critical
part of the embankment. A factor of safety of 1. 78 for the downstream
slope was determined using an "infinite slope" analysis. This analysis
results in the minimum static factor of safety for the downstream slope.
In the minimum normal operating conditions (headwater elevation 1090 feet
and tail water elevation 1065 feet) the stability of the downstream slope
will not be changed and the factor of safety will be 1.78 because the toe
is essentially dry. The upstream slope without the concrete face or
support from the water (i.e., dewatered case) would have a minimum factor
of safety of 1.66 based on an "infinite slope" analysis.
4329R/CG 6-4
The end of construction case is not relevant for a concrete faced rockfill
dam because there is no opportunity for consolidation pore pressures to
build up. This case then reduces to the minimum normal operating level
case above.
The probable maximum flood, headwater elevation 1190.6 feet, makes the
upstream slope more stable by adding more water to support it. This case
does not affect the downstream slope since seepage does not develop through
the embankment of a rockfill dam to a significant degree, so the minimum
factor of safety remains 1.78.
In the rapid drawdown case the stability of the rockfill will not be
affected because it is free draining and there is no build up of excess
pore water pressure. Because of the semi-pervious nature of the bedding
layer and the fact that it is well drained on its lower side, it is
reasonable to assume that pore pressures in this layer will also dissipate
as rapidly as the reservoir can be drawn down. This will leave the bedding
layer temporarily at a saturated unit weight. As the LEASE analysis shows,
the unit weight of the material makes very little difference in the factor
of safety. This case would also be very similar to the minimum normal
operating case.
Based on the static analysis, the embankment is stable in all design
cases. For the seismic analysis, the normal cases will be used since it is
highly unlikely for an abnormal high water condition and an earthquake to
happen simultaneously. Three cases will be considered for the seismic
analysis.
4329R/CG 6-5
The maximum credible earthquake (MCE) is the condition caused by nearby
crustal faulting having a peak ground acceleration o-f 0. 75g. The design
basis earthquake (DBE) uses a peak ground acceleration of 0.375g. This is
slightly higher than the final selected project criterion of 0.35g, but was
retained throughout analysis in order to avoid two variants in the same
analysis. The megathrust condition, which simulates a large earthquake at
a great distance, uses a peak ground acceleration of O.SSg.
6.4 LEASE II ANALYSES
A static stability analysis was conducted using the LEASE II program to
determine the critical potential slip surfaces and their critical
accelerations. The analysis was conducted for both the upstream and
downstream slopes for the two normal cases; normal maximum and normal
minimum operating levels. Two embankment geometries were used 1n the
initial LEASE II analyses. The old geometry which was based on the
conceptual cross-section did not include a downstream berm. The final
embankment geometry which was selected during the analyses, included the
downstream berm. Large, coarse grids of circle center points were used to
locate the areas of low factors o£ sa-fety on the original embankment.
Smaller finer grids were then used on the final geometry to determine the
slip circles with the lowest factors of safety. In this manner, six slip
circles of various types in various parts of the dam were selected for
further analysis. The circles on the downstream slope are designated A, B,
and C and the circles on the upstream slope are designated D, E, and F.
These slip circles are shown on Figure 8. The circles were chosen because
of their low factors of safety and because they are distributed over the
2-1223-JJ 6-6
height of the dam. The shallow "infinite-slope" cases were not included
because they are not critical to water retention or structural stability,
because they would result only in loss of dam crest width. If multiple
regressive surface slumps occurred, they would eventually regress to one of
the sloped sliding planes, illustrated on Figure 16.
The critical accelerations of the six slip circles were next determined for
use in the SARMA program. The LEASE II program was used with a 0 value
of 48° and a unit weight of 135 pcf. The program also requires input of
a vertical and horizontal seismic acceleration, so the ratio of vertical to
horizontal acceleration, a/~· was determined. Three values of
a/~ were used for each circle. The first is given by the equation
a/~ = tan (0-8) where 0 is the friction angle and e is the
slope of the slip surface (Fig. 7). The other two values assumed are
a/~ = 2/3 and a/~ = 0. Horizontal and vertical accelerations
were input according to which av/~ was used, and factors of safety
were determined. Plots were made of the factor of safety versus horizontal
acceleration. By entering the graphs at a factor of safety of 1, the
critical horizontal acceleration can be read off the acceleration axis.
Critical horizontal accelerations were then determined for a range of
values of friction angle, unit weight and the ratio of av/~. For
these determinations each of the design parameters were varied
independently while the other two parameters were held constant (Table 7).
The variation of critical acceleration for the main dam with each variable
is illustrated by Figure 9.
4329R/CG 6-7
6.5 SARMA ANALYSIS
The dynamic stability analysis was conducted using the SARMA computer
program on the six slip circles chosen by the LEASE II analysis. The
circles were subjected to the Hybrid and Taft earthquake records described
previously. The analysis was run with the range of values given previously
(Table S) for each of the input parameters. The case using the average, or
most likely values, of all the input parameters was designated the "normal"
case. Also, a "worst case" condition, containing the value of each input
parameter which resulted in the greatest deformation, was run. The results
of the analysis are presented in Table 7, along with data from the Taft
earthquake (scaled up) for comparison. The influence of the various input
parameters on the critical accelerations and vertical displacements of the
various slip circles is shown graphically in Figures 9 and 10.
For circles B, D, and E, representative critical case plots of displacement
and wedge response versus time for various cases are shown in Figure 11.
The plots indicate progressive slip-displacement cycles of each wedge.
These circles were selected as the critical modes which could result in
loss of freeboard or rupture of the concrete face.
6.6 INTERPRETATION OF RESULTS
The LEASE II analysis of the embankment showed a lowest minimum static
factor of safety of 1. 7, for circle C, a shallow circle through the
downstream berm. It should be noted that the embankment geometry used in
4329R/CG 6-8
the original
berm. The
analysis ("old geometry") did not include the downstream
larger, more
safety of 1.8 or higher.
under static conditions.
significant circles showed minimum factors of
The analysis shows that the embankment is stable
The maximum displacement calculated from the Sarma analysis for the MCE in
the normal case was 3.2 feet vertical and 5.7 feet horizontal. Even the
"worst case", with low 0 angle, low density and vertical acceleration
equal to 2/3 of the horizontal acceleration yielded total vertical and
horizontal displacements of only 4.7 and 8.4 feet respectively. The design
criterion requires that half of the available 10 feet of freeboard be
maintained, so the calculated vertical displacement is acceptable. The 5.7
feet of estimated horizontal movement is also acceptable because even if it
is considered to be the offset of the upstream face, it would leave a
signifi-cant part of the minimum 12 foot bedding layer intact. It should
be noted that the analysis assumes that the movement will occur along one
slip surface, while in an actual earthquake-induced movement, this type of
dam is more likely to experience settlement and bulging. For this reason
the horizontal displacement should be considered only a very rough
approxima-tion, with the actual horizontal movement being less. The
analyses show that the embankment will respond satisfactorily to earthquake
loadings produced by the MCE or any lesser event.
4329R/CG 6-9
6.7 SPECIAL STUDIES
A number of special studies were performed to see how the embankment
responded to smaller or different earthquakes, post-earthquake conditions,
or changes in slip surface or embankment geometry. These studies and their
results are explained below.
6.7.1 Megathrust (~ = O.SSg)
To simulate the effects of a large earthquake in the Benioff Zone, (the
megathrust event), a Sarma analysis was conducted using the Hybrid and Taft
earthquake records normalized to a peak acceleration of O.SSg. The
analysis used the same circles as the analyses conducted with a peak ground
acceleration of 0. 75g. The maximum loss of freeboard (vertical displace-
ment) was 1.4 feet, which is well within the criteria and approaching
insignificant damage for this type of darn. The results are presented in
Table 7 and plots of the displacement versus time curve and the earthquake
accelerograrn for the selected design cases are shown in Figure 12 for
circles B, D, and E. These plots are presented as the "design" estimates of
response of the circles which could be critical to darn safety. It can be
seen by inspection that the displacement is limited to about 30 "pulses" of
motion. The number of pulses of motion is illustrated on the displace-
ment/elapsed time plot. Each displacement is in response to an
acceleration on the accelerograrn plot where wedge acceleration exceeds the
dashed line representing critical acceleration.
4329R/CG 6-10
The records, as used, do not completely match the criteria suggested by wee
since the significant length is about 25% short. However, even if the
records were doubled in length, the deformation would not equal that
resulting from the nearby crustal faulting (peak ~ = .75g). Therefore,
it is reasonable to conclude that the peak ~ = .75g event is the
critical case. On that basis, the megathrust event will not be examined
further.
6.7.2 DBE (~ = .375g)
In order to determine the response of the embankment when subjected to the
Design Basis Earthquake (DBE), a Sarma analysis was conducted using the
Hybrid and Taft earthquake records normalized to a peak ground acceleration
of 0.375g. The analysis used the same critical circles as the analysis
conducted with a peak ground acceleration of 0. 75g. The maximum loss of
freeboard (vertical displacement) was 0.4 feet. Subjected to the design
basis earthquake, the embankment, and particularly the concrete face,
would probably need some repairs, but it would remain safely operational.
The results are shown in Table 7 and plots of the displacement versus time
curve and the earthquake accelerogram for circles B, D, and E are shown on
Figure 13.
It should be noted that the design criteria specify a DBE peak ground
acceleration of .35g. The 0.375g peak ground acceleration (which reflects
a preliminary criterion value) used should result in movements only
slightly greater than the .35g acceleration; therefore, it was decided not
to redo the computer runs.
4329R/CG 6-11
6.7.3 Influence of Downstream Berm
The effect of the downstream berm on the stability of the embankment was
evaluated using the LEASE II program. In order to directly compare the
factors of safety of the embankment with and without the berm, two computer
runs were made, one with the berm and one without, with all other
parameters being constant. This analysis was performed with a slope on the
berm of 1.25H:lV. The slope was changed to 1.6V:lH in the final design,
but that should not change the conclusions below.
In the LEASE II analysis, the downstream berm was shown to have a very
small effect on the stability of the embankment. For most of the potential
slip circles the factor of safety is essentially unchanged. The factor of
safety of the critical circle having the lowest factor of safety did not
change without the berm. This is an indication that a dynamic analysis of
the embankment without the berm would not change significantly from the
embankment with the berm. Based on this reasoning, a dynamic analysis was
not performed on the embankment without the berm. In the final design, the
berm has the same slope and same composition as the rest of the dam and so
is not expected to behave any differently.
6.7.4 Failed Concrete Face
A special study was performed to estimate the effects on the stability of
the downstream slope, of seepage through the embankment caused by a
hypothetically completely failed concrete face. A flow net was drawn,
4329R/CG 6-12
assuming no concrete face, to determine the effect of seepage on circles A
and B, described previously. In modeling the embankment, the effective
height was taken as 130 ft., the bedding layer was 12 ft. horizontal and
the downstream berm was ignored. It was felt to be appropriate to ignore
the downstream berm based on the study of the effect of the berm on the
stability of the downstream slope. The downstream berm should have little
effect on the flow-net since it is composed of oversized rock fill. Two
assumptions made in drawing the flow net were that the coefficient of
permeability of the rockfill is 10 times that of the bedding layer, and the
horizontal permeability of the fill is 10 times the vertical permeability.
The flow net was drawn following the transformed section construction
techniques outlined by Cedergren (Ref. 13) and by Casagrande (Ref. 14).
Figure 14 shows the final flow net including circles A and B.
From the figure it can be seen that seepage would not affect circle A at
all and would affect circle B only slightly. The seepage would result in
pore pressure along a small part of circle B. Since driving and resisting
forces would both be affected it would have a minimal effect on the factor
of safety. Under these conditions seepage would not adversely affect the
stability of the embankment.
6.7.5 Varying Embankment Height
In order to illustrate the effect of the embankment height on the slip
circle displacement, a series of SARMA runs was made using different
heights of the embankment. Only circle D remained within the embankment
4329R/CG 6-13
through the whole range of heights, so it was the only circle considered.
The vertical displacement of circle D went from 1.20 feet at a height of
145.6 feet to .03 feet at a height of 30 feet, showing that the slip circle
displacement decreases as the embankment height decreases, as might be
expected due to lower seismic amplification in the fill. Similarly, peak
acceleration generally decreased with embankment height, but not in as
consistent a manner due to modal effects which result in amplification
variations over the range of fill height (Fig. 15).
6.7.6 Planar Slip Surfaces
The variation in results of the Sarma method at different levels within the
embankment was investigated. Two approaches were taken. To estimate
maximum potential offset of the dam upstream face at various levels, sloped
planes through the toe of the downstream slope were first analyzed (Fig.
16). These slip surfaces through the downstream toe should give
approximately the greatest deformations that would be expected to result
from movement of various portions of the embankment. To provide
information on maximum accelerations at various levels within the
embankment, horizontal slip surfaces were also analyzed (Fig. 17).
Statics analysis was used to determine the static factor of safety of each
of the wedges, as well as the critical horizontal acceleration. The SARMA
program was used to model the seismic response of the wedges. The Hybrid
earthquake record was the only earthquake used in this analysis.
4329R/CG 6-14
The analysis of the sloping planes through the downstream toe showed a
maximum displacement of 2.3 feet vertically and 3.9 feet horizontally. Data
presented (Fig. 16) includes the variation of static factor of safety,
critical horizontal acceleration, and vertical and horizontal displacement
for each case. The static factor of safety decreased as elevation
increased, as would be expected due to the steeper overall slopes of the
slip surfaces. The critical horizontal acceleration also decreased with
increasing elevation, leading to the increasing displacements. The face
offset due to earthquake loading approximated by this type of analysis
might cause cracking of the concrete face but the bedding layer would not
be breached. It should be noted that below elevation 1180 (normal maximum
operating level) the displacements rapidly decrease 1n magnitude,
indicating that while measurable crest movements can be expected,
deep-seated instability is not a concern.
The analysis of the horizontal slip surfaces showed a maximum wedge
acceleration at the top of the embankment. The maximum wedge acceleration
increases with elevation (Fig. 17). This figure also shows the variation
of factor of safety, critical acceleration and displacement with
elevation. The static factor of safety for horizontal planes above the
water level is infinity because there is no driving force. It should be
noted for this case that the ratio of vertical to horizontal acceleration
is 2/3, resulting in very high relative seismic driving force. Again,
below elevation 1180 the net displacement drops off rapidly.
4329R/CG 6-15
The Sarma method generally does not take into account any dynamic water
load. The effect of dynamic water load was checked using Zangar' s method
and horizontal sliding planes. It was found that, due to the slope of the
impervious face, the seismic water load has a slight stabilizing influence
rather than causing a reduction of stability. On this basis, it was
neglected in the remainder of the analysis.
6.7.7 La Union Accelerogram
A special effort was made to utilize a record from the Michoacan, Mexico
earthquake of September 19, 1985 (Ref. 15) for comparison. This was
reported as a magnitude 8.1 event, and is one of the few great earthquakes
with good bedrock seismograms. The available records were scanned and the
one chosen was the La Union E-W accelerogram, which has a significant
duration of 26.4 seconds and a peak ground acceleration of O.lSg. The
recording station is about 52 miles from the epicenter.
The La Union record was scaled up to have a peak ground acceleration of
0.75g and applied to circle B. As shown in Fig. 18, the total movement
predicted was 4.9 ft. with a predicted loss of freeboard of 2.0 ft. This
is about 78% more movement than predicted using the Hybrid record, so the
results require some explanation. There are some key differences between
the La Union and Hybrid accelerograms which account for the increase in the
amount of predicted movement and make the Hybrid record more appropriate
for the current analysis.
4329R/CG 6-16
One key difference is that the magnitude of the Michoacan event was 8.1,
which is significantly larger than the magnitude 7.5 event specified as the
MCE. Among other effects, this results in a much greater number of
relatively large acceleration spikes in the La Union record. The
difference is obvious upon comparison of the accelerograms (Figs. 11 and
18, note difference in time scales).
A related effect is that the Arias Intensity (Ref. 16) of the La Union
record is 21.7 when it is amplified to a peak ground acceleration of
0.7Sg. The Hybrid record has an Arias Intensity of only 8.8 at the same
ground acceleration level. This indicates that the potential for damage
which could occur from an earthquake with the La Union accelerogram was
almost two and a half times that which the Bradley Lake project design
earthquake could produce.
Another key difference is in the epicentral distance of the two records.
The Bradley Lake MCE is supposed to be caused by rupture of one of the
faults on or very near the site. Thus, the Hybrid record, constructed to
approximate near field accelerograms, is more appropriate than the La Union
record, which was made 52 miles from the epicenter. The difference can be
seen in the response spectra (Figs. 5 and 19). At greater distances,
proportionately more of the high frequency content of the earthquake is
damped out. Thus, scaling a record from some significant distance up to
the same ground acceleration as a near field record results in a
disproportionate amplification of the lower frequency portion of the
record. This is the reason the La Union record is below the mean response
4329R/CG 6-17
spectrum at very short periods and significantly above it at most periods
of significance. The Taft record, which was utilized in unmodified form
for parametric analysis has the same problem to a slight extent. Its res-
ponse spectrum appears in Figure 20. A plot of the Arias Intensity of
these three records is shown in Figure 21. This figure illustrates the
difference among the three records which result from sealing low
acceleration records to high accelerations.
The object of this analysis was not to find the worst imaginable case, but
rather to find the worst appropriate case. The results based on the Hybrid
record are considered to be the best indication of the maximum deformations
which could be anticipated because this accelerogram has been produced to
fit the site conditions.
6.7.8 Parametric Analyses
As a part of this analysis, input parameters were varied over the range
considered reasonable for each parameter. This allowed examination of the
sensitivity of the analysis to changes in the input data. Most of the
resulting information of significance is contained in Figures 9 and 10 in
the form of graphs. These graphs visually present the information
effectively so the writeup of these analyses will be abbreviated.
Friction angle, av/~ and damping ratio all
tion in the way that would normally be expected.
affect permanent deforma-
Their effects appear to
be minor to moderate in magnitude. Unit weight of the fill within
reasonable limits generally does not have a significant influence. Shear
4329R/CG 6-18
wave velocity has what appears at first to be an odd effect on
displacement. However, it should be noted that changes in V have the s
effect of changing the natural period of the dam. This causes the natural
period to fall on different parts of the response spectrum as the velocity
is changed. The V versus movement curves are simply a reflection of the s
shape of the response spectrum for the Hybrid record. Circle C is not
affected in the same way since it is near the bottom of the dam and
responds at different frequencies than the other circles. Table 8 includes
the results of the SARMA runs based on a previous dam geometry ("old
geometry") without a downstream berm. It includes results from varying
V over a wider range than was utilized in the final design runs. s
As expected, the parameter that has the greatest effect on the amount of
deformation is the peak ground acceleration. As the peak acceleration
decreases by a factor of 2, the displacement drops by a factor of 7 or more.
Input not covered by the above graphs includes the earthquake accelerogram
and the embankment slope. Three accelerograms were used in the current
study. Of these the La Union record is the least appropriate, as discussed
in Sect ion 6. 7. 7. The Taft and Hybrid records are related to each other.
The Taft record has responses significantly above the mean response
spectrum throughout the period range of interest. It was modified by
insertion of a small segment from a similar record (Friuli) which had a
slightly higher peak acceleration. The resulting record (the Hybrid) has a
response spectrum that matches the mean curve fairly well, with the
exception of one prominent spike in the period range of interest.
4329R/CG 6-19
The results obtained using the Hybrid and Taft records are very similar, as
can be seen by examination of Tables 7 and 8, and by comparing Sheet 1 of
Figure 11 (Hybrid MCE for circle B) to Figure 22 (Taft MCE). The Hybrid
record is considered a better representation of the MCE, and was used for
design analyses.
The remaining input parameter, slope angle, was picked by engineering
judgement based on a number of factors, including infinite slope stability
factor of safety, site seismicity, precedents provided by other similar
dams and constraints created by the configuration of the site.
6 • 8 COFFERDAM
A static and dynamic stability analysis was conducted on the upstream
cofferdam. The general geometry of the cofferdam is shown on Figure 4.
The stabi 1i ty analyses were based on an "infinite slope" approach. The
static factor of safety was 1.35 and the factor of safety for the
construction earthquake was 1.1. The stability and
effectiveness of this cofferdam will be increased by the placement of waste
fill on its upstream side.
4329R/CG 6-20
7.0 CONCLUSIONS
7.1 CRITICAL CASES
In an analysis such as this there are numerous modes of movement that can
be considered. At tent ion must be directed toward the modes that are most
1 ikely to occur and the ones with the greatest potential consequences.
Some of the more relevant modes are sliding of the whole dam on its base,
surficial ravelling or slumping of slopes, general settling and bulging of
the dam, and movement on one or more well defined planar or circular
surfaces.
Of these various modes of movement, sliding of the whole dam on its base is
of little interest because it is extremely unlikely, being far from a
critical failure surface. In fact, part of the analysis showed that even
the MCE would produce no permanent deformation at the base of the dam due
to the overlying fill. At the other extreme, surficial raveling or minor
slumping of faces is much more likely, but it is of little interest because
it poses no threat to the integrity of the embankment as a whole.
Since only modes that have a significant connection to the overall
integrity of the dam or loss of freeboard are of interest, the remaining
two modes deserved the most consideration. Settling and bulging of slopes
is the more likely of the two.
4329R/CG 7-1
This would result in some loss of freeboard and in bending and cracking of
the concrete face. Movement along a well defined surface is significantly
less likely but would result in at least as much loss of freeboard and
probably a great deal more damage to the concrete face. For these reasons,
this last mode of movement was given the most attention with the assumption
that it would serve as a limiting or extreme case as far as effect on the
embankment, hence resulting in a more conservative design.
7.2 SUMMARY OF CRITICAL FAILURE SURFACES
Of the various types of surfaces along which movements could occur, only
certain ones are of interest. Planes or arcs that intersect the concrete
face below headwater level, and those that could remove all freeboard with
sufficient movement are· of the greatest concern.
Six circular arcs and numerous planes were considered in the present
analysis. Circles and planes that slope downstream and intersect the
upstream face below the water level were given considerable at tent ion. Of
these, circle B (Fig. 8) was considered most critical, since it intersects
the concrete face slightly below elevation 1180 feet and has one of the
steepest inclinations possible within the embankment. Planar surfaces 2 and
3 on Figure 16 are similar to it but were not analyzed as extensively.
Some attention was also devoted to upstream sloping surfaces. Because of
the supporting effects of the reservoir water, only circles almost
completely above headwater level were predicted to move significantly at
4329R/CG 7-2
full reservoir. Of the upstream-sloping surfaces, circle E was considered
most critical since it showed the highest amount of movement. Circle E
assumes the minimum normal operating level, el. 1090 ft., so the circle is
mostly above headwater level and receives little support from it.
Therefore, for the retained pool case, which is critical for dam safety,
the upstream face is more stable than the downstream. In the low-pool
case, the upstream slope is similar to the higher and identical downstream
slope, with the primary difference being in the lower 0 angle of the
bedding material, which will contribute to slightly greater surficial slip
surface motions.
7.3 PREDICTED DISPLACEMENTS
Each of the circular arcs chosen for analysis was analyzed with a range of
values for each input parameter, as previously explained. The calculated
permanent displacements, as well as the input values used, are given in
Table 7 and illustrated in Figures 10 through 13.
For the Maximum Credible Earthquake, circle B shows a maximum loss of
freeboard of 1.8 feet and total movement of 4.64 feet. Using the expected
values, rather than worst values, of the previous parameters results in
vertical and total movements of 1.1 and 2.76 ft., respectively. This is
comparable to the results calculated for planes 2 and 3, Figure 16.
With a peak ground acceleration of O.SSg, comparable to the megathrust
event, circle B moved 0.4 ft. vertically and 1.0 ft. total. This is a
reduction by a factor of about 2.7 from the MCE. Movement with a peak
4329R/CG 7-3
ground acceleration of 0.375g, slightly above the DBE, was 0.1 ft. vertical
and 0.22 ft. total. The movement is less than that caused by the MCE by a
factor of about 12.5 and is considered essentially insignificant and
unlikely to result in structural damage.
The other circles and planes on the downstream slope of the dam were
examined, but more steeply sloped surfaces failed to intersect headwater,
and the less steep surfaces did not move as much according to the
calculation.
On the upstream slope of the dam, circle E was predicted to have a maximum
vertical movement of 4.8 ft. and total movement of 9.6 ft. as a result of
the MCE. Using the. expected values of input parameters resulted in
movements of 3.2 ft. vertical and 6.5 ft. total. The 0.5Sg event produced
vertical movement of 1.4 ft. and total movement of 2.9 ft. These are about
a factor of 2.2 less than resulted from the MCE. The 0.375g event produced
vertical movement of 0.4 ft. and total movement of 0.9 ft. These are about
a factor of 7.1 less than resulted from the MCE.
Other surfaces on the upstream slope did not exhibit as much predicted
movement as circle E. One major factor causing circle E to show more
movement than other surfaces, including downstream-sloping ones, is the
influence of the lower friction angle used for the bedding layer. However,
considering the dependence of circle E on headwater level, the most likely
critical surface is circle B on the downstream slope of the dam for
retained pool cases.
4329R/CG 7-4
From the above information it can be seen that the amount of movement, and
therefore damage, drops off sharply with a decrease in peak ground
acceleration from 0. 75g. This knowledge is useful when considering the
relative probability of occurrence of the three seismic events used in the
calculation. It should also be noted that the greatest vertical movement
predicted would result in a loss of less than one half the freeboard,
without considering the added height of the parapet wall.
In addition to calculating maximum displacements and examining how they
vary with different parameters, some effort was devoted to evaluating how
peak acceleration and displacement vary with height above the base of the
dam. Analysis of horizontal surfaces (Fig. 17) within the dam showed that
peak acceleration varied from 0.75g at the base to a maximum of 2.7g at the
crest of the dam. Similarly, permanent displacement varied from 0 to 2.2
ft. in the downstream direction. It can be concluded that the response of
the dam to any seismic event will be minimal near the base and most severe
at the crest. This would also suggest that following a large earthquake,
some reshaping of the crest may be required. Parapet displacement may
result in the need for concrete repair, but the parapet is designed to
remain upright, with only translational motion of an estimated 2 - 3 ft.
Another area of interest is the probable effect of an earthquake on the
concrete face of the dam. First, it should be noted that even if dam
displacements were restricted to a single surface, the maximum predicted
movement would not breach the integrity of the bedding layer beneath the
concrete face. This is significant because the bedding layer is intended
to restrict flow through the embankment in the event of damage to the
4329R/CG 7-5
concrete face. For this reason it is graded to have a lower permeability
than the rockfill and to avoid piping into the rockfill. Thus, even the
most extreme horizontal movement predicted would not produce enough leakage
to affect the stability of the dam.
Further, as has been pointed out previously, it is much more 1 ikely that
deformation of the dam would take the form of movement distributed over
numerous surfaces or of general settling and bulging of slopes. The effect
of these more likely modes of deformation on the concrete face would be
bending and cracking rather than large discrete offsets. Thus, the more
likely modes of deformation would result in even less leakage and in more
readily repairable damage to the face.
In addition, the dam face joints have been designed to accommodate, without
damage, movements up to those predicted for the DBE. The joints are
expected to survive in functional, if not intact form, movements several
times that severity.
7.4 RESPONSE TO VARIOUS EVENTS
It has been shown that the embankment will remain stable, meeting the
design criteria, even in the case of the 0.75g MCE. Most of the response
of the dam was shown to take place in the upper part of the embankment.
This would remain true regardless of the magnitude of the earthquake. For
the DBE it was shown that the resulting deformation was relatively minor,
dropping off by a factor of roughly 7 to 12 from deformations caused by the
2-1223-JJ 7-6
MCE. This was based on use of 0.375g rather than 0.3Sg, and a longer
accelerogram than should be necessary for the DBE, so these values are
conservative.
With respect to the MCE, it should be remembered that the . 75g crustal
fault event is extremely unlikely compared to the . SSg mega thrust event.
The . 75g event is based on movement on existing faults at the site, but
there is no evidence that these faults are active and the microseismic
stations in the area have indicated no activity attributable to them.
Thus, it is likely that the most severe earthquake the project will
experience over any length project life is the mega thrust event with peak
ground acceleration of .SSg. This would produce only about one quarter the
amount of deformation predicted for the MCE so the maximum loss of
freeboard expected would be 0.4 ft. and 1.4 ft. for circles B and E,
respectively. These levels of movement would not be likely to cause damage
other than cracking and minor leakage. The leakage and loss of freeboard
should not be significant to stability of the dam.
It is therefore concluded the dam and cofferdam can readily serve their
intended purpose through events up to the DBE and construction case
earthquake, respectively, and will remain structurally intact (though
potentially aesthetically and functionally damaged) up to the MCE event.
4329R/CG 7-7
8.0 BIBLIOGRAPHY
1) Woodward-Clyde Consultants, Report on the Bradley Lake Hydroelectric
Project, Design Earthquake Study, submitted to Alaska District, Corps
of Engineers, 10 Nov 1981.
2) Woodward-Clyde Consultants, Seismicity Study, Bradley Lake
Hydroelectric Project, submitted to Alaska District, Corps of
Engineers, 28 March 1980.
3) Wilson, S.D. and R.J. Marsal ed., Current Trends in Design and
Construction of Embankment Darns, ASCE, 1979.
4) Hausner, G.W. et al, Safety of Darns: Flood and Earthquake Criteria,
National Academy Press, 1985.
5) Newmark, N.M., Effects of Earthquake on Darns and Embankments, Fifth
Rankine Lecture in Geotechnique Vol X'J No. 2, Institution of Civil
Engineers, 1965.
6) Bureau, Gilles et al, "Seismic Analysis of Concrete Face Rockfill
Darns" 1n Concrete Faced Rockfill Dams-Design, Construction, and
Performance, edited by J.B. Cooke and J.L. Sherard, ASCE, 10/21/85.
7) "Slope Stability Analysis, LEASE II," SWEC Program GT018, Version 01
Level 00, August 1980.
4329R/CG 8-1
8) "Seismic Amplification Response by Modal Analysis, SARMA," SWEC
program GTOSS Version 01 Level 00, September 1986.
9) Lepps, Thomas, M., "Review of Shearing Strength of Rockfill, Journal
of the Soil Mechanics and Foundations Division ASCE Vol. 96, No. SM4,
July, 1970.
10) Seed, H. B. et. al, "Seismic Design of Concrete Faced Rockfill Dams"
in Concrete Faced Rockfill Dams Design, Construction, and
Performance, edited by J. B. Cooke and J. L. Sherard, ASCE, 10/21/85.
11) Arias, A., "A Measure of Earthquake Intensity" in Seismic Design for
Nuclear Power Plants, edited by R. J. Hansen, MIT Press, 1970.
12) Seed, H. B. and Idriss, I.M., "Soil Moduli and Damping Factors for the
Dynamic Response Analysis", Report No. EERC 70-10, EERC, Berkley, CA.,
1970.
13) Cedergren, H., Seepage, Drainage and Flow Nets, 2nd edition, Wiley &
Sons, 1977.
14) Casagrande, A., Seepage Through Dams, Journal of the New England Water
Works Association, Vol. LI., June 1937.
4329R/CG 8-2
15) Anderson, J.G., Bodin, P., Brune, J. N., Prince, J., Singh, S. K.,
Quaas, R., and Onate, M., "Strong Ground Motion from the Michoacan,
Mexico, Earthquake," Science, Vol. 233.
16) National Oceanic and Atmospheric Administration, Earthquake History of
the United States, Publication 41-1, Reprinted 1982.
4329R/CG 8-3
Crest Elevation
Crest Length:
Crest Width:
Maximum Height:
Slopes:
Materials:
4329R/CG
TABLE 1
MAIN DAM CHARACTERISTICS
1190.0 feet
602.5 feet
18.0 feet (Inside of Parapet to edge
of dam)
120 feet
1.6H: lV
U/S Concrete face, 1 ft. thick,
reinforced both ways
Bedding layer beneath face (Zone 1)
12 ft. horizontal thickness
minus 3" well-graded, angular gravel
Remainder of dam-blasted rockfill (4 Zones)
Zone 2 Coarse base drain, 36 inch max.
size
Zone 3 Upstream shell, approx. 36 inch
max. size
Zone 4 Downstream shell, approx.
36 inch max. size
Zone 5 Downstream face slope
protection, 48 inch max. size
TABLE 2
MAIN DAM COFFERDAM CHARACTERISTICS
U/S Cofferdam
Crest Elevation: 1090 ft. min.
Crest Length: 200 ft.
Crest Width: 18 ft.
Maximum Height: 32 ft.
Slopes: U/S 2H = lV
DIS 2H = lV
Materials: Random fill with geomembrane and
geotextile protection
4329R/CG
Date
Aug. 24, 1898
July 11, 1899
Oct. 7, 1900
Oct. 9, 1900
Dec. 30 and
Dec. 31 , 1901
Sept. 19, 1909
Sept. 21, 1911
Jan. 31, 1912
June 6 and
June 10, 1912
June 21, 1928
TABLE 3
(REF 16)
HISTORIC SEISMIC EVENTS SUMMARY
(SUMMARY OF LARGER EARTHQUAKES OCCURRING WITHIN A FEW
HUNDRED MILES OF THE HOMER AREA IN THE PERIOD 1788-1980)
Local
time
0228
1000
1901
1012
2356
0606
0627
Approximate location
(lat. N.; long. W.)
Valdez
(61°, 146°)
Tyonek ( 61 °, 151 o)
do
Chugach Mountains Area
(60 ' 142°)
Kenai, on Cook Inlet
(60.5°, 151°)
Kenai Peninsula
Prince William Sound
and Kenai Peninsula
(60.5°, 149°)
Prince William Sound
(61°, 147.5°)
Cook Inlet
( 59°, 15 3°)
South-central Alaska
(60°, 146.5°)
Estimated
Richter
magnitude
(Msl
8.3
7.4
6.9
7.25
6.4 & 7.0
7.0
Page 1 of 3
Remarks
Heavy.
Severe.
Severe. Probably is
same as following one
Severe, felt from
Southwestern Yukon
Territory, Canada to
Kodiak
Volcanic eruption and
several sea waves.
Strong at Seward.
Severe, felt at
Kenai Lake.
Strong at Valdez.
Severe.
Ground waves at
Seward.
TABLE 3 Page 2 of 3
Estimated
Richter
Local Approximate location magnitude
Date time ( 1 at. N . ; 1 ong . w. ) (Msl Remarks
Mar. 25, 1932 1359 South-central Alaska 6.9 Water main burst at
(62.5°, 152.5°) Seward.
Sept. 13, 1932 2243 Prince William Sound 6.25 Stopped clocks in
and Kenai Peninsula Homer, Valdez and
( 61°. 1480) Wasilla.
Oct. 6, 1932 0705 Homer --------------Awakened all.
(59.5°, 151.5°)
Apr. 26, 1933 1703 Homer 7.0 At Homer, worst shock
(59.5°, 151.50) in 15 years.
June 13, 1933 0219 Old Tyonek 6.25 Severe shock with
(61°, 151°) some damage.
June 17, 1934 2314 South-central Alaska 6.75 Damage at Anchorage.
(60.5°, 151°)
July 29, 1941 1551 Kenai Pensinsu1a Area
(61°, 151°)
6.25 Damage at Anchorage.
Sept. 27, 1949 0531 South-Central Alaska 7.00 Strong aftershock
(59.750, 1490) also. Damage at
Seward and Anchorage.
June 25, 1951 0613 Chickaloon Bay
(61°, 150°)
6.25 Damage at Anchorage.
Oct. 3, 1954 0119 Kenai Peninsula 6.75 Damage at Homer.
( 60. 5°. 151°)
TABLE 3 Page 3 of 3
Estimated
Richter
Local Approximate location magnitude
Date time {lat. N.; long. w.) (Msl Remarks
Sept. 5, 1961 0135 Kenai Peninsula 6-6.25 Felt. Anchorage
(600, 150.6°) rocked.
June 23, 1963 1827 Cook Inlet 6.75 Damage at Homer,
(59.5°. 151. 70) Barbara Point and
Seldovia.
March 27, 1964 1736 Prince William Sound 8.5 Loss of life, severe
( 61°. 14 7. 70) damage.
April 14, 1964 1256 Kodiak Island Region 4.5-4.75 Damage at Kodiak.
(580, 152.6°)
Aug. 30, 1966 1021 South-central Alaska 5.75-6.0 Damage at Anchorage.
1023 (61.3°, 1476.50)
Dec . 17, 1968 0202 Southern Alaska 6.5 Slight damage at
(60.2°, 152.8°) Kenai and Ninilchik.
Jan. 15, 1970 2206 Southern Alaska 6.1 Shocks felt on Cook
(60.3°, 152.7°) Inlet and Kenai
Peninsula.
TABLE 4
DESIGN FACTORS OF SAFETY
Load Case
Normal Pool Level:
Dead + Live +
(Wind or ice)
Dead + Construction
Dead + MCE Earthquake
PMF Pool Level:
Dead + Live
Seismic Loss of Freeboard
(Max Loss/Total Freeboard)
Infinite Slope Stability
Static
Operational Drawdown
Emergency Drawdown
4329R/CG
Main Dam
1.5
1.2
N/A
1.5
< 50%
(MCE)
1.5
1.2
1.0
Cofferdam
1.2
1.05
N/A
N/A
N/A
1.2
N.A.
N/A
Variable Name
Friction angle (0)
rockfill
bedding layer
Unit weight,
moist
saturated
Shear wave velocity, Vs
Damping ratio, D.R.
TABLE 5
INPUT PARAMETERS
125-150 pcf
138-154 pcf
700-1000 fps
12%-20%
0-2/3
* e = slope angle of sliding surface
4329R/CG
Most Likely Value
135 pcf
145 pcf
800 fps
15%
tan (0-9)*
TABLE 6
MAIN DAM STATIC STABILITY SUMMARY
Loading Condition Definition
Normal Maximum Operating H.W. el. 1180
T. W. el. 1065
Normal Minimum Operating H.W. el. 1090
T.W. el. 1065
Design Flood-PMF H.W. el. 1190
T.W. el. 1082
Safety
U/S
1.66
1.66
1.66
Factors
DIS
1. 78
1. 78
1. 78
Criteria
Minimum
1.5
1.5
1.5
End of Construction]
Rapid Drawdown ]
These cases reduce to Normal Minimum oper-
ating case due to features of the upstream
cofferdam which result in headwater
retention at El 1090 even if lake level
drops lower.
Factors of safety calculated by "infinite slope"
analysis. Failures of significance to the integrity of
the dam (i.e., penetrating water retention zone) have
higher safety factors . as shown on subsequent figures
(e.g., Figure 16). U/S analysis neglects support from
concrete face slab and headwater.
H.W. = Reservoir Headwater Level
T.W. = Dam Toe Tailwater Level
4329R/CG
TABLE 7 page 1 of 6
Main Dam SARMA Results
CIRCLE A, DOWNSTREAM, H.W. El. = 1180 ft., T.W. El. = 1061 ft.
CENTER POINT (596.0, 1450.0), RADIUS 347.13
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-0 (Degrees) Density Static D.R. Vs Peak v h v h
meter Varied Rockf111/Bedding (pcf) F.S. av/ah a he % (fps) Ah• (ft) ( ft) ( ft) (ft)
Normal Case 48/46 135 2.0 0.335 0.31 15 800 1. 78 2.8 5.0 3.0 5.4
" 45/44 135 1.8 0.277 0.275 15 800 1. 78 3.4 6.0 3.6 6.3
" 50/48 135 2. 1 0.374 0.345 15 800 I. 78 2.4 4.2 2.5 4.5
Density 48/46 125 2.0 0.335 0.31 15 800 1. 78 2.8 5.0 3.0 5.4
Density 48/46 150 2.0 0.335 0.31 15 800 1. 78 2.8 5.0 3.0 5.4
av/ah 48/46 135 2.0 0 0.355 15 800 1. 78 2.2 4.0 2.4 4.3
avlah 48/46 135 2.0 2/3 0.285 15 800 1. 78 3.2 5.7 3.4 6. 1
D.R. 48/46 135 2.0 0.335 0.31 12 700 1.58 2.5 4.5 3.3 5.9
D.R. 48/46 135 2.0 0.335 0.31 20 700 1.16 1.3 2.3 1.7 3.0
Vs 48/46 135 2.0 0.335 0.31 15 700 1.40 1.9 3.4 2.5 4.4
vs 48/46 135 2.0 0.335 0.31 15 900 1.90 3.2 5.7 3.4 6. 1
Vs 48/46 135 2.0 0.335 0.31 15 1000 1. 73 2.7 4.8 3.8 5.7
Worst Case 45/44 125 1.8 2/3 0.245 15 900 1.90 4.3 7.6 4.5 7.9
Megathrust .55g 48/46 135 2.0 0.335 0.31 15 800 1. 31 1.2 2.2 1.4 2.4
DBE .375g 48/46 135 2.0 0.335 0.31 15 800 0.89 0.4 0.7 0.5 0.8
.. Peak horizontal acceleration of the failure circle.
TABLl page 2 of 6
Main Dam SARMA Results
CIRCLE B, DOWNSTREAM, H.W. El. = 1180 ft., T.W. El. = 1061 ft.
CENTER POINT (596.0, 1450.0}, RADIUS= 366.51
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-foJ (Degrees) Density Static D.R. Vs Peak v h v h
meter Var'led Rockfill/Bedding (pcf> F.S. av/ah ahc % (fps) Ah'" ( ft) (ft) (ft) ( ft)
48/46 135 2.34 0.456 0.375 15 800 1. 35 1.1 2.5 1.2 2.7
foJ 45/44 135 2.11 0.394 0.340 15 800 1. 35 1.3 3.0 1.4 3.3
0 50/48 135 2.52 0.499 0.400 15 800 1. 35 1.0 2.2 1.1 2.5
Density 48/46 125 2.34 0.456 0.375 15 800 1. 35 1.1 2.5 1.2 2.7
Denslty 48/46 150 2.34 0.456 0.375 15 800 1. 35 1.1 2.5 1.2 2.7
av/ah 48/46 135 2.34 0 0.435 15 800 1. 35 0.8 1.9 1.0 2. 1
av/ah 48/46 135 2. 34 2/3 0.345 15 800 1. 35 1.3 2.9 1.4 3.2
D.R. 48/46 135 2.34 0.456 0.375 12 700 1. 10 0.8 1.8 1.1 2.6
D.R. 48/46 135 2.34 0.456 0.375 20 700 0.89 0.4 1.0 0.5 1.3
Vs 48/46 135 2.34 0.456 0.375 15 700 1. 00 0.6 1.4 0.9 2.0
vs 48/46 135 2.34 0.456 0.375 15 900 1.50 1.3 3.0 1.5 3.5
Vs 48/46 135 2.34 0.456 0.375 15 1000 1.39 1.2 2.6 1.4 3.2
Worst Case 45/44 125 2.12 2/3 0.305 15 900 1.50 1.8 4.2 2.1 4.7
Megathrust .55g 48/46 135 2.34 0.456 0.375 15 800 0.99 0.4 1.0 0.4 1.1
DBE .375g 48/46 135 2.34 0.456 0.375 15 800 0.67 0. 1 0.2 0. 1 0.2
,. Peak horizontal acceleration of the failure circle.
TABLt.. page 3 of 6
Main Dam SARMA Results
CIRCLE C, DOWNSTREAM, H.W. El. ~ 1180 ft., T.W. El. : 1061 ft.
CENTER POINT (696.67, 1249.33), RADIUS 213.8
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-9 (Degrees) Dens1ty Static D.R. Vs Peak v h v h
meter Varied Rockf111/Bedd1ng (pcf) F.S. av/ah a he % ( fps) Ah * ( ft) (ftl (ft) (ft)
Normal Case 48/46 135 1.89 0.374 0.239 15 800 0.567 0.3 0.6 0.6 1.1
g 45/44 135 1. 71 0.315 0.202 15 800 0.567 0.5 0.9 0.8 1.6
0 50/48 135 2.03 0.414 0.265 15 800 0.567 0.2 0.4 0.4 0.9
Density 48/46 125 1.89 0.374 0.234 15 800 0.567 0.3 0.6 0.5 1.2
Density 48/46 150 1.89 0.374 0.242 15 800 0.567 0.3 0.5 0.6 1.1
avlah 48/46 135 1.89 0 0.265 15 800 0.567 0.2 0.4 0.4 0.9
av/ah 48/46 135 1.89 2/3 0.215 15 800 0.567 0.4 0.8 0.7 1.4
D.R. 48/46 135 1.89 0.374 0.239 1 2 700 0.626 0.3 0.6 0.7 1.3
D.R. 48/46 135 1.89 0.374 0.239 20 700 0.573 0.2 0.4 0.5 0.8
Vs 48/46 135 1.89 0.374 0.239 15 700 0.597 0.3 0.5 0.6 1.0
Vs 48/46 135 1.89 0.374 0.239 15 900 o. 728 0.5 0.9 0.5 1.2
Vs 48/46 135 1.89 0.374 0.239 15 1000 0.897 0.8 1.5 0.8 1.6
Worst Case 45/44 125 1. 71 2/3 0.180 15 900 0. 728 0.8 1.6 1.0 2.0
Megathrust .55g 48/46 135 1.89 0.374 0.239 15 800 0.423 0. 1 0. 1 0.2 0.4
DBE .375g 48/46 135 1.89 0.374 0.239 15 800 0.288 0.0 0.0 0.0 0.0
.. Peak hor1zontal acceleration of the failure circle .
TABL~ page 4 of 6
Main Dam SARMA Results
CIRCLED, UPSTREAM, H.W. El. = 1180 ft., T.W. El. = 1061 ft.
CENTER POINT (24.5, 2056.5), RADIUS= 938.17
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-~ (Degrees) Density Static D.R. Vs Peak v h v h
meter Varied Rockfill/Bedding (pcf) F.S. avlah ahc % (fps) Ah" ( ft) ( ft) ( ft) (ft)
MCE 0.75g
Nonnal Case 48/46 135 3.47 0.515 0.504 15 800 2.48 1.2 3.2 1.3 3.5
~ 45/44 135 3.24 0.450 0.475 15 800 2.63 1.4 3.7 1.6 4. 1
~ 50/48 135 3. 72 0.560 0.539 15 800 2.63 1.1 3.0 1.3 3.3
Density 48/46 125 3.54 0.515 0.506 15 800 2.48 1.2 3.2 1.3 3.5
Density 48/46 150 3.38 0.515 0.487 15 800 2.63 1.4 3.6 1.5 3.9
avlah 48/46 135 3.47 0 0.620 15 800 2.63 0.9 2.3 1.0 2.6
av/ah 48/46 135 3.47 2/3 0.486 15 800 2.63 1.4 3.6 1.5 4.0
D.R. 48/46 135 3.47 0.515 0.504 12 700 2.33 1.2 3.2 1.6 4.2
D.R. 48/46 135 3.47 0.515 0.504 20 700 1.63 0.5 1.4 0.7 1.9
Vs 48/46 135 3.47 0.515 0.504 15 700 2.03 0.9 2.3 1.2 3.0
Vs 48/46 135 3.47 0.515 0.504 15 900 2.53 1.4 3.6 1.5 3.9
Vs 48/46 135 3.47 0.515 0.504 15 1000 2.24 1.0 2.8 1.4 3.6
Worst Case 45/44 125 3.30 2/3 0.455 15 900 2.65 1.7 4.5 1.9 4.9
Megathrust .55g 48/46 135 3.47 0.515 0.504 15 800 1.82 0.5 1.2 0.5 1.4
DBE .375g 48/46 135 3.47 0.515 0.504 15 800 1. 24 0.1 0.3 0. 1 0.3
,. Peak horizontal acceleration of the failure circle.
TABL, page 5 of 6
Main Dam SARMA Results
CIRCLE E. UPSTREAM, H.W. El. = 1190 ft., T.W. El. = 1061 ft.
CENTER POINT (71.5, 1609.33), RADIUS= 522.97
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-lit (Degrees) Density Static D.R. Vs Peak v h v h
meter Varied Rockfill/Bedding (pcf) F.S. a 11 /ah ahc % (fps) Ah" ( ft) ( ft) (ft) (ft)
.75g
Normal Case 48/46 135 1.95 0.335 0.299 15 800 1.90 3.2 5.7 3.4 6. 1
" 45/44 135 1.77 0.277 0.215 15 800 1. 93 4.0 7.0 4.2 7.4
" 50/48 135 2.09 0.374 0.330 15 800 1.90 2.8 4.9 3.0 5.3
Density 48/46 125 1. 95 0.335 0.299 15 800 1.90 3.2 5.7 3.4 6. 1
Density 48/46 150 1.95 0.335 0.299 15 800 1.90 3.2 5.7 3.4 6. 1
a 11 /ah 48/46 135 1. 95 0 0.335 15 800 1.90 2.7 4.8 2.9 5. 1
a 11 /ah 48/46 135 1. 95 2/3 0.270 15 800 1.90 3.7 6.6 3.9 6.9
O.R. 48/46 135 1. 95 0.335 0.299 12 700 1. 71 2.9 5.2 3.8 6.8
O.R. 48/46 135 1.95 0.335 0.299 20 700 1. 25 1.5 2.7 2.0 3.4
Vs 48/46 135 1.95 0.335 0.299 15 700 1. 52 2.2 3.9 2.9 5.1
vs 48/46 135 1. 95 0.335 0.299 15 900 2.01 3.6 6.4 3.8 6.7
Vs 48/46 135 1.95 0.335 0.299 15 1000 1.82 3.0 5.2 3.6 6.3
Worst Case 45/44 125 1.77 2/3 0.235 15 900 2.01 4.8 8.4 5.0 8.6
Megathrust .55g 48/46 135 1.95 0.335 0.299 15 800 1.40 1.4 2.5 1.6 2.7
DBE .375g 48/46 135 1.95 0.335 0.299 15 800 0.95 0.4 0.8 0.6 0.9
"' Peak horizontal acceleration of the failure circle.
TABLE page 6 of 6
Main Dam SARMA Results
CIRCLE F, UPSTREAM, H.W. El. 1190 ft., T.W. El. = 1061 ft.
CENTER POINT (71.5, 1609.33), RADIUS= 550.07
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Case/Para-1.1 (Degrees) Density Static D.R. Vs Peak v h v h
meter Varied Rockfill/Bedding (pcf) F.S. av/ah ahc % (fps) Ah* (ft) ( ft) (ft) (ft)
MCE 0.75g
Normal Case 48/46 135 2.48 0.445 0.396 15 800 2.63 0.8 1.9 1.0 2. 1
0 45/44 135 2.24 0.384 0.359 15 800 1.20 1.0 2.3 1.1 2.5
" 50/48 135 2.67 0.488 0.425 15 800 1. 20 0.8 1.6 0.8 1.8
Density 48/46 125 2.50 ' 0.445 0.404 15 800 1.35 0.9 2. 1 1.1 2.4
Density 48/46 150 2.47 0.445 0.390 15 800 1.20 0.9 2.0 1.0 2.2
av/ah 48/46 135 2.48 0 0.479 15 800 1. 20 0.6 1.2 0.7 1.4
av/ah 48/46 135 2.48 2/3 0.370 15 800 1. 20 1.0 2.2 1.1 2.4
O.R. 48/46 135 2.48 0.445 0.396 12 700 1. 71 0.5 1.2 0.8 1.7
O.R. 48/46 135 2.48 0.445 0.396 20 700 1. 25 0.3 0.6 0.4 1.0
Vs 48/46 135 2.48 0.445 0.396 15 700 2.23 0.4 0.9 0.7 1.4
vs 48/46 135 2.48 0.445 0.396 15 900 2.65 1.1 2.4 1.2 2.7
Vs 48/46 135 2.48 0.445 0.396 15 1000 1. 29 0.9 2.0 1.1 2.5
Worst Case 45/44 125 2.25 2/3 0.330 15 900 1. 37 1.4 3.2 1.6 3.6
Hegathrust .55g 48/46 135 2.48 0.445 0.396 15 800 0.88 0.3 0.7 0.3 0.8
DBE ,375g 48/46 135 2.48 0.445 0.396 15 800 0.60 0.0 0. 1 0. 1 0. 1
.. Peak horizontal acceleration of the failure circle.
TABU:. -page 1 of 4
SARMA Results-Old Geometry
CIRCLED, UPSTREAM, H.W. El. = 1180 ., T.W. El. =
CENTER POINT (640.33, 1900.0), RADIUS 770.87
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Varied 0 (Degrees) Density Static D.R. Peak v h v h
v5 (fps) Rockfill/Bedding (pcf) F.S. av/ah ahc % A * h (ft) (ft) ( ft) ( ft)
700 48 1 35 4.4 0.52 0.585 15 2.29 0.90 1.95 l. 32 2.43
800 48 135 4.4 0.52 0.585 15 2.57 1.24 2.65 1.36 2.91
950 48 135 4.4 0.52 0.585 15 2.22 0.92 1.96 l. 25 2.67
1100 48 135 4.4 0.52 0.585 15 1.83 0.67 1.45 l. 01 2.18
1250 48 135 4.4 0.52 0.585 15 1.84 0.43 0.93 o. 70 1.49
1350 48 135 4.4 0.52 0.585 15 1.95 0.42 0.90 0.60 1. 28
1500 48 135 4.4 0.52 0.585 15 2.09 0.40 0.85 0.52 1.11
"' Peak horizontal acceleration of the failure circle.
TABLt. page 2 of 4
SARMA Results-Old Geometry
CIRCLE A, DOWNSTREAM, H.W. El. = 1190 ft., T.W. El.
CENTER POINT (497.8, 1386.86), RADIUS 246.17
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Varied 0 (Degrees) Density Stat'ic D.R. Peak v h v h v5 (fps) Rockfi 11/Bedding (pcf) F.S. ay/ah a he % Ah,. ( ft l (ft) (ft) (ft)
700 48 135 2.2 0.38 0.345 15 2.03 2. 10 4.49 2.45 5.26
800 48 135 2.2 0.38 0.345 15 2.32 2.91 6.24 2.93 6.30
850 48 135 2.2 0.38 0.345 15 2.30 2.82 6.07 2.99 6.41
950 48 135 2.2 0.38 0.345 15 2.04 2.24 4.80 2.76 5.91
1100 48 135 2.2 0.38 0.345 15 1.68 1. 74 3.75 2.32 4.97
1250 48 135 2.2 0.38 0.345 15 I. 71 I. 37 2.94 1. 81 3.88
1350 48 135 2.2 0.38 0.345 15 1.84 1.29 2.76 1.60 3.43
1500 48 135 2.2 0.38 0.345 15 1.99 1. 15 2.45 1. 41 3.03
" Peak horizontal acceleration of the failure circle.
TABLt. . page 3 of 4
SARMA Results-Old Geometry
CIRCLE G, DOWNSTREAM, H.W. El. = 1190 ft., T.W. El.
CENTER POINT (557.8, 1353.53), RADIUS 204.43
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
HYBRID TAFT
Varied 0 !Degrees) Density Static O.R. Peak v h v h
Vs (fps) Rockfill/Bedding (pcf) f.S. avlah a he % A " h (ft) (ft) (ft) ( ft)
700 48 1 35 2.04 0.31 0.348 15 1 .034 0.99 1. 82 1. 21 2.23
800 48 135 2.04 0.31 0.348 15 1. 33 1.63 3.00 1 76 3.24
850 48 135 2.04 0.31 0.348 15 1. 38 1.67 3.08 1.88 3.47
950 48 135 2.04 0.31 0.348 15 1. 28 1.41 2.60 1. 76 3.24
1100 48 1 35 2.04 0. 31 0.348 15 1. 10 1. 14 2. 10 1.57 2.89
1250 48 135 2.04 0.31 0.348 15 1. 10 0.86 1.59 1.24 2.29
1350 48 135 2.04 0.31 0.348 15 1.24 0.83 1.53 1. 13 2.07
1500 48 135 2.04 0.31 0.348 15 1.44 0.79 1.45 1.04 1.90
" Peak horizontal acceleration of the failure circle.
Varied
v5 (fps)
800
TABU:. w page 4 of 4
SARMA Results-Old Geometry
CIRCLED, UPSTREAM, H.W. El. = 1190 ft., T.W. El.
CENTER POINT (625.6, 1523.73), RADIUS= 432.82
INPUT PARAMETERS/INTERMEDIATE RESULTS DISPLACEMENT
II {Degrees)
Rockf111/Bedding
48
Density
{pcf)
135
Static
F.S.
1.98
av/ah
0.32
HYBRID TAFT
D.R. Peak v h v h
ahc % Ahw (ft) (ft) (ft) (ft)
0. 305 15 1.75 2.84 5.34 2.90 5.45
* Peak horizontal acceleration of the failure circle.
g
"' ~
2.115.000
~4/ V/u
1
KACHEMAK BAY ~s
I ~'"
~~~ y.O -.,o
d .... l ...... .
.... -
2.110
;·
2.100,000
-)
c
MUD
FLAT
'!.1
·~-1
'-1'"
A~
/<:>oO I 1 1 1 r=.........._ C>~ l 1 r--""" <@f? :::;::s:) T 1 , n 1 n ''' 1 )
i11~'\(" ~'
EXISTING SITE PREP
CONTRACTOR CAMP
~
NOTES I
I• IIILl FEA~ ltCMIII ME. EllSTU. £JCI:f'TI
A. MLN5 DEil..,.TED M -slTE"
8. I'IIIUN IWt
C. SI'ILLIMY
D. ..a.EI T\.NEL urr Met
0 1000 2000
(\ ~ I
SCALE • 1"•1000'
PROJECT LOCATION MAP
FIGURE 1
I
ACCESS OOAO-j
I
/
/~ASTE DISPOSAL
AREA B
• lL 1\00.0'MAX
~~~
'0 --~ 't 'b
(
\
\
WASTE
DISPOSAL AREA ~7 EL 1090,0' MA':f
\ ', "-, ', .......... , ', \ ', ~
\ "' \ 0
\ ( I
t \ r, 1 l \J ..........
\
\
\
\ I ' /
, __
• I
MAXIMUM NORMAL I 1
OPERATiNG WATER • C3~1 1
SURF!>CE EL\1800' I '
' I \ I I
1
WASTE DISPOSALl--ll ~ AREA F J \
rt EL 10900' \M~AA ~ ,: /
\ ~ / I I I
I f
I I
I \ I \
t \ \ \
\ \ \
\ I I
J I I -\
\
\
' \
\
\
' ............... ~ ........... ~ ........ ,
'.._, ', ',, ',:,;, ',,
', ', ',, ', ', ',',,o ', ',~,, ',,>·, ', ', <>o,
' ......... , ',' 'n"'' ......... ..... .... ' .... ',_'1:> ..... .... ..... -
tiJ SYMBOLS KEY
E) SURVEY MONUMENT
• SEISMOGRAPH INSTALLATION
G) WCRK POINT ..... ' .... ' .........
' '-b~ ....... ' .... , ......... ,
..... ~ ..... ' .... .... ........ , ,...., <~-G~~~E~~c::.
~ ..... ::----........ .:: ....... , ... __ ...... __
' .... .... --........ , ... , ........
............. : ......... : ......... .....
....... ...... , .... ,,
', ', ',
................ , ' .........
.......... ...... .....
1:>, 0~~
STAGtNG ~ EL 11 ~· 0 !'1'1£A ..,c?. ~.7 "'..,•"'
' ' ... ,,,\
\
\
\ /' \,.. .-' ....... __ _
GENERAL ARRANGEMENT
MAIN DAM AREA
FIGURE 2
3' PROCES!>ED CRUSHED STONE
39MAX ROCK FILL· GAP GRADED DRAIN
36°MAX ROCK FILL
36'MA)( ROCK FILL
B!> I 48"MAX ROCK F"LL (OIIERSIZE)
B1l I GRAIIEL ROAD SUIFACING
ZONE
TYPE 91
CONC FACE SLAB
1'·0' THICK
GROJT CURTAIN
ZONE 2
TYPE 62 FILL
El. 119_4'_1..__
ZONE J
TYPE 93 FILL
ZONE 4
TYPE 94 FILL
II
MAXIMUM DAM PROFILE ...
ICAl.t•nn
581f·6·
AO.RAPET EXP JT IT'IP) PARAPET WALL !10'·0• ~~~ 11
I
\
" " " " -
SEGMENT
A
TRANSfTKlN .DINT
W/WATERSTOP
3-3 .. ~
IC.U.I'. "IT
.......... -._ .......__ ---
REINFORCED
/'CONCRETE
.t' FI>C£ SLABS
,.J
21
2
SEGMENT D
VIEW LOOKING DOWNSTREAM---------
~
~
b ,!
. -· .. !!'liiil"'"<ii I
&4UllfUU
DISTANCE A
L, 6\.7•
T Jl.()'
Lz 4'·10f
SEGMENT
B c 0
.4•-7f 8'-1-f' e•.oo
2•~-4t• 2t.)• 2'-3'
!>'·!Of !>'·3r ~-3f
I.
-
ZONE 2
TYPE 62 FILL
RIGHT AEilJTMENT PLINTH
I
I
J
'i:"'-/--GROut
I CURTAIN
/
/
-.. ·-. ·~ . ~ '. :\
'· "'" .I. . .. ::!·'j
2-2
, r •' I"Siie; I
"' ~
b
tOi
FLOW-
MEMBRANE LINER
FILTER MINUS 3' MARTIN
RIIIER BORROW MATERIAL
UPSTREAM COFFERDAM PROFILE
o• 10• '2d w I
SCAL[ lflj Jl'f;(T
10'·0'
PARAPET WALL
SL1190t .· ...... ·F-'·0::·~ .· · ..... ~-<~~-
<!'·91
1 -1 '. ~ ~ e---Q I
.C&Lt: * rUI'
MAIN DAM SECTIONS
FIGURE 3
2.25 ...-....
01 ..._,
ns
(/) 1.88
z
0
~1.50
0:: w
_j
w u 1.13
~
_J
<{ 0.75
0::
1-w
0.. 0.38
(j)
0.00
RESPONSE SPECTRUM
FOR HYBRID EARTHQUAKE
BRADLEY LAKE HYDROELECTRIC PROJECT
MEAN RESPONSE SPECTRUM FOR M C E
(NEARBY SHALLOW CRUSTAL FAULT)
REF: WOODWARD-CLYDE CONSULT
REPORT1 "DESIGN EARlliQUAKE STUDY'
NOV 10,1981
--
SPECTRUM
--~-
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00
PERIOD (SEC)
MCE RESPONSE SPECTRA...,
MEAN AND CHOSEN
~------------------------FIGURE 5
-
LOW DENSITY,
POORLY GRADED#
-+----+ WEAK ROUNDED
PARTICLES
ANTICIPATED RANGE-
BRADLEY LAKE MAIN DAM
HIGH DENSITY 1 WELL
GRADED I STRONG
ANGULAR PARTICLES
ROCK FILL
304---~----+---+---~----r---+-----+---~
1 2 5 10 20 50 100 200 500
NORMAL PRESSURE
(PSI)
FROM LEPS ,1970
ROCKFILL FRICTION ANGLES
~-----------------------------FIGURE 6
ELEMENT OF
DAM FACE A RESULTANT OF FORCES
ACTING ON THE ELEMENT
WHICH FALLS ON ONE OF THE
11 CRITICAL LINES" GIVES A
FACTOR OF SAFTY OF 1.0
AGAINST SLIDING. IF RESULTANT
IS OUTSIDE THE CRITICAL LINEJ
THEN ELEMENT IS UNSTABLE.
THE PRIMARY FORCE IS THE
ELEMENT WEIGHT (w).
THE MIN SEISMIC FORCE (amm)
NEEDED TO REACH THE CRITICAL
LINE WILL BE PERPENDICULAR
TO THE CRITICAL LINE.
THE RATIO BETWEEN av & aH
WHICH GIVES THAT MINIMUM
SEISMIC FORCE IS
0YciH = t~n (;-e)
FOR MOST VALUES OF ¢ & e J
THIS GIVES A RATIO BETWEEN
0 AND~.
IN THE CASE OF CIRCULAR
ARCS RATHER THAN PLANES:
THE SLOPE ANGLE FROM
pt. A TO pt. B WAS USED TO
APPROXIMATE e.
INTERMEDIATE 0 %H RATIO
----------------------------------FIGURE 7----~
CIRCLE SLIP SURFACES
USED IN THE DETAILED ANALYSIS
1250
1200
1150
1100
1050
NORMAL MAX
OPER LEVEL
EL 1180.0'
\7
(PMF)
EL 1190.0'
\7
~DAM SHELL~
CIRCLE X CENTER Y CENTER
A
B
c
D
E
F
---..::::.::=:; s >.. B
" <
POINT POINT
596.0 1450.0
596.0 1450.0
696.67 1249.33
24.50 2056.50
71.50 1609.33
71 .50 1609.33
MAX
TAILWATER
(PMF)
EL 1077.01
'\7
-----
1000;-------~----~------~------,-------~-----.------~------~------.-------.------.------~
100 150 200 250 300 350 400 450 500 550 600 650 700
SELECTED SLIDING SURFACES -MAIN DAM
~----------------------------------------------------------------------------------FIGURE 8
.
RADIUS
347.13
366.51
213.80
938.17
522.97
550.07
NORMAL
TAILWATER
EL 1061.o'
\7
0.6 ___.,~-or---.--.
D
0.5 I I :J.tC I I
u
I: 0.4 I I I ~ :31 I
0
0.31 I ~
Q2 ' II"" I t I
~· 50•
44• 49 6
¢-ROCKFILL I BEDDING
0.6
0.5 D
u
J0.4 -F
0.3
0.2
1 20
-
I
140
B
A
E-
c
1tJ 0
lROCKFILL ~ MOIST
( PCF)
~ c
0.7 ~-----,----,
0.6 --Pr-------1---~
Q5 t ~D I
04 I -.. """"" I I • -.:c:: -..::::
F
B
0.31~ A I
E
c 0.2 I I I
0 0.5 1.0
Oy/OH
?_) s
CIRCLE LOCATIONS
SHOWN ON FIGURE 8
CRITICAL ACCELERATION PLOTS
----------------------------------------------------FIGURE 9 ----
4
~
~ w u 3 <!
...J n.
Vl
0 -~~2 --~
<!
~
....J
:::> 1 u _,
<! u
i= a::: 0 w. w 45°
0 > 44•
.¢'-ROCKFILL I BEDDING
4
3
2
1
0
120
-
E
A
~ D __..
8
F
c
140 160
ROCKFILL ~ MOIST
( PCF)
HYBRID RECORD
PERMANENT DEFORMATION PLOTS
4------~------~
E
A
3 1 " ...,. I ,.-:ir
2 -1------+--------1
g
1 (;;,,............-__............. l::::;;....o-F I
c
0 -+------t-----i
0 0.5 1.0
av/aH
CIRCLE LOCATIONS
SHOWN ON FIGURE 8
L------------------------------FIGURE 10 _ ___.
SHEET 1 OF 4
ffi
~ w u
~
(/)
B
w
>-~~
~
L: :::> u
_J
<{ u
1--
0:: w
>
4 __ ___,.. __ ...,......,
J I ., " I
E
2 'A
1 'D
0 I I I I I I
600 1000
SHEARWAVE VELOCITY
( FPS)
4~-r-----t-___,..--,
3,E
2 I I ,._ t. I I
1 I I ..,._ I I I
c
' I ~ I I I 0 I 20 12
DAMPING RATIO
( Ofo)
HYBRID RECORD
.55 g MEGATHRUST
4------~~~~--
I
g~
"a M
3 j I I
2 I I I II I I
1 I I Ill lA-I
0 I I 16<)' 11 I
0
OH PEAK
(0/o OF g)
PERMANENT DEFORMATION PLOTS
.___---------------------------FIGURE 10 ---4
SHEET 2 OF 4
~ 4
}:
w u ·~ ... 31 EC 3~ I --
t/)
0
w >-~t j ._ 2 T1T 1 -------'1 ~ I I
}:
}: a
..J
<{ u -...._
1 I I...,__ 1-1
0: w > Ql I I I I
45°
44°
50°
48°
.¢-ROCKFIL L I BEDDING
4
E
3 A-
2
~ D -----8
1
...._ F -
.....,. c
0
I' {') 1 ..1("\ H 0
ROCKFILL ~ MOIST
(PCF)
TAFT RECORD
PERMANENT DEFORMATION PLOTS
4 E
A
3 I I -I I .-1-
2~----~~----~
1 I~ ==-~· I
c
o~---+-------1
1.0
Oy/OH
~-------------------------------------------------------FIGURE 10 --~
SHEET 3 OF 4
..... z
~ w u
<(
...J a..
Vl -0
~-
4 -r-----r----r_,
E
3 ~ I I I I
Eit: 21 I II
...J-
::>
~ :::>
U 1~ lA I
_j
<( u
.....
a:: w > 0 t I I I I I
600 1000
SHEARWAVE VELOCITY
( FPS)
4 -r-r----,----r-,
E
3 ~ 1\ '.r I I I
2 I I I ~ ~ I
1 I -1"-""= I I
0 I I I I I I I
12 20
DAMPING RATIO
( o/o )
TAFT RECORD
.75g-MCE
4 w ~ tn
I :::> ~ rn~ JE
,......_ l!H-
3 I ~ Ll1 ~ JJA
UJ
~
2. I Ill I I
1---1 I II I /II
0 t I ...,...-l 1 11 I
0 50 100
aH PEAK
(0 /o OF g)
PERMANENT DEFORMATION PLOTS
~-----------------------------------------------------------------------FIGURE 10----~
SHEET 4 OF 4
1-0 .... w.
WN
lL.
V,)o ~--~ z-w :c w u cz:o
....I~
a._o
V,) ,_..
0
o.oo s.oo
TIME-HISTORY
CIRCLE B,Vs=
.oo 25.00
OF· WEDGE MOTIONS
800 FPS "NORMAL ..
Earthquake accelerogram and
acceleration plotted
Wedge acceleration is of larger
60.00
magnitude and is more prominent U
_of .tbe._tv.cLcu.:r:ves~ -------------------------~
--INDICIITCi [11-I"QU~kfl "YIIIIIO llttCI..f~OOAM 1 kt•N to. 111ft IUtfl ~ flliULI HIILl '~N •otco lfWI
----INDICIUU Clllflti!L IICCfLCilllrlON. Q.JU D'l
--INOitiiTU llfODC oarLACC"fH" ~ IICt[LUIITIOMI.
0
II)
0
-I
0 ....
N
0
(j)
0
II)
0
0 0
0 0 . .
s.oo 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 ss.oo 60.00 °
TIME -SECONDS
MCE RESPONSE I DISPLACEMENT PLOTS
~------------------------------------------------------FIGURE11
SHEET 1 OF 6
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE B~Vs= 11 000 FPS
o.oo 5.00 .oo 55.00 60.00
0
U)~
~0
1-0
"'# w.
W"'
LL
U)o
I-"?
z-w
L:
w
u a:o
_J~
(LO
U)
~
0
u,
--------------------------------------------------------------------------
Jf\AAJ.It~~--------
--------------------------------------------------------------------------U,
--J"OICAIU CAAIHIIUAKCI HUAID ACCCLCAOOIIAn 1 KCA" CD. TArT IIIICI ~ rAJULI JIALY 'A" AOCCD 1(~1 DATA AAC CDCHICIC"U or D--f7S7 rDI"" •I • D.D%0 HCD"D'
----I"DICAIU CAlli CAL ACCCLCAAIJDM. o.,S D''
I"DICAIC' ~COD( DllrLACCn£"" ~ ACCCLCUIJD"'·
0
CD
0
0
0
0
0
CD
0
I
0
10 -I
0
"'#
N
0
10
0
CD
0
0 0
0 0 . .
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 ss.oo 60.00 °
TIME -SECONDS
MCE RESPONSE/ DISPLACEMENT PLOTS
FIGURE 11-----~
SHEET 2 OF 6
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE B, Vs = 900 FPS
o.oo s.oo .oo
0
(f.)~
._o
I.U~
LLJI"l
u..
(f.)o .... ~
ZN
l.LJ
I:
I.U u a:o
...Jo a....:
(f.) ......
Cl
~
-------~----------------------------------------------------~--------------
0
CD
0
0
0 WJU~~~~~~~~.~~w~~~~~~~~~~~~~~----------+~
----------~---·----------------------------------------------------------------------~ I~
--IIIOICIITU CIUITIIQUM[I HYII.IIIO AttC~UOOAAII 1 KUII tO. tArT UIUI ~ fltiULI ITA~Y IIIII ltOCCO I [Ill
---IIIOICIIIU CltltiCA~ IICCCLCIUHIOII. O.SU 0'1
DATA AU tOC,ICICHU Of 0--ll5l rOINU •T • Q.OtO UCOHOI
IIIOICIITU 11[00( Ollrl~t!ft[llfl ~ IICCllCUTIONI •
~~F=~~~======~========~
0
ID -I
0
0
.....
0
0
N
0
0
~I I I I I I I I 1 I I I I I I~ u.oo s.oo to.oo ts.oo zo.oo 25.oo 3o.oo 35.oo ~o.oo 45.oo 5o.oo 55.oo Go.oo
TIME -SECONDS
MCE RESPONSE/ DISPLACEMENT PLOTS
L-------------------------------------------------------FIGURE11-----
SHEET 3 OF 6
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE B J V5 = 700 FPS
o.oo s.oo
0
V,)IJ)
cc;;
0 1-N w· w-
u...
v:lo ~--~ zo w :c w u a:o
...J~
Q..o
V) -0
0
0
------------------------------------------------------------------------------------~
0
IJ)
0
0
0
~MI~~·Iwl-'\fl\o~.MJI'I'\'lAA.M.IMjW""""~-----+c;;
0 -------------------------------------------------------------------------------------:u--11J)
<
--IIIDICAJU (IUIJ"OUM(I lllUID ACtCLfltOOI!M 1 lUll CO. Iliff IIUU l UIULI ITALY ,1111 aGCCO Ifill DlliR Rill: COUflCJ[IIU Of 0--flU rOIIIU •J: Q.OZO UCOHOI
----INOICAIU CltlriCAL ACCILUAJIOII, o·.nl 0'1
IIIOICIIIU 11(00( DIII'LACinCIIII 4 ACC(L(UJIOIIf,
0
1:'1
0
<I)
0
0 ....
0
0
0
•I II I j ' I 9J.oo s.oo to.oo t5.oo 2o.oo 2s.oo 3o.oo 3s.oo ~------~r--------r--------,---------r--~ . 4o.oo -4S.oo sb.oo ss.oo sb.oo o
TIME -SECONDS
MCE RESPONSE/ Dl SPLAC EM ENT PLOTS
~--------------------------------------------------------FIGURE11----~
SHEET 4 OF 6
TIME-HISTORY OFWEDGE MOTIONS
Cl RCL E D J Vs = 800 FPS ~" NORMAL"
0 .QQ 5.00 10.00 IS .00 20.00 ZS .00 30.00 35.00 ~0 .Otr ~5 .00 50.00 55.00 60.00
~-------L------~~------1
0
(.1')0
d~
VJo
0
0
N
zo o~~Jamt~.' !Jl84Jv-\l.ftl.)l\1(Uillli.lm-~, irh-.-.;'~•I.~-~~T.:::~~------------------------------------~ 1 g
1 ur'Y~ _r~r:~r~-~-v: --------~ ·
>---
a:
et: wo ..J~
WN
u' u a:
---------------------------0 --------------------------------------:u--
'
--llfOICIII(I UA!!<IIIJUC1 "lUIO •ttCLUOO«M 1 HU co. T~fl IUUI 'riiULI IIAlf ~<• JOC<O 4(wl o•U Uf CDCfriCIUU or 0--t?n IOI•U •I • O.QtO Hto•DI
0
0
N
I
----llfDICA!£1 tJ<IIICAi. ~CCC\.C.AIIO•, 0.101 0'1 g I --lifO I caru wcooc DllfLacc•nu ' ~cccLC•AT 10"'. 0
0
-.:
I
,___g
uJ •
tu"" u...
(/)0
--~ ZN
w
r
w
a...:
(/) _,
0
s.oo 10.00
MCE
..
I
0
0
1'1
0
0
N
0
0
0
0 ,.----,--------·--.-------.-----.---.----.--------1. .
\5.00 20.00 25.00 30.00 35.00 40.00 45.00 so.oo ss.oo 60.00 °
T I tiE -SECONDS
RESPONSE/DISPLACEMENT PLOTS
Fl GURE 11------~
SHEET 50F 6
o.oo
0
(/)0 . .
(!)-
s.oo 10.00
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE E , Vs = 800 FPS >"NORMAL"
ts.oo zo.oo zs.oo '3o.oo 3S.oo •o.oo-•s.oo so.oo ss.oo 1 1 1 ___________ L __ ____L L.. _ _____.__._1_ ___ _._.
60.00
Cl
0
(f)o zo
oc;;~•
u ~~----------------------------------------------~ ~~
......
t-
cr:
0:::0
Wo
_J •
-------------------------------------------------------:u--
'
w-
u' u cr:
0
0
~---I~Oitl'UI fAA!l10V~M[I •UUO ~ttClUOGUII I MU• tO. T~ff llUCI ~ f~IUll ll~ll ~~~OttO I[Wl
---I•Oitl!IU CAI!ltftl AttCl[~~IID•, a.Ut 0'1
.,_o
0
W• w10
lL.
U)o ~--~
z~'"~ w
I:
w
0
--IMDitl!IU WCO!K OUrLAtCoC•II ~ /ltC(lUA!I O•l,
OA!A ~A[ COCfrltlC•U Of ~--nn tOI•U •I • O.Oto UtO•GI
Cl
0 -I
0
0
!:'ol
I
0
0
ID
.::;,
0
"'
0 0
0 0 . .
~ "' 10.oo s.oo to.oo ts.oo zo.oo zs.oo 3o.oo 3s.oo 4u.oo 4S.oo so.oo ss.oo so.oo 1
TIME -SECONDS
MCE RESPONSE/DISPLACEMENT PLOTS
L--------------------------------------------------------FIGURE11
SHEET 6 OF 6
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE 8, aH = 0.55G
o.oo 50.00 55.00 so.oo
0
U)IJ)
de;;
1-0 wO? wo
LL.
U)o ~--~ zo
w
:1:: w u a:o
_jeri lLc;;
<n ......
0
~
0
Ill
------------------~------------------------------~------------------------------------------
0
0 IIIWINIIJIIMit'WIMR'\AI\IVW'\Jidfo\liMitn~ ... l\j.\6;/~J~Aw-e-· ------1-c:;;
--INDICJIJ[f fiiiJ"IIUIIII[t 11lUID IICCfU:It08111111 t Kfllll CO. JAn IIIIKI 4 fRIULI ITIILY filii lottO 1[111
----INDitiiTfS tltiJ I tilL IICCfL[IIIIUDII. 0 ,,IJ O'S
--IIIOICIITU 11[118( PIII'LIItfllfiiTI 4 IICtfLfiUITJDIIIJ.
0
-r-lln
c
OATil Ill[ COf,ltlfiiTI Of 0--Ulrl I'OIIITI •T • O.OU UtONOS
0
(I)
0
0
II) .
0
0
crl
0
0 0
0 0 . .
5.00 10.00 I .00 20.00 2 .00 30.00 3 .00 40.00 4 .00 50.00 55.00 60.00 °
TIME SECONDS
MEGATHRUST RESPONSE/ DISPLACEMENT PLOTS
L-------------------------------------------------------FIGURE12----~
SHEET 1 OF 3
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE D 1 OH = 0.55 G
o.oo 5.00 lO.OO 15.00 20.00 25.00 30.00 35.00
0
U)O
0 1-c.~ w. w-
ll...
wo ~--~ zo
w r
w
u a:o _, ....
Q._~
U) .....
0
0
0
u
------------------------------------- ------------~
0
0
0
0
0
----------------------------------::ij I o --t 0
JIIOICIITU fiiiiT"IlWIIIft "YIIIIID ACCflfiOIIIIAII : !!fill CO. TArT 1688fl ~ raJULI ITAU 6AII IOCCO Ifill
----IIIDICIITU CIITICfll flttflf:IIHJDII. o.S04 0'6
--IIIOICfiTfi llfDOf Ol6rlfltfllf11Ti ~ ACULUfiHDIIi·
DATA AI[ CO[frltlfiHi Or 0--t757 rOIHTll •T : o.ao I[COH06
-I
0
0
0
N
0
liD
0
0 ....
0
+---~--~------~------,-------.-~.00 5.00 10.00
.-------------y
25.00 30.00 35.00 40.00 45.00 55.00 50.00 LS.OO 20.00
TIME -SECONDS
MEGATHRUST RESPONSE/ DISPLACEMENT PLOTS
FIGURE 12----'
SH T 2 OF 3
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE
o.oo 5.00
0
Ul"!
1-0 ..
W•
WN
LJ...
U)o ~--~
z-w
:E w
u a:o
_JCIO
a...~
U) .......
D
------------------------~
0
CD
0
0
0 lll.l!tiMVII~'Wli!Mh\W~IINWPA~6ll~J\.Aw~ ... ·~-I~
-------------------------------------------------------------------------=u--
c
--INDICIITfl fll~THUUilKfl HTIIIID llttfLfiiDOIIIln 1 KfiiN to. TllP'T llllfl ' P'~IULI ITIILT IRII ~DCCD IfNI
----INDICilTfl ~ITICIIL llCC[L[~IlTIDN. D.UI D"l
--INDICIITfl WfDDf DllrLiltfllfiiTI 'llttfLUilTIDIII.
(
rP
Dlllll llllf COfP'P'ICI[IITI OP' D--!757 roiNTI •T 2 o.oto lfCDNDI
0
CD
0
ID -I
~ ~0 ---..
N
0
ID
0
CD
0
0 0
0 0 . .
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 °
TIME -SECONDS
MEGATHRUST RESPONSE I DISPLACEMENT PLOTS
FIGURE 12-----~
SHEET 3 OF 3
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE = 0.375G
o.oo s.oo .oo
0
U) ....
de;;
I-liD
w-:' wo
l.i..
U)N
1--:' zo
w
l::
w
u
a:"" _Jc;
CLo
U) .......
0
0
--------------------------------------------------------~ I.,..
------------------------------------------:u--
c
IIIOICIITU (llllfiiQUAII[o IIYUID IICCfLlllOOIIIIII o l!.fllll CO. ffl(f 1&18[1 <f. riiiULI ITIILT SAil I!DttO lUI
----IIIOICIITU tltiUCfll IICC:flfi!ATIOII. O.US D'S
IIIOICIITU MfOOf Ol&tlfiUIIfNT6 <f. RCUU~RTIOII&.
J
ORTR lilt[ tKrntlfllf& Of 0 •• t751 tOIIITS at : D.otO &ftOIIO&
0
0
0
0 ....
liD -
0
N -
0
(I)
0
0
0 0
0 0 . .
s.oo 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 so.oo 55.00 60.00 °
TIME -SECONDS
DBE RESPONSE /DISPLACEMENT PLOTS
~--------------------------------------------------------FIGURE13----~
SHEET 1 OF 3
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE D ,aH = 0.375 G
o.oo
0
U)ll'l
d~
~--· w~ wa
lL.
U)<O
t-':
:z:a
w ::c w
(J
a:"' _.~o
a...~
(/) ......
0
u, --------------------~--------------------------------·· -------------~----.......,_
0
"' 0
a
0
a
------r 1 ~
'
--INOICIITfi flll!THIIOMft HlUIO IICCfl[IIOO«A" 1 IIUN Co. TIII'T 1618[1 ~ Fllllll.l JTIIll &liN UCCO Ifill
----lNOICIITfi CIIJT I CAl IICCflfiiiiTIOH. 0.504 0'1
INOICIITU II£0Gf: 0161'lACf"fNrli ~ IICCflfiiiiTION6.
UIITA flftf C(l[f"f"ICifHf' Of 0--t757 I'OINT6 •f = O.O'lO UCOII06
....
cv
a
<0 -
0
(I)
a
a
a a o a . .
s.oo to.oo 15.00 20.00 2 .oo 30.00 35.00 40.00 45.00 50.00 55.00 60.00 a
TIME -SECONDS
DBE RESPONSE/DISPLACEMENT PLOTS
~-------------------------------------------------------FIGURE13----~
SHEET 2 OF 3
TIME HISTORY OF WEDGE MOTIONS
CIRCLE E 0.375 G o.oo s.oo .oo 45.00 .oo 60.00
0
([JII)
~~
t-:i
LIJ •
LIJO
lL.
(J)o
t-~ z:o
LIJ
I::
LIJ u a:o
_J(\')
a...~
(fJ ._
0
-----------------------------------------------------------------~
--L·-----------------------------------------------------------~
--IMOICIIfU fllltfH8lllllf1 "YUIO llttfl.fiiOGIIIIft 1 11011 CO. flirT 1&18[1 ~ rlUUI.I UIII.T &liM liOCCO Ifill 011111 ur c«P'P'Itlfiiiii or G·-ns, reuna .r = o.oto arcoHG&
--INDICIIff& ClllfiCIIl lltCfLfliiiUOM. o.raa 0'1
--IMDJCIIJfl llfOIN' DJfjrLIICfftfNfl ~ IICC[L[~IITIOMti.
0
II)
0
0
0
0
0'1
0
0
CD
0
0
(\')
0
0 0
0 0 . .
s.oo to.oo t5.oo 2o.oo 2s.oo 3o.oo 35.oo 40.oo 4S.oo 5o.oo 55.oo so.oo o
TIME -SECONDS
DBE RESPONSE/DISPLACEMENT PLOTS
Fl GURE 13 __ __.
SHEET 3 OF 3
EL 1180
CONDITIONS : Kv = 1b K H
K BEDDING = Tb K ROCKFlLL
BEDDING LAYER = 12FT HORIZONTAL
DIS BERM IGNORED
FLOW THROUGH DAM WITHOUT FACE
MAX
BEL 1066
\!
~--------------------------------------------------FIGURE14
I-
I
<.9 -....-...
20
15
~~ 10
~
L:
~
5
)~--~--~--~~--~--~---
) -+--+--+--t
I -t-----t.rfiL----f-----t-
1~---+----~--+---~---+--~
0 0.2 0.4 0.6 0.8 1.0 1.2 1.4
VERTICAL DISPLACEMENT
( FT MCE CASE)
1-
I
<.9 -..........
WI-
IIJ...
....__.
L:
<(
0
200~------~-----.-------r------.
150 ---+-----·----+···-~-----·---+--···--------1 •
100 ~-+------------·-----+-·----·· -------~-
•
50
•
04-------~-----+-------r----~
1.0 1.5 2.0 2,5
MAX WEDGE ACCELERATION
(G)
3.0
DAM HEIGHT VS. ACCELERATION AND DISPLACEMENT
~--------------------------------------------------FIGURE15 ---~
FOR PLANAR SLIDING SURFACES
ILLUSTRATED BELOW I 1200 End Pt. F.S. <XV
\ledge I (Left} Slope 8 Static o( H
366,1190 30.05° 1.92 .324
2 350,1180 26.74° 2.20 .389
334,1170 23.65° 2.51 • 453
318,1160 20.77° 2.83 .515
5 302,1150 18.10° 3.19 .575
6 286,1140 15.62° 3.59 .634
270,1130 13.32° 4.04 2/3
254,1120 11.19° 4.55 2/3
9 238,1110 9.21° 5.14 2/3
10 222,1100 7.38° 5.81 2/3
11 206,1090 5.68° 6.59 2/3
12 190,1080 4.11° 7 .so 2/3
13 174,1070 2.64° 8.57 2/3
14 158,1060 1.27° 9.86 213
15 142,1050 0.0° 11.41 213
o<.v < 2 ---<XH 3
1200-,
t .........
1-
!::S 11 50~
z
0
~
Gj 1100~
_j
w
1050
Vert. Horiz.
c< HC 6...fh_ 6rt.
.293 2. 3 3. 9 I
.338 1. 5 2.9
.375 1.0 2. 3
.405 .6 1. 8
.431 .5 1. 5
.453 .3 1.1
.476 .2 .9
.506 .1 • 7
. 534 .1 .s
.561 o.o .3
.588 0.0 .2
.614 o.o .1
.640 0.0 0.0
.665 o.o o.o
.690 o.o 0.0
1180
z 1150
0 -!-....-...
~~
W'-'
_j
w 1100
1150
6 5 .
10 15
STATIC FACTOR OF SAFETY
1
E L 1180
-
-;<.__
--------
~
-
I~ --------------
0 0.2 0.5
0 Hc
~~
0.7 0 1.0 2.0 3.0 0
6 VERTICAL
(FT)
~
15~------~------~------~------~------~------~----~--~~~~~--~
150 200 250 300 350 400 450 500 550 600
WEDGE STABILITY: SLOPED SLIDING PLANES
1.0 2.0 3.0 4.0
6 HORIZONTAL
( FT)
r 1200
~ 1150
l-1100
1050
~--------------------------------------------------------------------------------------FIGURE16
Wedge I Elevation Static F .S. <?< HC
Top 1190 oO .638
1170 49.83 .648
2 1150 19.29 • 666
3 1130 14.92 • 676
4 1110 13.23 .682
5 1090 12.34 .685
6 1070 11.78 .688
7 1050 11.4 1 .690
el.y - 2
o<H -3
1200
.........
1-
lL
.......... 1150~
z
0 -1-:g;
~ 1100-1
w
1050.
1200
Ha.x. Accel. 6H I 1180
2.724 2.2 .........
2.4563 t: 1150
2.1598 1.5 ..........
.9 • z
0
1. 6219
.5 -1-1.2578
1.0566 .2 :g; 1100
.92 97 • 1 ,w
_j
0 .w
1050 o 1o 20 30. 40 ·5o .62
STATIC FACTOR
OF SAFETY
2
/
.65
o(HC
0.7 1.0 2.0 3.0 0
MAX. ACCELERATION
(G)
""
1 2
6 HORIZONTAL
(FT)
1200
l-1150
l-11 00
7~------~------~------~----~------~------~------~-----.:-----~~ 1050
150 200 250 300 350 400 450 500 550 600
WEDGE STABILITY: HORIZONTAL SLIDING PLANES
~---------------------------------------------------------------------------------------FIGURE 17--~
o.oo
0
(1)10
~.;
.......
1-
a:
a::: wo _Jo:;
we
u' u a:
0
"'
s.oo \2.00
TIME-HISTORY OF WEDGE MOTIONS
CIRCLE B ,aH = 0.75 G, LA UNION E-w
16.00 2 4. 00 30.00 60.00 66.00 .oo
I 1 ~~
u I ------- ----------------~-----~ 1
------------u,
0
0
0
0
(0
0
I
0
"' --lHOICRTES EARTHQUAKE• RffORTE DEL ARCHIVO: UNI0850S19AI.l 616nQ: 8509190A E61ACION: UNIO N90E• EVEN TO: OA HOftA: 13 ;J1 :<9 OUftRClOH: e2 .7
7
~--~ w. ww
lL
UJo 0 1-. z..t
ILl
:L
w
CLN
(j)
0
62H fOINI6 •1 0.010 SECOHD6
INDICATES CRITICAL ACCELERATION, 0.375 0'6
INDICATE& WEOOE Ol6FLACEnfHI6 < RCCELERATJOH6.
~========================================~
0
0
------,------.-9Jj.._o_o __ __,6 '. o-o---, r-2 ~.:-o o 1 B • oo 2 4 . o o T I ME
LA UNION RESPONSE I DISPLACEMENT PLOTS
0
0
"'
0
0 ...
0
0
N
0
0
FIGURE 18
3.38
3.00
...........
rn2.63 -
z
0 2.25 t-
<(
0:: w
_J 1.88
w
u u
<( 1.50
_j
<(
0:: 1. 13
t-w
a_
If) 0.75
0.38
0.00
\ -\
" ~ " 1\VV
J \
/'JI -----------·
\ v\ r/ /
v ~
--
--r----
---------
I ---
,
----------
/ LA UNION
-·----~~ -~---~--·---
~PROJECT DESIGN
MCE RESPONSE
~SPECTRUM
~ ---·---··--------------------
....... ~::-:r-::-r---------
~ --......... 1---~
-------1------~------L. .. --
0.00 0.25 0.50 075 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00
PERIOD (SEC)
RESPONSE SPECTRUM -LA UNION E-W RECORD
~-------------------------------------------------------FIGURE19----~
2.63
-2.25
m ........
z 1.88
0
1--
~ 1.50
w
_J
w
u 1.13 u
<!
_J
<! 0.75
0::
1--u w 0.38
Q_
l/)
0.00
---
j
( ~M ~v
~
1--u
----------------------------------
__ ,.,., .. _______ ~_
1--------------
I/ ~-PROJECT DESIGN
MCE RESPONSE
SPECTRUM
~ (\ ----f---~t--TAFT
.. __
~~ -----· ~ ~t--~ -r----
-----~---. ----------------~
-------------------· ----------'-----------------
0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 2.75 3.00
PERIOD (SEC)
RESPONSE SPECTRUM -TAFT RECORD
~-------------------------------------------------FIGURE20 ----~
25
20
>-15 r-
l/) z w r-z
-10
l/)
<!
0:::
<!
5
o.
0 10 20
LA UNION
-
HYBRID
NOTE : PLOTS OF ARIAS INTENSITY FOR THREE
ACCELEROGRAMS USED 1 SCALED TO
0.75g PEAK GROUND ACCELERATION.
30 40
TIME
(SEC)
ARIAS INTENSITY
50 60 70
~--------------------------------------------------FIGURE 21
0
U)<D
d~
•I
1-0 • W•
W""
lL
U)o
1-"':: z-w
:E
w u a:o
...Ja:>
(L~
U)
D
.oo 5.00
TIME-HISTORY OF WEDGE MOTIONS
CIRCL B JOH = 0.75 G
15.00 20.00 25.00
I I
.oo 55.00 .oo
0
(D
0
u -----------------~· -----------------------__:._
0
-::u:-I o
(D
0
<.0 ------INDICATES EARTHQUAKE: IIA004 52.002.0 TAFT LINCOLN SCHOOL TUNNEL COMP S69E --2720 POINTS •T : 0.020 SECONDS I
-----INDICATES CRITICAL ACCELERATION, 0.375 G'S
-------INDICATES HEDGE OJSI'LACEMENTS «. ACCELERATIONS.
...-----.------------·-··-·-····-·······--·-·--~ -----y--------,
10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
TIME SECONDS
TAFT RESPONSE /DISPLACEMENT PLOT
55.00
0 •
""
0
<.0
0 co
0
L-----------------------------------------------------FIGURE22--~
SPILLWAY
STABILITY REPORT
SPILLWAY STABILITY REPORT
BRADLEY LAKE HYDROELECTRIC PROJECT
Prepared for
ALASKA POWER AUTHORITY
March 1988
STONE & WEBSTER ENGINEERING CORPORATION
DENVER, COLORADO 80111
FERC PROJECT NO. P8221-000
BRADLEY LAKE HYDROELECTRIC PROJECT
ALASKA POWER AUTHORITY
SPILLWAY STABILITY REPORT
TABLE OF CONTENTS
Section Section Title Page
1.0 INTRODUCTION 1-1
1.1 PURPOSE 1-1
1.2 SCOPE 1-1
1.3 SPILLWAY SAFETY CRITERIA 1-1
2.0 DESCRIPTION OF PROJECT FEATURES 2-1
2.1 GENERAL 2-1
2.2 OGEE SECTION 2-2
2.3 NON-OVERFLOW SECTIONS 2-3
2.4 GEOLOGIC CONDITIONS 2-4
3.0 DESIGN EARTHQUAKE REGIME 3-1
3.1 SEISMOTECTONIC SETTING 3-1
3.2 DESIGN RESPONSE SPECTRA 3-3
3.3 ACCELEROGRAM DEVELOPMENT 3-4
4.0 STABILITY CRITERIA 4-1
4.1 GENERAL 4-1
4.2 LOADS 4-1
4.2.1 Deadweight 4-1
4.2.2 Ice 4-2
4.2.3 Hydrostatic 4-2
4.2.4 Earthquake 4-2
4.2.5 Wind 4-3
4.2.6 Uplift 4-3
4.2.7 Temperature 4-4
4.3 LOADING CONDITIONS 4-4
4.4 ACCEPTANCE CRITERIA 4-5
4.4.1 Stability Requirements 4-5
4.4.2 Minimum Allowable Stress 4-7
4.4.3 Shear-Friction Factor of Safety 4-8
5.0 METHODS OF ANALYSIS 5-l
5.1 STATIC METHOD 5-1
5.2 FINITE ELEMENT METHOD 5-l
5.3 SARMA METHOD 5-2
6.0 STATIC ANALYSIS 6-1
6.1 STABILITY ANALYSIS 6-1
6.2 RESULTS 6-2
7.0 FINITE ELEMENT ANALYSIS 7-1
7.1 STRESS ANALYSIS 7-1
7.2 RESULTS 7-3
TABLE OF CONTENTS (Cont'd)
Section Section Title Page
8.0 SARMA ANALYSIS 8-1
8.1 STABILITY ANALYSIS 8-1
8.2 RESULTS 8-2
9.0 CONCLUSIONS 9-1
9.1 CRITICAL CASES 9-1
9.2 SUMMARY OF STABILITY CONDITIONS 9-1
10.0 BIBLIOGRAPHY 10-1
Figure
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
LIST OF FIGURES
Title
Project Layout Map
Main Dam Area -General Arrangement
General Arrangement -Spillway
Project Response Spectra
Hybrid Accelerogram
Static Spillway Model
Case I -Static Analysis-Base El 1124
Case II -Static Analysis-Base El 1124
Case IV -Static Analysis-Base El 1124
Finite Element Model -Base El 1160
Finite Element Model -Base El 1150
Finite Element Model -Base EL 1124
Finite Element Analysis: Case III -Max Tensile Stresses -Base
EL 1160
Finite Element Analysis: Case III -Max. Compressive Stresses -
Base EL 1160
Finite Element Analysis: Case V -Max. Tensile Stresses -Base EL
1160
16 Finite Element Analysis: Case V -Max. Compressive Stresses -
Base EL 1160
17 Finite Element Analysis: Case III -Max. Tensile Stresses -Base
EL 1150
18 Finite Element Analysis: Case III -Max. Compressive Stresses -
Base EL 1150
19 Finite Element Analysis: Case V -Max. Tensile Stresses -Base EL
1150
20 Finite Element Analysis: Case V -Max. Compressive Stresses -
Base El 1150
21 Finite Element Analysis: Case III -Max. Tensile Stresses -Base
EL 1124
22 Finite Element Analysis: Case III -Max. Compressive Stresses -
Base EL 1124
23 Finite Element Analysis: Case V -Max. Tensile Stresses -Base EL
1124
24 Finite Element Analysis: Case V -Max. Compressive Stresses -
Base EL 1124
25 SARMA Analysis Model, Ogee Sections -Sheet 1
26 SARMA Analysis Model, Ogee Sections -Sheet 2
27 SARMA Analysis Model, Non-Overflow Sections
28 SARMA Analysis: Base EL 1160 -Ogee
29 SARMA Analysis: Base EL 1150 -Ogee
30 SARMA Analysis: Base EL 1130 -Ogee
31 SARMA Analysis: Base EL 1124 -Ogee
32 SARMA Analysis: Base EL 1160 -Left Abutment
33 SARMA Analysis: Base EL 1124 -Right Abutment
34 Spillway Stability Analysis Summary -Sheet 1
35 Spillway Stability Analysis Summary -Sheet 2
36 Spillway Stability Analysis Summary -Sheet 3
SPILLWAY STABILITY REPORT
1.0 INTRODUCTION
1.1 PURPOSE
The purpose of the stability analysis reported herein is to document that
the design of the Bradley Lake Hydroelectric Project spillway meets the
required stability criteria as defined for the Project.
1.2 SCOPE
This report presents the analysis methods and results for stability
analysis of the spillway located adjacent to the main dam at Bradley Lake.
The spillway was analyzed to determine stresses and factors of safety under
various static and dynamic loading conditions. The spillway was also
analyzed to predict maximum potential movement under seismic loading
conditions. Static loading conditions included normal maximum reservoir
operating level and Probable Maximum Flood (PMF). Dynamic loading
conditions included analysis of the Maximum Credible Earthquake (MCE) with
normal maximum reservoir level and with a low reservoir level, as well as a
pseudostatic analysis of a lower intensity earthquake during construction.
A Design Basis Earthquake (DBE) was also evaluated to predict potential
movement in order to provide a comparison of MCE and DBE results.
1.3 SPILLWAY SAFETY CRITERIA
The basic requirement which must be met by the spillway design is that the
reservoir must be retained under all conditions evaluated.
Spillway safety criteria were established to aid in evaluating spillway
stability and performance. These criteria established limiting allowable
stresses and minimum factors of safety. Each loading condition was
classified as either a Usual, Unusual, or Extreme Condition, and the
acceptance criteria were based on currently accepted recommended minimum
factors of safety for the applicable loading condition (Ref. 1).
3777R/20SR/LS 1-1
2.0 DESCRIPTION OF PROJECT FEATURES
2.1 GENERAL
The Bradley Lake Hydroelectric Project is located on the Kenai Peninsula in
southcentral Alaska. The project will develop the hydroelectric energy
potential of Bradley Lake, a natural lake at El 1080, and will have an
initial (2 unit) nominal generating capacity of 90MW. The Project
consists of water diversion facilities, a concrete-faced rockfill dam, a
spillway and an underground power conduit leading to a surface powerhouse
with tailrace discharging into Kachemak Bay. See Figure 1 for the general
Project layout.
The spillway included for reservoir discharge is a mass concrete ungated
ogee crest spillway with the crest at El 1180. The spillway was designed
to accommodate flows within the range from zero flow up to and including
the Probable Maximum Flood (PMF) flow of 23,800 cfs. See Figures 2 and 3
for the location and general arrangement of the spillway.
The spillway will be constructed in the low saddle area to the right of the
main dam (looking downstream) and will be in line with the main dam
baseline. This saddle is formed by a steep rock cliff on the right side of
the spillway and a knob of rock on the left that constitutes both the right
dam abutment and left spillway abutment. Non-overflow concrete sections
will connect the ogee section with the rock abutments on each side.
Although the
neglecting any
structure, that is spillway is designed as a gravity
load transfer to the abutments, keys are provided at
and to provide better contraction joints
seismic stability.
to allow some load transfer
The spillway will be founded on bedrock. A grout curtain will be developed
upstream of the spillway baseline to a depth of 2/3 of the design head at
the foundation level. A line of foundation drains will be provided
downstream of the grout curtain.
3777R/205R/LS 2-1
The spillway will be provided with an 8 ft wide by 8 ft high drainage
gallery generally at the rock surface to allow for inspection and
maintenance of the foundation drains. A V-notch measuring weir is provided
at the discharge of the drainage gallery to allow measurement of foundation
drain seepage rates.
The vertical rock cuts adjacent to the spillway aprons will be lined with
concrete training walls to prevent erosion and undercutting of the rock.
These walls will be tied to the rock with rock bolts. The wall along the
west side of the spillway channel will be extended beyond the end of the
rock cut to control flow near the diversion tunnel outlet structure and to
prevent damage from lateral flow.
The agee and non-overflow sections will be constructed of a lower strength
mass concrete core (3,000 psi compressive strength at 28 days) and a shell
approximately three feet thick of a higher strength concrete (4,000 psi
compressive strength at 28 days). The core concrete mix is designed to
limit heat of hydration on the mass sections of the pour, whereas the shell
concrete mix is designed to provide durability. The shell concrete will be
placed concurrently with the core concrete to assure a monolithic
structure. For the purpose of establishing allowable stresses, the effect
of the higher strength of the shell concrete is neglected in the analysis.
2.2 OGEE SECTION
The agee section will have a 175 foot crest length (E1 1180). A 5 foot
long concrete apron at the end of the reverse curve on the downstream slope
will be cast against rock. The drainage gallery will be provided for the
full length of the agee section.
In order to minimize rock excavation and concrete quantities, the spillway
is designed to discharge to two different downstream elevations. The
spillway apron will discharge at approximately El 1135 for a 105 foot
length and at approximately El ll65 for a 70 foot length. The
corresponding ogee section base elevations on natural rock contours at the
3777R/205R/LS 2-2
upstream edge of the cross section range from approximately El 1120 to 1140
at the lower section and El 1140 to 1160 at the upper section. The natural
and excavated rock contours at the downstream edge vary from approximately
El 1125 to 1135 at the lower section and El 1160 to 1165 at the upper
section.
The agee section at each discharge elevation will consist of two blocks
separated by a contraction joint. Although keys are provided at these
joints, the spillway was analyzed using a unit-width gravity
neglecting any load transfer between blocks and to the abutments.
method,
Static
stability analyses were performed for the agee section with base elevations
at El 1124, 1135, 1150, and 1160. The spillway is founded on the varying
existing rock surface. The above range of assumed base elevations
appropriately brackets actual foundation elevations. Finite element
analyses were performed for base elevations 1124, 1150, and 1160.
2.3 NON-OVERFLOW SECTIONS
Non-overflow sections, extending up to El ll95, will be provided at both
ends of the agee section to provide the transition to the adjacent rock.
The left non-overflow section (looking downstream) will be approximately 72
feet long and have a foundation on natural and excavated rock contours
varying from approximately El 1160 to 1190. The right non-overflow section
will be approximately 30 feet long and contain an emergency exit and
ventilation shaft for the drainage gallery. Its base will be founded at
elevation 1145 and keyed into an inclined rock cut over its entire height.
The keying of the right non-overflow section into rock and to the adjacent
agee section will provide added stability. To verify stability of the left
non-overflow section, sections with bases at El 1160 and El 1180 were
evaluated by the static method neglecting the additional restraint
provided by keys and varying cantilever heights. The overturning and
3777R/205R/LS 2-3
sliding stability of the right non-overflow section with base El ll45 was
deemed not to be critical in the upstream-downstream direction and was not
analyzed separately for this condition. However, due to its short length
relative to its height, the right non-overflow sect ion was evaluated for
seismic stability in the east-west direction.
2.4 GEOLOGIC CONDITIONS
The crest of the main dam structure is located approximately 500 ft
downstream of the existing Bradley Lake outlet and will dam the Bradley
River between two bedrock knobs. The right dam abutment, spillway, and
diversion tunnel/gate shaft are founded predominantly in massive graywacke
that is generally slightly weathered and hard with occasional deep open
joints and fractures, with weathering generally limited to joints. Jointing
is well developed, widely to very widely spaced and cross-oriented, giving
the exposed rock a large-scale blocky appearance.
The spillway lies east of the main dam in a low saddle. Bedrock in the
spillway saddle, exposed during excavation performed under the Site
Preparation Contract, displays glacial striations and chatter marks left by
glacial ice that once occupied the spillway saddle and Bradley River
channel. The bedrock also exhibits a smooth polished finish, caused by ice
scouring or by glacial meltwater flowing through the spillway saddle.
The east cliff has continuous bedrock exposure, and the west abutment knob
has numerous outcrops near the crest elevation of the spillway. These
outcrops are mapped as graywacke (98%) with very minor argillite (2%). The
rock appears fresh and is hard to very hard. The bottom of the saddle has
bedrock outcrops and the rock appears identical to the high cliff and rock
knob. Hence, the saddle does not appear to have been formed in weaker
lithology. This implies a structural influence for the formation of the
spillway saddle.
3777R/205R/LS 2-4
A 20-foot-wide, rubble-filled notch (Lineament 3 in the Geotechnical
Interpretive Report, Ref. 13) was located, prior to Site Preparation work,
at the base of the steep cliffs that form the eastern edge of the spillway
site. The lineament is oriented approximately Nl0° W with a near vertical
dip. Borings intersected limited zones of highly fractured rock and clayey
gouge, and encountered loss of circulation water and high flows during
pressure testing. There is no apparent change in lithology across this
feature in outcrop or in the boreholes. From aerial photographs, a
lineament can be projected through the area and is suspected to be a minor
fault or a swarm of closely spaced joints. A small gouge zone is visible at
the lake edge, roughly projecting along the cliff face and perpendicular to
the spillway baseline. The saddle represents a minor fault or shear zone
that has been eroded by glaciation and partially obscured by subsequent
glacial and post-glacial deposition. Trenching exposed a 1 to 3 ft wide
shear zone, paralleling the base of the cliff.
Borings throughout the spillway area show a preponderance of graywacke. Two
boreholes, RM43 and DH7EX, were drilled at an angle across the spillway from
west to east. Neither boring penetrated all the way to the plane of the
cliff east of the spillway. However, these two borings, and others drilled
near to the east cliff, showed evidence of highly to intensely fractured
rock with some gouge present in one area. In several borings, circulation
water returns were lost and not recovered.
Several small groundwater seeps, which originate from bedrock fractures,
flow from the high ground to the east and were noted on the face of the east
cliff. Major joints in and adjacent to the spillway trend and dip N84° W,
84° S and N25° E, 72° SE. Where trenching exposed joints, they often made
small amounts of water (up to 0.5 gpm) for periods of less than 12 hours.
Overburden depths in the area of the spillway are quite variable, originally
ranging from 2 ft on the west flank to an estimated 17 ft near the east
cliff. As a result of excavation during Site Preparation, very little
overburden remains on the crest of the saddle. Throughout the spi 11 way
area, the soil materials consist of till and alluvium with cobbles and
boulders (greater than 5 ft in dimension) in a silty, sandy gravel matrix.
Very coarse and bouldery graywacke talus lies below the eastern cliff and
extends from the saddle to the bottom of the spillway.
3777R/205R/LS 2-5
All overburden from the upstream edge of the spillway to the downstream pool
(to El 1060) will be removed to the bedrock surface. Overburden varies in
thickness from 2 ft near the crest to greater than 60 ft (locally) near the
downstream pool. Due to the presence of geologic structure through the
saddle and the proximity of steep cliffs to the east, the overburden is
expected to contain talus in addition to the glacial till. The bedrock
surface is expected to be locally somewhat fractured and weathered. This
will require some grouting and limited dental concrete for foundation
improvement. Some ripping and limited blasting will be necessary to meet
design grade and flow path requirements in the apron area, in addition to
the specified controlled blasting in the concrete foundation area.
A grout curtain is designed to transect upstream base of the spillway
structure. The curtain extends eastward about 80 ft from the cliff face.
This will be done to seal a series of joints, which are oriented
approximately N48° W. These fractures may connect the reservoir with the
downstream pool, so they will be grouted to reduce potential leakage.
More detailed geologic data is provided and referenced in the Geotechnical
Interpretive Report, contained in the General Civil Construction Contract as
Volume 6 and in the report entitled 1987 Geotechnical Exploration Program
(Ref 12).
3377R/205R/LS 2-6
3.0 DESIGN EARTHQUAKE REGIME
3.1 SEISMOTECTONIC SETTING
The detailed project seismic design studies and parameters are provided in
two reports by Woodward-Clyde Consultants (Ref. 2 and 3).
Southern Alaska is one of the world's more seismically active regions. The
primary cause of seismic activity in southern Alaska, including the site
area, is the stress imposed on the region by the relative motion of the
Pacific and the North American tectonic plates at their common boundary.
The Pacific plate is moving northward relative to the North American plate
at a rate of about 2-1/2 inches/year causing underthrusting of the Pacific
plate. This underthrusting results primarily in compressional stresses
which cause folds, high-angle reverse faults, and thrust faults to develop
in the overlying crust. A counterclockwise rotational effect also induces
strike-slip faulting parallel to the plate boundary.
The boundary between the plates where the underthrusting occurs is a
northwestward-dipping megathrust fault or subduction zone. The Aleutian
Trench marks the surface expression of this subduction zone and is located
on the ocean floor approximately 185 miles southeast of Bradley Lake. The
orientation of the subduction zone is inferred along a broad inclined band
of seismicity, referred to as the Benioff Zone, that dips northwest from
the Aleutian Trench, and is approximately 30 miles beneath the surface at
the Bradley Lake site. Great earthquakes (Richter magnitude M =8 or s
greater) and large earthquakes (M =7 or greater) have occurred s
historically throughout the region and can be expected to occur in the
future. Historically (1899 to date), eight earthquakes ranging between
M =7.4 and M =8.5 have occurred within 500 miles of the site. (Table 3 s s
in the Dam Stability Report provides a representative sunnnary of historic
seismic events within 100 miles of the area.)
3777R/205R/LS 3-1
Bradley Lake is situated on the overriding crustal block above the
subduction zone and between the Castle Mountain fault to the northwest and
the Patton Bay-Harming faults to the southeast. All of these faults have
documented Holocene or historic surface ruptures. Because of the active
tectonic environment, activity is conceivable on other faults, such as
those found near or at the project site between the known active faults
mentioned above.
Two faults of regional extent exist at or near the site.
Fault trends southwest beneath Kachemak Bay to the
The Border Ranges
northwest of the
project, and the Eagle River fault crosses the southeastern end of Bradley
Lake at about the same trend. While no evidence of recent activity along
these faults has been found in the site area, recently reported data
indicates recent activity on the Eagle River Fault near Eklutna (125 mi NE
of the site). Given the tectonic setting, it is reasonable to consider
these faults as potentially active.
In addition to the nearby regional faults, the site is crossed by two large
local faults, the Bradley River Fault and the Bull Moose Fault, and a
number of confirmed and probable smaller faults. The dominant trend is
northeasterly, paralleling the regional trend. The larger local faults,
particularly the Bradley River, are considered as potentially capable of
independent earthquake generation, while any of the local faults could
possibly move in sympathetic response to earthquakes occurring on the
regional faults.
It is therefore concluded that the site will probably experience at least
one moderate to large earthquake during the life of the proposed project.
The possibility of significant ground rupture exists but is much less
subject to prediction and is considered to have a much lower probability.
3777R/205R/LS 3-2
3.2 DESIGN RESPONSE SPECTRA
The response spectra utilized for this analysis were taken from a report
prepared by Woodward-Clyde Consultants (WCC) for the Army Corps of
Engineers (Ref. 2). The report documents the work performed by WCC to
develop parameters for what the Corps terms the "design maximum earthquake"
and the "operational base earthquake" , henceforth called Maximum Credible
Earthquake and Design Basis Earthquake, respectively. The Maximum Credible
Earthquake (MCE) is defined as the most severe earthquake believed to be
probable which could affect the site. The Design Basis Earthquake (DBE) is
less severe, and is defined as the seismic level which is considered as
likely to occur during the life of the project. Maximum Credible
Earthquakes are normally used as a basis for determining whether or not
certain structures can structurally withstand extreme events having remote
probabilities of occurring, regardless of damage level. Design Basis
Earthquakes are used as a basis for estimating the maintenance and other
costs resulting from events expected to occur, and for design of
non-critical structures where severe damage and loss of .function in a major
seismic event are considered an acceptable risk. The response spectra for
the MCE will be used in the seismic stability analysis of the spillway.
Based on their work on the seismicity of the site, wee proposed two
possible response spectra for the "design maximum earthquake", the
equivalent of the MCE. The one which was expected to control was based on
rupture of one of the faults nearest the site. The resulting earthquake
would have a magnitude of M =7. 5, peak ground acceleration of 0. 75g at s
the site, and a significant ground motion duration of 25 seconds. This
event would have a response spectrum roughly corresponding to the upper
smooth curve on Figure 4.
The other possible MCE was an event tied to the Benioff Zone roughly 30
miles beneath the site. This event would have a magnitude of M =8.5, s
peak ground acceleration of O.S5g at the site, and a significant ground
motion duration of 45 seconds. It was not expected to be the controlling
event unless the faults in the innnediate vicinity of the site could be
shown to be inactive.
3777R/205R/LS 3-3
A third response spectrum proposed by wee was an event with a peak ground
acceleration approximately one half that of the MCE. This was used for the
DBE, resulting in a peak ground acceleration of 0.35g (Fig. 4).
3.3 ACCELEROGRAM DEVELOPMENT
Because the near field
mega thrust M =8. 5 event, s
crustal M =7.5 event is more severe than the s
in terms of both peak parameters and spectral
accelerations throughout the frequency range of interest, an accelerogram
for the crustal event is of primary interest. The mega thrust event is
considered in detail in the wee reports, but was utilized in design only
for parametric comparative purposes.
Since all critical structures of the Bradley Lake Project are founded on
bedrock, accelerograms recorded on rock from large magnitude earthquakes
having similar parameters to those listed above for the crustal event would
ideally be used for the required analyses. More importantly, the response
spectra of any accelerograms used for design should match, in an average
sense, the curve shown in Figure 4. At the time the analysis was being
performed, no accelerograms recorded on rock in the near field of large
magnitude earthquakes (M =7. 5) were available from anywhere in the world, s
including Alaska. Consequently, available accelerograms from historical
earthquakes having appropriate peak and spectral characteristics over a
broad period range, even when scaled, were not available for use.
Since no actual accelerogram was available, a composite hybrid accelerogram
was derived for the dam and spillway stability analyses from the historical
accelerograms of two earthquakes having appropriate characteristics. This
approach has been previously used for other studies, including those
performed by the California Department of Water Resources for Oroville Dam,
and is considered an appropriate state-of-the-art method for simulation of
strong motion events.
3777R/ 205R/ LS 3-4
After examining the response spectra for recorded accelerograms from a
number of earthquakes in the United States and abroad, it was concluded
that a suitable accelerogram for the M =7.5 crustal event could be s
obtained by combining the S69° E component of the Taft record from the 1952
Kern County, California earthquake and the East-West component of the San
Rocco record from the September 15, 1976 Friuli, Italy earthquake. The
Taft record was scaled by a factor of 3.5 and was used to represent the
hybrid earthquake from time 0.00 to 2.32 seconds and from 4.32 seconds to
the end. The portion of the Friuli record from time 2.14 to 4.10 seconds
was scaled by a factor of 3.2 and inserted into the scaled Taft record,
replacing the port ion of the Taft record from time 2. 34 through 4. 30
seconds in the hybrid record. In effect, the portion of the Friuli record
with the highest accelerations was spliced into the high-acceleration
portion of the Taft record, resulting in a record with greater duration and
a greater proportion of relatively high acceleration peaks. The resulting
accelerogram, called the Hybrid record, is shown on Figure 5, and its
response spectrum is compared to the spectrum recommended by WCC in Figure
4. The significant duration of the Hybrid record is 28.8 seconds, which is
slightly longer than the 25 second MCE proposed by WCC. This longer event
duration, when combined with the greater density of high acceleration peaks
from the combined records, results in a design record which is
conservatively intense and definitely on the safe side when used to
simulate the project MCE.
3777R/205R/LS 3-5
4.0 STABILITY CRITERIA
4.1 GENERAL
The spillway was evaluated for sliding stability, for maximum compressive
and tensile stresses, and for the maximum permanent deformation under
dynamic loading. The evaluation of the sliding stability and stresses was
performed in general agreement with the gravity method as presented in
"Engineering Guidelines for the Evaluation of Hydropower Projects", FERC
0119-1 (Ref. 11). The gravity method was used in both the static and
finite element analyses. Permanent dynamic deformations were evaluated
using the Sarma method of analysis.
The maximum stresses were evaluated for loading Cases I, II, and IV by the
static method and for Cases III and V (MCE) by the finite element method.
For these analyses, uplift in an uncracked section was not included as an
active external force (i.e., uplift does not contribute to overturning)
since it is an internal pressure resisted by the structure until cracking
occurs. The calculated stress was combined with the uplift pressure by
superposition to determine the effective stress. For static loading
conditions, uplift in a cracked section was considered as an active force
(i.e., uplift in a crack acts as an external hydrostatic force).
The sliding stability was evaluated for loading Cases I, II, and IV by
calculating the shear-friction factor of safety in the static analysis.
The sliding stability for dynamic loading conditions (Cases III and V) was
evaluated by determination of the maximum permanent dynamic deformation
using the Sarma method of analysis.
4.2 LOADS
4.2.1 Dead Weight
The weight of the spillway was evaluated using a concrete unit weight of
145 lbs/cu ft.
3777R/205R/LS 4-1
4.2.2 Ice
An ice load of 12,000 lbs/lin ft was applied at El 1179. This was based on
an assumed ice thickness of 28 inches.
4.2.3 Hydrostatic
Hydrostatic pressures were calculated based on a unit weight of water of
62.4 lbs/cu ft.
For the Probable Maximum Flood (PMF) condition, the hydrostatic forces
downstream of the crest were considered negligible. While the horizontal
hydrostatic force on the upstream face was based on the PMF reservoir
level, the vertical hydrostatic force on the upstream face was reduced by
five feet to account for drawdown at the spillway crest.
Due to the spillway elevation relative to lake bottom, forces due to
accumulated sediment were determined to be negligible.
4.2.4 Earthquake
Two earthquakes were evaluated for the stress analyses: the Maximum
Credible Earthquake (MCE) with a peak horizontal ground acceleration of
0.75g, and a Construction Condition Earthquake with a peak horizontal
ground acceleration of O.lOg. The Sarma analysis also evaluated the Design
Basis Earthquake (DBE) with a peak horizontal ground acceleration of
0.35g. Vertical ground accelerations in all cases were assumed equal to
2/3 the horizontal accelerations. The pseudostatic and finite element
analyses considered horizontal and vertical seismic forces to act
simultaneously. The
acting alone after
Sarma analysis
both components
evaluated horizontal seismic forces
of acceleration were utilized to
calculate the critical horizontal acceleration. Static analysis assumed an
acceleration equal to the peak ground acceleration. Finite element
analysis was based on the Project Response Spectra (Figure 4) and the Sarma
analysis was based on the developed Hybrid Accelerogram (Figure 5).
3777R/205R/LS 4-2
The hydrodynamic earthquake pressure was based on the Westergaard added
mass approach (Ref. 7 & 8) for the finite element method and based on the
Zangar formula (Ref. 1 & 6) for the static and Sarma methods. The
difference in magnitude of these forces utilizing the two differing methods
is not significant to the results.
4.2.5 Wind
A wind loading of 90 psf was evaluated as an alternate loading to the
Construction Condition Earthquake but was found to be a less critical
case. This loading is equivalent to Uniform Building Code (UBC) Exposure C
with a 120 mph wind speed and a 1.15 Importance Factor.
4.2.6 Uplift
The uplift pressure at all lift lines above the foundation was assumed to
vary linearly from the full headwater pressure at the upstream face to zero
at the downstream face. The PMF headwater pressure at the upstream face
was based on 50% headrise above normal headwater pressure. Uplift
pressures at the downstream face were assumed as negligible for the PMF
case.
At the rock-concrete interface the foundation drains were considered SO%
effective, so the uplift was assumed to vary linearly from full headwater
pressure at the upstream face to 50% of headwater pressure at the
foundation drains, then decreasing linearly to zero at the downstream face,
(Ref. 4). The drainage gallery will permit inspection, monitoring, and
maintenance of drains.
For static cracked sections, the uplift was assumed at full headwater
pressure for the length of the crack then decreasing linearly to zero at
the downstream face. Uplift for the dynamic analysis was not revised from
that developed for the Normal Condition.
Uplift was assumed to act over 100 percent of the base area.
3777R/205R/LS 4-3
4.2.7 Temperature
Loadings from volumetric changes due to temperature change were not
considered in the analysis since the joints are not grouted. While the
loadings were not evaluated, consideration was given to temperature effects
in location of contraction joints and in material selection.
4.3 LOADING CONDITIONS
Five loading combinations were considered, as follows:
Case I Normal Reservoir -Usual Condition
1. Normal Max. Reservoir El 1180
2. Uplift and seepage forces
3. Dead loads
4. Ice at El 1179.0
Case II Probable Maximum Flood (PMF) -Unusual Condition
1. Max. Reservoir El 1191
2. Uplift and seepage forces
3. Dead loads
Case III Earthquake -Extreme Condition
1. Normal Max. Reservoir El 1180
2. Uplift and seepage forces
3. Dead loads
4. Ice at El 1179.0
5. Maximum Credible Earthquake (0.75g)
Case IV Construction Case -Unusual Condition
3777R/205R/LS
1. Reservoir water surface at El 1065
2. Dead loads
3. Construction Condition Earthquake (O.lg) or
wind load
4-4
Case V Low Reservoir with Earthquake -Extreme Condition
1. Reservoir below El 1124 (no hydrostatic)
2. Dead 1 oads
3. Maximum Credible Earthquake (0.75g)
Ice loads were not included in the SARMA analysis.
4.4 ACCEPTANCE CRITERIA
4.4.1 Stability Requirements
Maximum allowable stresses and minimum required factors of safety are as
specified in Table 4-1. These values are based on factors of safety as
recommended in Design of Gravity Dams (Ref. 1) for Usual Loadings, Unusual
Loadings, and Extreme Loadings and factors of safety specified in project
design criteria.
TABLE 4-1
Case I Case II Case III Case IV Case V
Normal PMF Earthguake Construction Low Res.
Stresses:
Concrete (f'c = 3000 psi)
Assumed Safety factor 3.0 2.0 1.0 2.0 1.0
Compression, psi 1000 1500 3000 1500 3000
Tension, psi ;'r 60 90 270 90 270
Rock (40 ksf = 280 psi )~·r:r:r
Assumed Safety factor 2.0 1.5 1.1 1.5 1.1
Compression, psi 140 185 250 185 250
Tension, psi -:c 0 0 ..,•c 0 0
Sliding:
Shear -Friction in
Concrete
Safety factor 3.0 2.0 1.0 2.0 1.0
(in Concrete and at
Rock/Concrete Interface)
On Rock Foundation Joints
and Faults
Safety factor 4.0 3.0 1 • 2-lrlr 3.0 1. 2-lr:r
3777R/205R/LS 4-5
*For Usual and Unusual Conditions, tensile resistance is allowed only above
the rock-concrete interface. For dynamic stress analysis by the finite
element method, the tensile stress at the rock-concrete interface shall not
exceed the allowable tensile capacity of the concrete. Tensile capacity of
concrete is increased 50 percent above static tensile capacity for dynamic
loading conditions.
~rlcSafety factors are not relevant to the SARMA analysis.
~rlrlcSafety factors for rock bearing capacity are applied against the
allowable bearing capacity, which already includes a minimum factor of
safety of two relative to ultimate bearing capacity. Thus actual factors of
safety are at least double those shown and exceed the recommended factors of
safety given in References 1 and 11.
The structure may be considered stable against overturning when the minimum
calculated stress, without uplift, meets the requirements of paragraph 4.4.2
and when the maximum compressive stresses are 1 ess than those specified
above. The structure may be considered stable against sliding when the
shear-friction factor of safety, as given in paragraph 4.4.3, is greater
than the value specified above.
3777R/205R/LS 4-6
4.4.2 Minimum Allowable Stress
The evaluation of stresses assumes that the concrete has tensile strength.
In order not to exceed the allowable tensile stress, the minimum allowable
stress without uplift (assuming compression is positive and tension is
negative) is determined by:
where:
Minimum allowable stress = WH -ft
w =
H =
unit weight of water
depth below reservoir surface
ft = allowable concrete tensile strength at lift surfaces which
includes the safety factor as given in Table 4-1 for
concrete; zero at rock-concrete interface.
For static cases, tensile strength = 61 x compressive strength
For dynamic cases, tensile strength = 1501 x static tensile strength
If the calculated minimum stress (without uplift) is less than the minimum
allowable stress given above then cracking is assumed to occur. If the
minimum stress for a Usual or Unusual case is less than zero, i.e., tension
present, then reinforcing steel is required to limit crack development.
3777R/205R/LS 4-7
4.4.3 Shear-Friction Factor of Safety
The shear-friction factor of safety is the ratio of resisting to driving
forces as computed by:
where:
Q
Q
c
A
N
0
H
=
=
=
=
CA + N Tan 0
H
shear-friction factor of safety
unit cohesion
300 psi at concrete on concrete (10% f'c)
= 160 psi at concrete on rock
=
=
=
area of base in contact (uncracked section)
summation of normal loads including uplift forces
internal friction angle
= 45° at concrete lift lines
= 45° at concrete on rock
= summation of horizontal driving forces
The passive resistance of the vertical rock against which the spillway
apron is cast was not included in the resisting forces, due to potential
water scouring and sloping of the bedrock contour surface away from the
spillway apron.
3777R/20SR/LS 4-8
5.0 METHODS OF ANALYSIS
5.1 STATIC METHOD
The static method analyzes the spillway using the gravity method by solving
for the summation of forces and moments acting on a theoretical one foot
vertical strip of the structure. The effects of load transfer across the
keyed joints and the effects of transversely sloping foundations were
conservatively omitted. Seismic loads for Case IV, Construction Case, were
included as pseudostatic forces.
This analysis calculates the maximum compressive and tensile stresses,
shear friction factors of safety, summation of vertical and horizontal
forces, location of resultant force and, if present, the crack length.
This is the simplest of the methods of analysis used and is useful in
evaluating the overall stability for various loading conditions. It does,
however, possess a number of limitations. A significant limitation is that
it does not take into account the frequency characteristics of a dynamic
loading or the natural frequency of the structure. Where large peak
accelerations are expected, as for Cases III and V, a pseudostatic analysis
is considered to be inappropriate for analysis of dynamic stability. For
this reason, a dynamic analysis has been performed.
5.2 FINITE ELEMENT METHOD
Due to the potential seismic risk and the magnitude of the seismic design
accelerations, with capable faults near the spillway site and located in a
seismic Zone 4 region as specified in "Engineering Guide! ines" (Ref. 11),
the spillway was dynamically analyzed. In order to take into account the
frequency characteristics of the structure and the seismic loading, a
two-dimensional finite element analysis was performed on the concrete ogee
sections. The non-overflow sections were not analyzed for dynamic stresses
for Cases III and V, but are considered adequate for seismic stresses by
comparison of cross sections and loadings with the ogee sections analyzed.
3777R/205R/LS 5-1
STARDYNE (Ref. S) was used to perform two-dimensional linear elastic
analyses to obtain the stresses in the spillway and the reactions at the
base for the loading conditions considered.
The STAR program of STARDYNE was used to do the static analysis for water
load, ice thrust force, and dead weight of spillway and to perform the
frequency and modal shape analysis. The seismic analysis was done by the
DYNRE4 program of STARDYNE.
5.3 SARMA METHOD
In order to predict the maximum potential permanent deformation of the
spillway under seismic loading conditions, the spillway sections were
analyzed by the SARMA (Seismic Amplification Response by Modal Analysis)
computer program (Ref. 9). This program has been developed by Stone &
Webster Engineering Corporation and qualified for the Nuclear Regulatory
Commission to evaluate potential deformations utilizing the methods of S.
K. Sarma and N. N. Ambraseys for seismic amplification in fill structures
and N. M. Newmark for the cumulative displacement under dynamic
excitation. This program was used to evaluate potential movement of the
various spillway sections, conservatively neglecting cohesion at the
rock-concrete interface.
This method is one that is often used to model the response of dams to
earthquakes when the pseudostatic method is inappropriate. It is commonly
used by the Army Corps of Engineers in modeling for new dam design and for
analysis of existing dams.
The Sarma method starts with the calculation of resonant frequencies and
modal response shapes of the structure. The next step is calculation of
participation factors for a given potential failure wedge or block. These
factors describe how much effect each of the modes of oscillation will have
on the potential failure wedge. Once this is accomplished, the
accelerations of the wedge in each mode in response to the earthquake
accelerogram are calculated and the displacements from the various modes
3777R/205R/LS S-2
due to each pulse are combined. The result is a time-history of the
accelerations the
earthquake if it
foundation.
wedge would experience as
remained attached to the
a result of the chosen
rest of the structure or
Once the time-history of acceleration pulses of the individual wedge is
known, the cumulative displacement is calculated by Newmark's sliding block
procedure (Ref. 10). In this procedure, the wedge is assumed to remain
fully attached as long as the average acceleration of the wedge is less
than a specified critical (or break-free) acceleration. When the
acceleration exceeds the critical acceleration, the wedge slides relative
to its support until it comes to rest during a subsequent reversal of the
acceleration. The total displacement of the wedge is taken as the sum of
all the increments of movement that occur during a particular earthquake
record.
In this case, the entire spillway was taken as a single wedge supported by
the foundation for analytic purposes.
3777R/205R/LS 5-3
6.0 STATIC ANALYSIS
6.1 STABILITY ANALYSIS
Load Cases I, II and IV were evaluated by the static method. Seismic loads
in Case IV were included as pseudostatic forces.
Due to the relative flexibility of the spillway apron with respect to the
remaining structure, the apron area was neglected in calculations and
assumed to be detached. The toe was assumed at a hypothetical extension of
the downstream face slope, as shown in Figure 6. Since the apron
deadweight exceeds the uplift in that region, uplift under the apron was
also neglected. Uplift was also conservatively assumed to act across the
gallery width.
Any Usual or Unusual case where the minimum stress, without uplift, at the
upstream face is less than that required by paragraph 4.4.2, will require
reinforcing steel. The concrete and reinforcing steel stresses for such
cases have been evaluated based on concrete working stress methods with all
concrete tensile capacity neglected. Shear-friction factors of safety were
calculated based on net uncracked areas. It should be noted that the
effective stress, including uplift, may be tensile without assuming
cracking provided that the tension is caused solely by the uplift pressure.
The spillway agee section was evaluated at base elevations of El 1124,
1135, 1150 and 1160, and for lift lines at El 1140, 1150, 1160 and 1170.
Due to tension on the upstream face at El 1170 in the Usual Condition,
evaluations of additional lift lines at El 1165 and El 1175 were added.
The remaining lift lines were evaluated by interpolation.
Non-overflow sections at the left spillway abutment were evaluated with
base elevations at El 1160 and El 1180. The non-overflow section at the
right spillway abutment is keyed into rock along its base and sloping east
face, with most of its 15 foot base width keyed horizontally into rock and
a minimum of 3 feet keyed horizontally into rock above the base.
3777R/205R/LS 6-1
A summary of sliding factors of safety and maximum effective tension and
compression stresses within the agee section is given in Table 6-1.
Detailed information for each case and elevation, with agee base El 1124,
is provided in Figures 7 to 9.
TABLE 6-1
STATIC RESULTS-OGEE SECTION
Case I Case II Case IV
Usual Unusual Unusual
Shear-Friction F.S.
Min Calculated 5.2 13.6 61.0
Allowable 3 2 2
Concrete Compression (psi):
Max Calculated 45 34 48
Allowable 1000 1500 1500
Concrete Tension, Incl. uplift (psi):
Max Calculated 3.6* (No Tension) (No Tension)
Allowable 60 90 90
Rock compression (psi):
Max Calculated 32 34 48
Allowable 140 185 185
'" Maximum tensile stress due to uplift without cracking assumed is 3. 6
psi. Maximum tensile stress including ice load and uplift is 7. 3 psi,
neglecting reinforcement. Section is reinforced in all tensile zones.
3777R/205R/LS 6-3
Cohesion and shear-friction along the rock cut will assist in resisting
overturning in the north-south direction. Based on these facts and
evaluation of the non-overflow section with base El 1160, the right
non-overflow section was deemed acceptable for stability in the north-south
direction, but required further evaluation for seismic loading in the
east-west direction.
6.2 RESULTS
The static analysis indicates that the ice loading in the Usual Condition
results in tension (5 psi) on the upstream face of the spillway agee
section at El 1170 and El 1175. However, stresses for the Usual Condition
without ice load or uplift remained compressive. Reinforcing was added to
control cracking due to the ice load, with very low resulting steel
stresses (f = 3. 4 ksi). Since some reinforcing was being added in the s
upstream face near the crest due to tension, it was extended down the
upstream face to the base elevation and over the crest to the point of
inflection to limit thermal and shrinkage cracking and to improve overall
stability.
It should also be noted that if the reservoir level were at El 1178 with
the corresponding iceload applied at El ll77. the static analysis for the
Usual Condition, without uplift, indicates no concrete tension or
cracking. As ice will occur primarily during winter months when the
reservoir is lower, the potential for cracking due to ice load is very
small.
All remaining cases and levels were found to meet the stability criteria
without additional reinforcing. The effective stress, including uplift,
for El ll60 and El ll65 in the normal reservoir case was tensile, but
these tensile stresses were due to uplift so cracking was not assumed.
With the exception of those levels requiring reinforcing due to ice load
(El 1170 and El ll75), the resultant for all Usual and Unusual Conditions
(Case I, II, and IV) is located within the middle third of the section. No
tension was indicated at the rock-concrete interface for any of the static
analyses.
3777R/205R/LS 6-2
The non-overflow sections analyzed with bases at El 1160 and El 1180 were
found to be statically stable in all cases. For Cases I & II the stresses
at El 1160 without uplift were always compressive at 23 psi to 40 psi. The
minimum shear-friction factor of safety for these cases was in excess of
22. Case IV indicated stresses of 1 psi to 72 psi and a shear-friction
factor of safety over 56.
The right non-overflow section was evaluated for seismic stability in the
east-west direction using pseudostatically applied accelerations at 0.35 g
and 0.75 g. In order to improve the stability of the right non-overflow
section, the concrete section will be tied back into the rock abutment
using rock bolts.
3777R/205R/LS 6-4
7.0 FINITE ELEMENT ANALYSIS
7.1 STRESS ANALYSIS
Required inputs for the finite element analysis are as follows:
(1) Physical geometry of the ogee sections.
(2) Elastic modulus and Poisson's ratio of concrete.
(3) Deformation modulus and Poisson's ratio of rock.
(4) Density of concrete.
(5) Hydrostatic loads, corresponding to reservoir EL 1180.
(6) Ice thrust force, equal to 12000 lb/lin ft at EL 1179.
(7) Ground response spectra.
(8) Uplift and seepage forces. (Combined with finite element results
by superposition.)
In addition, to account for the hydrodynamic effects of reservoir water
during the earthquake, the Westergaard added mass approach was used to
calculate the additional mass (Ref. 7 and 8).
Two types of concrete m1x were selected for the spillway construction. The
mass concrete core will be constructed with a specified concrete
compressive strength of 3000 psi and the outer 3 foot shell will be
constructed with a specified concrete compressive strength of 4000 psi.
The spillway was evaluated with properties for 3000 psi concrete. The
moduli of elasticity and Poisson's ratio used in the finite element
analysis are given in Table 7-1.
3777R/205R/LS 7-1
TABLE 7-1
ASSUMED CONCRETE PROPERTIES
f' = 3000 psi c
Modulus of Elasticity:
(psi)
Static
Dynamic
Poisson's Ratio 0.2
The assumed rock foundation deformation modulus was taken as 4xl0 6 psi.
Poisson's ratio for the rock was taken as 0.27 for static conditions and
0.35 for dynamic conditions. These values were based on Beiniawski's Rock
Mass rating method for moderately fractured rock.
The spillway was designed for an earthquake with the response spectrum
shown on Fig. 4, Mean Horizontal Response Spectrum, with a normalized peak
acceleration of 0. 75g and 5 percent damping. Vertical earthquake ground
motions were assumed equal to 2/3 the horizontal motions.
The spillway may be considered to have the so-called "plane strain" state
of stress. Therefore, a finite element model with two-dimensional elements
was adopted for this analysis. These elements can be either quadrilateral
or triangular in shape.
To account for the contribution from the stiffness of the rock foundation,
the foundation was included in the model. The model included a foundation
with its depth equal to the height of the concrete spillway and extending a
distance equal to the height of the spillway upstream and downstream from
the spillway.
The seismic stress was obtained by response spectra modal analysis. In
each direction of earthquake, the contribution from each mode was combined
by square root of the sum of the squares (SRSS) of all modes considered.
The results from each direction (vertical and horizontal) were then
combined by vector sum.
3777R/205R/LS 7-2
Three finite element models were prepared. Models were for ogee sect ions
at base El 1160, 1150, and 1124 (Figures 10,11, and 12, respectively).
7.2 RESULTS
The stresses for the ogee sect ions for Cases III and V are presented in
Figures 13 to 24. All stresses at each of the sections of spillway
analyzed under the extreme loading conditions were within the allowable
stresses based on 3000 psi concrete. Uplift pressures were combined by
superposition with the computer analysis results for Case III to obtain
maximum concrete tensions. For Case V the section was analyzed without
hydrostatic loads but for simplicity included the Westergaard. added mass in
the seismic analysis. This approach resulted in slightly conservative
seismic stresses but well within allowable values. A summary of maximum
stresses for Case III and Case V is given in Table 7-2.
The shear-friction factor of safety was not calculated using the finite
element method. Sliding stability was evaluated using the Sarma method of
analysis.
TABLE 7-2
FINITE ELEMENT RESULTS
CASE III CASE V
Base at: El ll24 El ll50 El ll60 El ll24 El 1150 El ll60
Concrete Compression (psi)
Max Calculated 155.7 77.9 52.3 165.5 86.3 48.8
Allowable 3000 3000 3000 3000 3000 3000
Concrete Tension (psi)
Max Calculated: W/o uplift 87.3 32.1 45.4 77.5 18.3 22.5
Incl. uplift 111.6 45.1 54.1
Allowable 270 270 270 270 270 270
Rock Compression (psi)
Max Calculated 155.7 77.9 52.3 165.5 86.3 48.8
Allowable 250 250 250 250 250 250
3777R/205R/LS 7-3
8.0 SARMA ANALYSIS
8.1 STABILITY ANALYSIS
The sliding stability of the spillway was evaluated by dynamic analysis.
The analysis utilized the SARMA computer program (Ref. 9) to model the
response of the spillway under earthquake loading. In order to use the
SARMA program the following assumptions were made: the concrete spillway
sections act as an intact failure wedge, the failure plane is the contact
between the spillway sections and the foundation rock or a horizontal
approximation of same, and the spillway sect ions will remain intact. All
downstream rock restraint was ignored in the analysis, resulting in
significant conservatism in the final sliding stability.
The shear wave velocity, the mass density, and the critical accelerations
of the spillway are needed as input to the program. The shear wave
velocity (V ) was calculated using the following relationship between the s
modulus of deformation (G) and the mass density:
V = (G/mass density)112
s
The mass density was calculated assuming the unit weight of the concrete
was 145 lb/ft 3 . The critical acceleration of a wedge section is the
horizontal earthquake acceleration necessary to initiate sliding. The
critical acceleration depends on the assumed loading conditions and is
found by statics.
Six spillway cross sections were evaluated in the analysis, including four
spillway ogee sections and two non-overflow sections. The SARMA program
requires that the section to be analyzed be modeled as a symmetrical
triangle. To model the spillway sections most accurately, the model
triangle was configured to have the same center of gravity, base elevation,
and area as the spillway section being modeled, as shown in Figures 25
to 27.
3777R/205R/LS 8-1
Several loading cases and special conditions were considered. It was
assumed in all cases that the reservoir was at El 1180, as this was more
critical than the low water reservoir condition. In the first case, the
static head was included and the only resisting force considered was the
friction (45 degree friction angle) between the concrete and the rock. The
uplift pressure was assumed to be the full head at the upstream face,
decreasing linearly to 1/2 the head at the drainage gallery, and then
decreasing linearly to zero at the downstream toe. The vertical earthquake
acceleration, 2/3 of the horizontal acceleration, was used in the static
analysis to determine the critical horizontal acceleration. In the second
case, the inertial force of the water created by the earthquake was also
included by using Zangar's formula, (Ref. 6). In the third case, the
inertial force of water was again taken into account as well as 500 psf
cohesion between the rock and the concrete. The worst case providing the
greatest cumulative displacement is the second case. This case was
analyzed for peak horizontal ground accelerations of 0.35g and 0.75g.
8.2 RESULTS
The maximum displacement occurred in
horizontal ground acceleration of 0.75g
eration. The displacement was 0. 5 feet.
the second case, with a peak
combined with vertical accel-
It should be noted that the
inertial force of the water is not usually considered in such a dynamic
analysis and resulted in slight decreases in critical acceleration for all
sections. There was no movement for the construction case earthquake of
O.lg and essentially no movement for the design basis earthquake of 0.3Sg,
with its worst displacement being about 1/100 of a foot with no cohesion
assumed. It is concluded that the movement of the spillway under
earthquake loading will be small and considering the keyed and fixed-edge
plate configuration will likely be zero even in the MCE case.
The amount of intact rock-concrete area needed to force critical wedge
accelerations of at least 0.75g and 0.35g was also calculated. The
required amounts were 2.4% and 0.3% of the surface area respectively in the
worst case based on a rock shear strength of 1500 psi. As intact
shear-capable rock is expected to be in excess of 75% under all sections,
3777R/205R/LS 8-2
the resultant stability is expected to be far in excess of that needed to
prevent movement. Similarly, with no intact rock, but using a contact
shear strength of 160 psi, the percent bonded area to prevent movement
during the MCE and DBE would be 22 and 2.5 percent, repsectively.
The sununary of MCE load displacements for Case 2 is presented 1n Table
8-1. Seismic displacement plots for the four ogee sections, and two
non-overflow sections for the MCE case are presented in Figures 28-33.
TABLE 8-1
SARMA RESULTS
Case 2: 0 = 450, no cohesion, Zangar's water force
Critical Max Gnd Displacement
Base El. Section Acceleration Acceleration (ft)
1160 Ogee 0.231 0.75g 0.51
1150 Ogee 0.282 0.75g 0.32
1130 Ogee 0.258 0.75g 0.38
1124 Ogee 0.263 0.75g 0.37
1160 Non-overflow 0.443 0.75g 0.06
1124 Non-Overflow 0.334 0.75g 0.20
3777R/205R/LS 8-3
9.0 CONCLUSIONS
9.1 CRITICAL CASES
Static loading Case I requires the addition of reinforcing steel on the
upstream face near the ogee crest to prevent cracking due to the ice
loads. All other static loading conditions, including the Usual Condition
without ice loads considered, meet the stability criteria without cracking.
The dynamic analyses for the seismic loading conditions indicate that the
structure is stable under the Extreme Loading Conditions. The finite
element analyses indicate that the concrete stresses due to the Extreme
Loading Conditions are within acceptable limits. The Sarma analyses
indicate that the potential spillway deformation is not extreme under the
earthquake condition even assuming no cohesion or intact rock at the
spillway base. Consequently, the spillway stability is considered to be
acceptable under the seismic loading for the given Maximum Credible
Earthquake and all lesser events.
9.2 SUMMARY OF STABILITY CONDITIONS
The calculated stresses and factors of safety for the bases of the analyzed
structures are summarized on Figures 34-36. The maximum calculated tensile
stress is 87 psi and the maximum calculated compressive stress is 166 psi,
well within allowable stresses as specified in Table 4-1. The minimum
calculated shear-friction factor of safety for Cases I, II, or IV is 5.2,
also with a margin of safety to minimum requirements. The maximum
calculated MCE permanent mass displacement, assuming a hypothetical
continuous and cohesionless failure plane, is small enough (6 inches) that
it would not result in breaching of the spillway. With the anticipated
rock conditions and specified foundation surface preparation, the MCE is
expected to result in no measurable permanent base displacement.
Consequently, the spillway is considered stable under all given loading
conditions.
377R/205R/LS 9-1
10.0 BIBLIOGRAPHY
1. Design of Gravity Dams, United States Department of the Interior,
Bureau of Reclamation, 1976.
2. Woodward-Clyde Consultants, "Report on the Bradley Lake Hydroelectric
Project, Design Earthquake Study," submitted to Alaska District,
Corps of Engineers, Nov. 10, 1981.
3. Woodward-Clyde Consultants, "Seismicity Study, Bradley Lake
Hydroelectric Project," submitted to Alaska District, Corps of
Engineers, March 28, 1980.
4. Stability Criteria of Existing Concrete Gravity Dams, FERC Guideline,
dated Nov. 7, 1985.
5. STARDYNE program, by R. Rosen, Mechanical Research Inc., 9841 Airport
Boulevard, Los Angeles, CA.
6. Design of Small Dams, United States Department of the Interior,
Bureau of Reclamation, (1977).
7. Water pressure on Dams During Earthquakes, by H.M. Westergaard,
Transactions American Society of Civil Engineers, Vol. 98, pg. 418,
1933.
8. Finite Element Methods in Analysis and Design of Dams, International
Commission on Large Dams (ICOLD), Bulletin 30, January 1978.
9. Seismic Amplification Response by Modal Analysis, "SARMA," SWEC
Program GT-055 Version 01, Level 00, September 1986.
10. Effects of Earthquake on Dams and Embankments, Newmark, N. M., Fifth
Rankine Lecture in Geotechnique Vol. XV No. 2, Institution of Civil
Engineers, 1965.
3777R/205R/LS 10-1
11. Engineering Guidelines for the Evaluation of Hydropower Projects,
FERC 0119-1, Federal Energy Regulatory Commission, Office of
Hydropower Licensing , July 1987.
12. Stone & Webster Engineering Corp., "1987 Geotechnical Exploration
Program, Bradley Lake Hydroelectric Project", February 1988
13. Stone & Webster Engineering Corp. , "Geotechnical Interpretive Report,
Bradley Lake Hydroelectric Project", Addendum No. 1, March 1, 1988.
3777R/20SR/LS 10-2
~
KACHEMAK BAY
MUD
FLAT
~~ ~ EG3 _-tfF
l/
ELEVATIONS SHOWN ARE BASED ON PROJECT LAYO,UT
PROJECT DATUM.
MEAN SEA LEVEL DATUM • PROJECT
DATUM PLUS 4.02 FT. MAP t' ~'-,.,__) ( s ~~£ . ,_ ' .:'..::': '"" -FIGURE 1
s=~=r---= ::::: . ...----------.. r----loeo
' \ \ \
\ \ \ ~wg.x ROCK
, \ I I
\ \ I I
\I\\\ ,,,, '',
'~'~'->-''"'' '~" '\ ,-..,_ --:::: .... ,, ... _'.: .... ,', ,~, ,,,, ,,, '\
,\\ \ ' \ \ \ \ \ \ ,,,\1
1\\ \ ,,,
i\ \ \ 51 I \I
\ \ '11, \ II I ·~ \ • I! t\
;-,\ \I loll
.\ 'l?tl ,, '
1 II
'
i
ACCESS ROAD7
I
/
/~ASTE DISPOSAL
AREA B
. LL 1100.0'MAX
~-?;-0.0
00 0
(
WASTE
DISPOSAL AREA ~h
I!L 1090.0' MAX/
( ...... '
l
I
\
\
\
',',,, ,,
'
' \
~
\
I
I
\
\
\
\
'b
I
\ I \ r, I \1 ,.,...
I I ', / ---
' ' MAXIMUM NORMAL I 1
OPERATING WATER • C3~1 1 ~RF.OCE EL 1\80 O' I 1
I I
\ \ \
t
WASTE DISI'OSAL}-l I ~ AREA F 1 1
EL 10900' MAX , / \~ ~ / I \ / /
\ I I
I I
I 1
I \ I \
I \ \ \ I I I I
,.-.,06 \ :
\ I I I $,_, \
', \
\
I
........ ,.,,:, .... ·,, '~ '... '' ' ' '', ... , '':..',',, '-. ',, ', ', ', ',,
",,'' '-.., ', ',,'-, ''o '
.... , ', ', ', '', ~o ....
',,', ', ', 't> ', '',, "'-,', ' ... _: .... ~.... .............. --
' ' -h', ', ......... ,,~, ........ ', " ....
... ' .... ' .... '-s_\~~~~~~-:__
~
-------
SYMBOLS KEY
G SURIIEY MONUMENT
• SEISMOGRAPH INSTALLATION
@ WCRK POINT
troo----_
GENERAL ARRANGEMENT
MAIN DAM AREA
FIGURE 2
~ /,/ct>" _.. ..-/ , / / I ~" -/~~ / / -./ / I I I I I /{-/ / " f// I / /-/ I -/ / \\\ }/// ,.\' // ----/ / I I / I I/" / / // /
" ~~'"--1 I(// ,/ ,...-;_---:::_~,... / _.... / //I//\./ / /..-,.--
"-\"i-.\~:::::_-....~1/!/ /P' /~,...,..-~---//:,..-"'..-_.///I l( I I \ \ '{~\~\:-'...--/; // (;' / ,...",t ,..._.... # 1 f I , ! I _;
\ ';.~ ~~'---/ / / / 1 1 I g 1 / .... ~" I I 1 ) y I /--
\
\
I
J
I
\ 'b~\\"-,..,..,..( ( -~/ \ 1 / I I 1 I ;l\.1 1
\ '~~~-) \ \ r\ ,/r~YMOf'U.£~\ \ I / _../ I I
' \. \ ' -I J ' / ' I I I P~,...BBB\ J I
' I
HANOR
LOCA liON OF ROCK I
CONCRETE INTERFACE
WILL VARY WITH ROCK
TOPOGRAPHY
1
I
I
\
\
\
I
0
0
"'
!
l
\
I IJ.1
I I I 11
Ill /1
)/It!
/t/!1
,../If//
J
MONUMENT
I
~
\
\ \
\ \
l -t.L "Jq.>~v [ ~ , , \ \ \ INTERMEDlAT~ '-... '-\
I t"\ \ TRAINING WALL----~ '-\ \ I \ \ ~I W£S T TRAiNING "' \
I rt! ~ WALL ' " [;..~tL Tr11NING \ \ ' "l I ' \\ \ \ . " II \\ , I \\\\ \ '\
J J (\ \ \ ~'-",, , __ -.. "', "-, \\," "-, ....... , \ ' '' -----"'"' "" ,_ --1 -/ / J J I \ '-'-'--...... -...._ '\ "\ -....... "'-----, / /:/
// J \\ ........... -------\''"' ---//// / I '-... --....... ""----::-' \ ' \ ' __.-·-///,/' "/ (\"' -"' --...-.. ' ,,___ .,.-_ ,/,.-
!"_., / " --....... ------' " "1170·----::.;://,:;: / / \ \ ""' ( '----, "'-~, -........ , ',-,,So'\ ", --.:::::::::: __ .......... ~~ //;:/':/if \ \\'\\ \ r----.. '~"'"-"'-'----""' --....--:::._,_,":.__,~60 -=----/////I !1 II;!
l
\ ' "-' ""''--...._, -....._ ' -------...-// I 111/11 1 \ 1 ' -...._ -----"" ', --/ / / _ __,nr;::;"i
J I l !TOR "" "---=:--...., -,....._ --.,,~_ ',._:--_ _:::---I;!
1/ :/1/
.::-1/
I jl;
I 1/1
I I I I / ' ' '--..._, -----/1/1 I I " " '--.... --'//
o/ I f I I ""' '--~''<'o -----·-/ /_/
"' / I / .,..-, ---'...._ // /,· / I / / --.... --I/ 1-" / , / / I ------, ----r:;%: I / jl
NON-OVER!" LOW
f'IX£0 LOlNtF..
E.l.._ll~
... ILAlteN'
>ER SHAf
EL 114
113.4-90' ~
EAST TRAINING WALL
/ 1-/ // / PLAN-SPILLWAY ---no---------
/ / ----
1()!)•_,....
OVERFLOW
52(.6t4
.,,,..------__ ,_ ____ ,
/
~/\DRAINAGE GALLERY
INTERMEDIATE TRAINING WALL
GALLERY
ACCESS
& DOOR
[L_1190,00:_
UIS CREST
",;• 2.62'
Yc' 1.00'
Rl ' 5.66'
R2:: 1.72 1
UPSTREAM
FACE
X
FLOW
'(
!SPILLWAY
1BASELINE
I
Yc
DIS CURVE COORDINAl]S
X I y
PC 1 14.~4' 9.~4'
PI I 19.04' 1$.00'
PT 1 26. 11 ' 15.00'
PC 2 35.4 1' 35.62'
PI 2 42.92' 45.00'
PT 2 !>4.93' 45.00•
DIS CREST COORDINATES
X
0 t.oo·
2.00'
l.oo•
4.00'
~.oo· s.oo•
'~00' a.oo•
9.00'
10.00'
11,00'
12.oo~
13.00'
14.00'
1!1.07'
y
0
0.07'
0.24'
0.52' o.aa•
1.3Jf
1.66'
2,4 7'
J, 161
3.93'
4.76'
5. 70t
6.69'
7.76'
6.90'
10.19'TANGENT
CURVE
INTERSECT
,---------CURvE EQUATION
OVERFLOW SECTION GEOMETRY
NTS
~SPILLWAY BASELINE
LON ::
7-t,-t--· 16
·.o• ---1 HIGH PT
EL ll!il:C 00' ~LOPE I !;L tt95.QII'
f!,OWh
_!;:l__ill!5_,QQ!
FILL CONCRETE
EL VA?::fivk;w:«; 1 f \ I 14-IUPU fll
NON-OVERFLOW SECTION GEOMETRY
NTS
V•0.0678X 1.8<1 6
Et.D OF SPILLWAY A PRClN
(CAST AGAINST ROCK!
0 10 40FEET
------:-1
ELEVATION-LOOKING UPSTREAM SCALE A: 1•• 20'
GENERAL ARRANGEMENT
SPILLWAY
FIGURE 3
REF WOODWARD-CLYDE CONSULT
REPORT' "DESIGN EARTHQUAKE STUD{
NOV 10,1981
~2 25 l-~--~T··-1
Vl 1.881 ~ [ ~FOR HYBRID rARTHOUAKt +-~-+~ ro RESPONSE SPECTRUM
z
0
~ 1 50
0:::. w
_j
w u 1.13
u
<(
~~~ 1 l
BRADLEY LAKE HYDROELECTRIC PROJECT
MEAN RESPONSE SPECTRUM FOR MCE
(NEARBY SHALLOW CRUSTAL FAULT}
r
t I
_j
<( 0.75
0:::.
MEAN RESPONSE SPECTRUM
FOR DBE
1-w
0.. 0.38
(})
000
0.00 0.25
--
050 075 100 25 150 1
PERIOD (SEC)
2 00 2 25
PROJECT RESPONSE SPECTRA
2 50 2
~-------------------------------------------------FIGURE 4
00
MODIFIED ACCELEROGRAM 0£1TAINED FROM THE FOLLOWING TWO ACCELEROGIIAMS
KERN COUNTY EARTHQUAKE 7-21-52
IIA004 TAFT LINCOLN SCHOOL TUNNEl, COMP S69E
SCALE FACTOR = 3.50
FRIULI. ITALY EARTHQUAKE 9·15 · 7ti
AND 1·3-16!) ITALY SAN ROCCO, COMP EW
SCALE FACTOR = 3.18
0.75r------------------------------------------------------------------------------------------------,
0.50
,...
Cl .....,
z
0 0.25
i=
c(
a: w
..J
~ 0.00
0
c(
a z
::l -0.25 0 a:
(!)
-0.50
0.00 SEC TO 2.32 SEC OF MODIFIED • 0.00 SEC TO 2.32 SEC 'OF KERN CO.
2.34 SEC T0·4.30 SEC OF MODIFIED • 2.14 SEC TO 4.10 SEC OF FRIULI
4.32 SEC TO 55.14 SEC OF MODIFIED • 3.58 SEC TO 54.40 SEC OF KERN CO.
THIS PLOT LIMITED TO FIRST 48.0 SECONDS OF THE MODIFIED.ACCELEROGRAM
-0.75L-----~-------L------~----~~----~-------L------~------L-----~------~------~----~
0.00 4.00 8.00 12.00 16.00 20.00 24.00 26.00 32.00 36.00 40.00 44.00 46.00
TIME (sec)
HYBRID ACCELERGRAM
FIGURE 5-------~
NEGLECT
1. oo"
CREST
CREST EL 1180
10.19'
---EL 1165
NEGLECT APRON
AT EL 1150 a 1160
4~0 11 RAD
ALLERY
G_DRAINS \.
~~---N----------\
51.7'
' \
\.
\.
SPILLWAY SECTION
NEGLECT UPLIFT
ON APRON
STATIC
SPILLWAY. MODEL
ASSUMED
GEOMETRY
NEGLECT
APRON
EL 1135
EL 1130
EL 1124
...._--------------FIGURE 6
CASE I-NORMAL RESERVOIR
RESULTANT PRESSURES INCLUDING UPLIFT
(psi)
\7 WS EL 1180
EL 1175
fs = 3.4ksi
EL 1170
F5 = 1.4ksi
EL 1165
3.6 psi
TENSION
EL 1160
1.1 psi
TENSION
EL 1150
3.2 psi
EL 1140
I!Spsi
16 psi
NOTE:
GALLERY SLAB ISOLATED FROM STRUCTURE
SO WILL NOT PROVIDE RESISTANCE. SLAB
DEBONDED FROM ROCK SO ACTUAL UPUFT
WILL BE NEGLIBILE (TYP).
STATIC ANALYSIS
BASE EL 1124
..__---------------FIGURE 7 __ __,
CASE II-PMF
RESULTANT PRESSURES INCLUDING UPLIFT
(psi)
\] WS EL 1191
EL 1180
0.3 psi
EL 1170
0.9pai
EL 1160
I. 4 psi
EL 1150
3.4 psi
EL 1140
15 psi
EL 1135
II psi
I
} CREST
I
STATIC ANALYSIS
BASE EL 1124
......_---------------FIGURE 8
CASE Til-CONSTRUCTION
( 0.1 g HORIZ)
RESULTANT PRESSURES
(psi)
CREST
EL 1180
EL 1170
9.6 psi
EL 1160
18.5 psi
EL 1150
27.1 psi
EL 1140
4!5 psi
EL 1135
48psi
GROUND
ACCELERATION
STATIC ANALYSIS
BASE EL 1124
..._---------------FIGURE 9
EL 1140
EL 1180 ......
I l r..........
11 I'...
.I 1 l ~
__ ___.,.:..:CO:..:..:N:.;.;CR.;.;;;;ETE 1 A ...... ,.,. / ~
ROCK I J T / 't-,_,r--it--~-\__.;tt-~~...-.,._..,.....::E..!:::..L ..!..!.11!:.::63~
EL 1160 l f ] .I
I~ 20' 39. 45' I 20' .....
FINITE ELEMENT MODEL
BASE EL 1160
~I
........ --------------FIGURE I 0
EL 1141
EL 1120
30' -
EL 1180
J
1
1
]]
If
J 7":. / EL II 57
42.25 1 -
FINITE ELEMENT MODEL
BASEEL1150
30'
'----------------FIGURE II
EL 1068
I~
--=..E L::......:..:...ll .:::..::80:.___, r,._,
56'
J ""'\
1 L'....
.. ,. 79 1
FINITE ELEMENT MODEL
BASE EL 1124
56'
.. J
....._--------------FIGURE 12
CASE ill-EARTHQUAKE
(0.75g HORIZ +0.50g VERT)
MAX VERTICAL TENSILE STRESSES W/0 UPLIFT
(PSI)
+45.4 -17.5
+TENSION
-COMPRESSION
-22.2 -7.8 -t2.1
-23. 8 -10. 8 -3. I +I. 4
EL 1160
FINITE ELEMENT ANALYSIS
BASE EL 1160
...._-------------FIGURE 13
CASE ill-EARTHQUAKE
( 0.75 g HOR IZ + 0. 50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES W/0 UPLIFT
CPS!)
-36.8 -16.6
-6.6 -:52.3 -44.4 -22.2 -17.!
+TENSION
-COMPRESSION
~---._~~E~L~tt~so~---------_.---------~----~--~--~
FINITE ELEMENT ANALYSIS
BASE EL 1160
---------------FIGURE 14
-14.0
CASEY -EARTHQUAKE
( 0.75g HOR I Z + 0.50g VERT)
MAX VERTICAL TENSILE STRESSES
(PSI)
CREST EL 1180
+TENSION
-COMPRESSION
+0.8
-3.4 +5.8
-16.5 -4.9 +1.2 +3.9
EL 1160
FINITE ELEMENT ANALYSIS
BASE EL 1160
...._-------------FIGURE 15
CASEY -EARTHQUAKE
(0.75g HORIZ +0.50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES
(PSI)
-29.5 -48.4
CREST EL 1180
+TENSION
-COMPRESSION
-II. 8
-12.2
-37.1 -16.3 -13.0 -10.1
EL 1160
FINITE ELEMENT ANALYSIS
BASE EL 1160
..._-------------FIGURE 16
CASElli -EARTHQUAKE
{0.75g HORIZ + 0.50g VERT)
MAX VERTICAL TENSILE STRESSES W/0 UPLIFT
(PSI)
CREST EL 1180
+ TENSION
-COMPRESSION
1"17.8 +17.3
-1.8 -5.1
-0.3 -6.8 -3.9
-0.3 -6.6 -7.4
-2.2 -6.0
-5.0 -10.2 -7.9
-5.9 -11.6 -10.3
-+6.3 -9.8 -12. B -12.3
EL 1150
FINITE ELEMENT ANALYSIS
BASE EL 1150
+2. 3 -t-1.0
-4.5 -+6.0
-6.6 -I. I
-9.3
...__--------------FIGURE 17
CASElli -EARTHQUAKE
(0.75g HORIZ + 0.50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES W/0 UPLIFT
(PSI)
-59.0 -77.9
CREST EL 1180
+ TENSION
-COMPRESSION
-10.2 -11.7
-13.5 -13.0
-15.7 -15.8
-25.4 -16.6 -21.4
-22.8 -20.9
-25.2 -22.1
-66.1 -34.6 -29.2 -25.7
EL 1150
FINITE ELEMENT ANALYSIS
BASE EL 1150
-9.8
-20.9
-14.9 -38.6
-21.4 -21.9
-25.3 -31.1
------------~--------------FIGURE IS
CASEY -EARTHQUAKE
{0.75g HORIZ + 0.50g VERT)
MAX VERTICAL TENSILE STRESSES
(PSI)
CREST EL. 1180
+ TENSION
-COMPRESSION
+0.2 +_9.1
+0.2 -2.4
-0.4 -:3. B -3.2
-1.7 -5.0
-7.9 -4.3
-9.4 -7.4
-1.5 -10.2 -II. 2 -9.6
EL. 1150
FINITE ELEMENT ANALYSIS
BASE EL 1150
+2.9
-2.0 +14.1
-3.5 +2.9
-5.8 +5.5
---------------FIGURE 19
-63.1 -84.3
CASEY -EARTHQUAKE
(0.75g HORIZ + 0.50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES
. (PSI)
CREST EL 118 0
-2.4
+ TENSION
-COMPRESSION
-8.2 -6.5
-13.0 .-8.6
-15.8 -13.0
-30.4 -16.1 -20.4
-20.5 -17.3
-37.0 -23.0 -19.2
-35.0 -27.6 -23.0
EL 1150
FINITE ELEMENT ANALYSIS
BASE EL 1150
-15.8 -7.9
-12.4 -30.5
-18.3 -17.9
-21.8 -25.9
~-------------FIGURE 20
CASE ill -EARTHQUAKE
(0.75g HORIZ + 0.50g VERT)
MAX VERTICAL TENSILE STRESSES W/0 UPLIFT
. (PSI)
-0.2 + TENSION
-COMPRESSION
+ 3.9 +0.2
t5.8 -7.5 +13.6
+7.1 -10.3 +2.9
+3.8 -11.8 -6.0 +9.7
-12.5 +1.6 +23.0
-17.0 -5.9 +8.2 +21.1
-14.4 -2.1.1 -13.1 -1.6 +8.7
FINITE ELEMENT ANALYSIS
BASE EL 1124
+17.2.
+9.5
.__-------------FIGURE 21
EL 1124
CASE ill -EARTHQUAKE
(0.75g HORJZ + 0.50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES W/0 UPLIFT
(PSI)
CREST EL II 8 0
J. -5.6 ~-14.8
f-, •• -15.4 -2~ + TENSION
-26.B~ -COMPRESSION
. -40.1 -23.9
1-49.2
-735-t -57.7 -32.0 -24.1 -49.4 ~
tra -69.0 -39.3 -27.1 -43.3 -68.1 ~
I -77.6 -41.6 -33.4 -40.4 -54.7 -120.1__......~
hl5.6 ---
/-126 1 ~t&~J_--46-.2-l.---4-0._7 .l---55-.o-l---7-0 ._6--l,.-i-l_l._a --r-_ 1 _,.2 . 5 ~35., ~,.., -53-3 -45.6 -54.9 -··. 2 - 6 3. • _ 3... _, o.
0
/
1---,5-5-. 1-~-1---s 5--.~~ -5-5-e6-.. ooj--6.;-3.-5+---6-0.-2 +---50-.7-+--5-7.-5+---s-o-. o-+-_-6-4.-r -+---4-B-.3--f......_-~
L----L---v~ \/'--~t-...1.----L--.L....----L----1-----~,.;..--...L..-----EL 1124
FINITE ELEMENT ANALYSIS
BASE EL 1124
~--------------~----------FIGURE22
CASEY. -EARTHQUAKE
(0.75g HORIZ +0.50g VERT)
MAX VERTICAL TENSILE STRESSES
(PSI)
+1.6 + TENSION
-COMPRESSION
+3. 6 +5.9
FINITE ELEMENT ANALYSIS
BASE EL 1124
~--------------------------FIGURE23
CASE --sL -EARTHQUAKE
(0.75g HORIZ ·+ 0.50g VERT)
MAX VERTICAL COMPRESSIVE STRESSES
(PSI)
-13.6 + TENSION
-COMPRESSION
-48.6 -24.2 -21.1
-32.9 -20.2 -41.7
-74.9 -40.5 -24.4 -37.5
-37.5 -48.8 -61.3
-52.6 -42.2 49.2 -58.3 -55.6
-60.4 -47.1 -52.0 -57.8 -56.6 -42. 9
FINITE ELEMENT ANALYSIS
BASE EL 1124
....... -------------FIGURE 24
CREST EL I I 80 CENTER OF GRAVITY
ACTUAL OGEE--------~~
ASSUMED MODEL OF EQUAL
MASS AND CENTER OF
GRAVITY. GEOMETRY
EL 1160
BASE EL 1160
CREST EL II 80.0
ACTUAL OGEE~-,'
GEOMETRY ;r' ASSUMED MODEL
I 38.8' ~
BASE EL 1150
SARMA ANALYSIS MODEL
OGEE SECTIONS
SHEET I
EL 1150.0
'------------------FIGURE 25
CREST EL 1180.0
ACTUAL OGEE--ASSUMED MODEL
GEOMETRY
CREST EL 1180
ACTUAL OGEE~,
GEOMETRY ;'
I
19.1 1
61.8 1
BASE EL 1130
.---ASSUMED MODEL
69.01
BASE EL 1124
SARMA ANALYSIS MODEL
OGEE SECTIONS
SHEET 2
EL 1130
..._--------------FIGURE 26
ASSUMED~
MODEL ""'
EL 1195.0
ACTUAL~r--NONOVERFLOW :
GEOMETRY
I
I
35.5'
ACTUAL~ NON OVERFLOW
GEOMETRY ,-EL 1195.0
BASE EL 1160
LEFT ABUTMENT
1
I
' ' \
CENTER OF ',
GRAVITY
29.3'
BASE EL 1124
RIGHT ABUTMENT
ASSUMED MODEL OF
EQUAL MASS AND
CENTER OF GRAVITY
EL 1145.0
EL1124.0
SARMA ANALYSIS MODEL
NON-OVERFLOW SECTIONS
EL 1168.0
EL 1160.0
~-----------------------------+IGURE 27
TIME HISTORY OF WEDGE MOTIONS
o.oo 5-00 10.00 !5.00 20.00 25.00 30.00 35.00 40.00 45.00 so.oo 55.00 60.00
0
UJ~ do
-----------------------------------------------~
UJo zo o,; ~INIIliUI ......
1-
a:
a::: a w.,..
_J • wo ul
u a:
0
(0
";'-INDICRT!S ERATHVUAK[• HYD~ID ACC~LfKDORAn • K(ftN CO. TAfT '"9[1 4 fMIULI ITALY SAN MOCCO lfWl
INDICATES CftiTICAl ACCELERATION, 0.231 0'5
0 1-w w. wo
I.J_
UJO .... 1-• zo
w
l::
w
CLo
UJ
0
INDICATES WEDGE 016PLACfftfHTS 4 RCC!L!RRTIOIIS.
SARMA ANALYSIS
MCE CASE W/0
BASE EL 1160-OGEE
FOUNDATION COHESION
DRTR ARE CDfffiCifNT$ Of 0--2151 fOIIITS •!" • 0.020 5£CONOS
.oo 45.00 so.oo ss.oo
0 ....
0
0
0
0
0 ....
0
I
0
CD
0
I
0
<D
0
0 ....
0
0
N
0
FIGURE 28
TIME-HISTORY OF WEDGE MOTIONS
o.oo 5.00 55.00 60.00
0
(/)~ do
C>
([)
0
------------------------------------~-----------------------------------------------~
0 ...
0
lliflNIIIP qp,A M~liV' 1-\&uY\L::/i'I.A,# \f'W.'WV•'vui_..·.M.,•~ 1 ~ 0
-------------------------U< 0 ...
I INOICATU fARTHQUAM£1 HYBRID IICCfl!ROOIIRn 1 kfRH CO. TAFT 1519!1 ' FUUU ITALY IAN ROCCO IfNI DATA RRf COffr!CifNU Of G--Z157 tOINT& •T • 0.020 5fCOIIDS
1-0 ("}
W• wo
lL
(1)0
~---~ z.o
w
:c w
u a:o
.....J":
(LC>
(f) ......
0
----IHOICATU CRITICAl ACCflfftATION. Q.t82 11'8
IHOICRTU WfDOf OISrUC!"fHU ' ACCflURTIONI.
r---
J
0
I")
0
0
N
0
0
C>
0 0
0 C> . .
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 °
TIME SECONDS
SARMA ANALYSIS: BASE EL 1150-0GEE
MCE CASE W/0 FOUNDATION COHESION
FIGURE 29
o.oo
D
w"' . .
s.oo
TIME-HISTORY OF WEDGE MOTIONS
-----------------------------------------------------------
.oo
D ...
D
D
--·-~-~-~------------ --------- ------------------------~ 0
D
a:>
'7---INOICATU fiiUHGURKfl HYIKIO ACtflfROOilA" 1 KfU CO. TAH IU&fl ~ fRIUll ITALY 8AN ROCCO IfNI
----INOICATU UHICRL ACCfLUAIION. o.t51 0'6
D
f-to w· wo
lL..
U)D ._":
zo
w
:c w u a:D
_j~
CL.D
U)
D
INDICRTU NfDO£ oUrLACfnfNT5 ~ RCCflf~IITION&.
r-
DATA Allf COfff!CifNU Of 0--2151 rOINT5 •T • 0.020 UCONO&
...
D
to
D
0 ...
D
D
N
D
Dj D D D . .
~.oo s.oo to.oo ts.oo 2o.oo 25.oo 3o.oo 35.oo 4o.oo 4s.oo so.oo ss.oo Go.oo o
TIME -SECONDS
SARMA ANALYSIS: BASE EL 1130-0GEE
MCE CASE W/0 FOUNDATION COHESION
FIGURE 30
'
TIME-HISTORY OF WEDGE MOTIONS
o.oo 5.00 10.00 15.00 20.00 25.00 30.00 35.00
0
rn"" ~c;;
1-0 ...,
W• wo
lJ.._
(f)o
~--~ zo
w
:I:
w
u a:o
...J":
Q...o
(f) ,_,
Cl
IHOICATr& rAUifQUIIKrr I!YUIO Rtt£LODOftA" 1 KCI!H to. TAfT ISU£1 < F!IULI ITALY SAN I!OCCD lfWl
IHDICATU CKITICAL IICC£LEI!ATIIIH. 0.261 0'6
--lNDICATU W£01)[ lll&tLIICE"[HI& < lltt[LEIIIHIONS •
ANALYSIS: BASE EL 1124-0GEE SARMA
MCE CASE W/0 FOUNDATION COHESION
40.00 45.00
u c -
-~ -u,
DATA Ill!£ CIICFFICIEHTS 0~ 0--t751 ,OINIS •I = 0.020 5CCOHD6
C) ...
0
I
0
C)
0
I~
C) .,
0
0
N
C)
C)
C)
FIGURE 31
TIME-HISTORY OF WEDGE MOTIONS
0 .oo 5 .oo
0
U)"" -------------------------------------------------------------------------------------~ j~ d~
U)o zo
0~
>---<
1--
a:
0:::: wo _J~
wo
~I 1------a----------~------------------------------------------------------------------------------------------------U,
U:
0
IX>
--INOICIIT[S ~llftiHOUIIKfl HYIRIO IICCHUOO~~" 1 Kf~N CO. TAfT IU9fl l fUULI ITALY &lllf ~OCCO IfNI DATA liRE CO[ff ICUNTi Of 0 --2151 tO lifT& •T • 0.020 UCIIND5
1--t.P w~
wo
IJ...
({) ....
~---~
w
u a:N
_JD
(Lo
U)
0
0
0
---INOICIITU CRltiC~l RCCflfRIITIDH. Q.U, O"f
·--IHOICRTU NfOOf Dl6rLACfHfNT6 l RCCfL£1RT IONS.
+-----~~-------.-------.--------.-------.-------.r-------.-------.-------.----~.00 s.oo 10.00 15.00 20-00 25.00 30.00 35.00 40.00 45.00 50.00
TIME SECONDS
SARMA ANALYSIS: BASE EL 1160-LEFT ABUTMENT
MCE CASE W/0 FOUNDATION COHESION
55.00
0
0
0
0
0 ...
t.P
0
0
....
0
0
N
0
D
~-----------------------------------------------FIGURE 32
o.oo
0
V'Joq
d.;
V'Ja zo o.; ......
f-a:
Ck:o w ...
_J •
wo
u' u
cc
0 co
0
5.00
TIME-HISTORY OF WEDGE MOTIONS
10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00
~ -------------------------------------------------------------------------------------
-----------------------------------------------------------------------------
I --INDICAJU (llliTHDUAKfl HYBRID ACCflUOIN!Aft 1 MUN tO. TAfT 166$(1 l fUUll ITALY &liN l!OCCO IfNI
----INOICATU UH JCIIL IICCfLURTJON, 0.3U G'&
DATil ARE C0[ffJClfNJ6 Of G --t1$1 POINT& •T : 0.0~0 &fCONDI
f-(D
w":'
wo
lL..
V'J"" f.--: zo
w
::c w u
a:"' -lo o__.;
If.) ......
Cl
--lNOicATU WfOGf DI&PLIICfnfNTI l IICCfLUATJ0118.
0
oq
0
0
0
0
0
oq
0
I
0 co
co
0
N
0
<.0
0
0
o a a a . .
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 °
TIME SECONDS
SARMA ANALYSIS: BASE EL 1124-RIGHT ABUTMENT
MCE CASE W/0 FOUNDATION COHESION
~-----------------------------------------------FIGURE 33
PMF EL 1191'
EL 1179' -
-5' EST DRAW DOWN
WATER SURFACE
CREST EL 1180'
64.3'
EL 1130'
LOADING DIAGRAM
NOTES:
1. Stability analysis based on gravity method.
Static analysis for Cases 1, 2 and 4.
Finite element analysis for Cases 3 and 5.
2. Loads:
D ~ Dead weight of structure at 145 lbs/cu. ft. (concrete).
EH ~ Horizontal inertial force due to earthquake
Ev ~ Vertical inertial force due to earthquake
Hw"' Horizontal hydrostatic force
Vw= Vertical hydrostatic force
I = Ice force at 12 kips/lin ft.
HE Hydrodynamic earthquake force
LJ = Uplift force
Numeral subscript indicates load case
3. Load Cases:
Case 1 -Normal
A. • Dead weight
B. -Hydrostatic forces for normal maximum reservoir
level of El 1180'
C. Ice
D.· Uplift and seepage forces
Case 2 -Probable maximum flood (PMF)
A. -Dead weight
B. -Hydrostatic forces for maximum reservoir level
ofEI1190.6' (rounded up to 1191')
C. · Uplift and seepage forces
Case 3 • Earthquake
A.· Dead weight
B. -Hydrostatic forces for normal maximum reservoir
level of El 1180'
C.-Ice
D. · Earthquake inertial and hydrodynamic forces for
maximum credible earthquake (0.75g horizontal
& 0.5g vertical)
E. Uplift and seepage forces
Case 4 · Construction
A.· Dead weight
B. · 1) Earthquake inertial forces for operational basis
earthquake (0.1 g horizontal) or;
2) Wind
Case 5 • Low reservoir level earthquake
A. • Dead weight
B. -Earthquake inertial forces for maximum credible
earthquake (0.75g horizontal & 0.5g vertical)
4. Base pressures for Case 3 and Case 5 determined by two
dimensional finite element analysis with earthquake inertia
load computed from response spectrum analysis and hydro-
dynamic effects approximated by Westergaard added masses.
5. Uplift pressures assume a drain efficiency of 50% at the base.
6. Uplift assumed to act over 100% of base area.
7. Base pressures for uncracked sections calculated without in-
cluding uplift as an active external force. Uplift pressures were
combined with the resulting base pressures by the superposition
method.
8. Allowable stress in PSI:
Concrete (3000 PSI) Rock (40KSF = 280 PSI)
Tension Compression Compression
Case 1 60 1000 140
Case 2 90 1500 185
Case 3 270 3000 250
Case 4 90 1500 185
Case 5 270 3000 250
9. Sliding factor of safety for Cases 1, 2 and 4 is based on shear
friction factor of safety formula with 160 PSI coheston and
an internal angle of friction of 45 degrees
SPILLWAY STABILITY
ANALYSIS SUMMARY
SHEET I
FIGURE 34
CASE I
NORMAL
CASE 2
PMF
BASE PRESSURE DIAGRAMS
EL 1124
t t t t t t'
..
.. • • t r • •
CASE 4
CONSTRUCTION
CASE RESULTANT KIPS
NUMBER :Ev :EH
1 237 110
2 222 136
4 294 32
BASE EL 1124
BASE PRESSURE ·PSI
X W/UPLIFT W/0 UPLIFt
FT U/S D/S U/S DIS
135.5 10 32 ~4 ~4
34.8 4 34 33 36
42.9 47 11 47 11
SAFETY
FACTOR
SLIDING
17.0
13.6
61.0
CRACK
LENGTH
FEET
0
0
0
STATIC STABILITY RESULTS
BASE PRESSURE DIAGRAMS
EL 1135
•• t t t t j
t_tjt t t j
~·
BASE EL 1135
BASE PRESSURE DIAGRAMS
EL 1150
DOJ
Q[IJ
r--[J7
t)
BASE EL1150
CASE rESUlTANT, KIPS
NUMBER :Ev I 1: H
X
FT
1 I 76 I 40 19.1 I 1 o I 22 I 23 I 22
2 I 67 I 49 20.0 f 9 I 20 I 26 I 20
__ 4 I 95 10 24.91 37 I 5 I 37 I 5
BASE PRESSURE ·PSI SAFETY CRACK
CASE RESULTANT KIPS X W/UPLIFT W/0 UPLIFT FACTOR lENGTH
NUMBER :Ev :EH FT U/S D/S U/S D/S SliDING FEET , 126 75 28.9 16 26 35 28 17.2
2 137 94 28.7 11 26 35 29 13.8 0
4 198 21 35.5 48 8 48 8 64.3 0
SAFETY
FACTOR
SliDING
21.6
17.6
86.6
BASE PRESSURE DIAGRAMS
EL 1160
CRACK
LENGTH
FEET
_o_
0
0
O[J]
om
orv
BASE El1160
BASE PRESSURE ·PSI SAFETY
FACTOR
SLIDING
I. CASE RESUlTANT KIPS X W/UPLIFT W/0 UPLIFT
NUMBER LV :EH FT U/S DIS U/S D/S
1
2
4
31 25 12.3 2 18 11 2_0
22 26 14.7 5 13 18 15
45 5 18.3 23 5 23 5
SPILLWAY STABILITY
ANALYSIS SUMMARY
SHEET 2
23.0
21.0
118
CRACK
lENGTH
FEET
0
0
L---------------------------------------FIGURE 35 -~
FINITE ELEMENT ANALYSIS RESULTS
MAXIMUM STRESSES (PSI)
BASE CASE TENSION TENSION SLIDING
ELEV NUMBER COMPRESSION W/0 UPLIFT WITH UPLIFT STABILITY-tc
1124 III 155.7 87.3 111.6 Stable
v 165.5 77.5 Stable
1150 III 77.9 32.1 45.1 Stable
v 86.3 18.3 Stable
1160 III 52.3 45.4 54.1 Stable
v 48.8 22.5 Stable
*Based on negligible calculated displacements from SARMA analysis.
SPILLWAY STABILITY
ANALYSIS SUMMARY I SHEET 3
... ---------------FIGURE 36 _..,