Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Paleoseismology of blind-thrust faults beneath Los Angeles, California: implications for the potential of system-wide earthquakes to occur in an active fold-and-thrust belt
(USC Thesis Other)
Paleoseismology of blind-thrust faults beneath Los Angeles, California: implications for the potential of system-wide earthquakes to occur in an active fold-and-thrust belt
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
PALEOSEISMOLOGY OF BLIND-THRUST FAULTS BENEATH LOS ANGELES,
CALIFORNIA: IMPLICATIONS FOR THE POTENTIAL OF SYSTEM-WIDE
EARTHQUAKES TO OCCUR IN AN ACTIVE FOLD-AND-THRUST BELT
by
Lorraine Annette Leon
A Dissertation presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)
August 2009
Copyright 2009 Lorraine Annette Leon
ii
ACKNOWLEDGEMENTS
Thanks are due, first and foremost, to Professor James Dolan, my primary advisor
during my PhD at USC. He has guided me, nagged me, and encouraged me throughout
my graduate studies. I am indebted to him for his willingness to take me on as a graduate
student and see the potential in me that I could not see myself. I couldn’t have asked for a
better mentor.
In addition to James a number of other people participated in this dissertation and
deserve to be recognized. My qualifying exam committee of James Dolan, Greg Davis,
Charlie Sammis, Steve Lund, and Iraj Ershaghi challenged and advised me how best to
move forward with my research. My dissertation committee James Dolan, Greg Davis,
Steve Lund, and Iraj Ershaghi found this dissertation acceptable and were willing to sign
off on it. I appreciate their guidance and support throughout this process. Tom Pratt,
John Shaw, and Steve Lund were enthusiastic collaborators and generously shared their
time and expertise with me. The many field assistants, interns, and undergraduates that
helped me during my research were a key part of this dissertation. Many city employees
helped me acquire the permits required to make this research possible. My officemates,
Ozgur Kozaci, Kurt Frankel, Erik Frost, Plamen Ganev, and Ben Haravitch, provided me
with support, and friendship. John McRaney always found a way to make sure that I
could move forward with my research. Cindy Waite, Vardui Ter-Simonian, John Yu,
Barbara Grubb, Karen Young, and Stanley Wright helped my time at USC run smoothly.
James Dolan, Tom Pratt, John Shaw, and Greg Davis each took time to write numerous
letters of recommendation, which in no small way contributed to me getting job offers.
iii
But first and foremost I would never have made it this far without the love and support
from my family. My husband, Carl, and daughter, Gemma, deserve many thanks for
their love and devotion. Thanks to my Mother who has never failed to support me with
patience and love and to my Father who cannot share this milestone with me.
Of course, none of the work presented in this dissertation would have been
possible without the generous support of the National Science Foundation, the Southern
California Earthquake and the University of Southern California Department of Earth
Sciences.
Finally, this dissertation includes parts of the following manuscripts:
Leon, L. A., S. A. Christofferson, J. F. Dolan, J. H. Shaw, and T. L. Pratt, 2007,
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard,
Journal of Geophyical. Research – Solid Earth, v. 112, B03S03, doi:
10.1029/2006JB004461.
Leon, L. A., Dolan, J.F., Shaw, J.H., and Pratt, T.L., 2009, Evidence for large Holocene
Earthquakes on the Compton Thrust Fault, Los Angeles, California, Journal of
Geophyical. Research – Solid Earth, doi: 10.1029/2008JB006129.
Leon, L. A., Dolan, J.F., Shaw, J.H., Pratt, T.L., and Lund, S., in prep, Paleoseismology
of the Los Angeles segment of the Puente Hills blind-thrust fault, Los Angeles,
California: Implications for the potential of system-wide earthquakes to occur in
a segmented blind-thrust system.
Leon, L. A., Dolan, J.F., Shaw, J.H., Pratt, T.L., and Lund, S., in prep, Evidence for
Holocene- Late Pleistocene activity of the Compton blind-thrust fault, Lakewood,
California: Implications for system-wide Earthquakes and seismic hazard.
The co-authors listed above shared data, assisted with field work, and helped to supervise
the published research results.
iv
TABLE OF CONTENTS
Acknowledgements ii
List of Tables viii
List of Figures ix
Abstract xii
Chapter 1: Introduction 1
1.1 Introduction 1
1.2 Active Blind-Thrust Folding and Faulting in the Los Angeles 2
Basin
1.2.1 Puente Hills Blind-Thrust Fault 5
1.2.2 Compton Blind-Thrust Fault 7
1.3 Methods 8
1.3.1 High-Resolution Seismic Reflection Data 8
1.3.2 Borehole Excavations 10
1.3.3 Cone Penetration Testing 12
1.4 Research Implications 13
Chapter 1 References 16
Chapter 2: Earthquake-By-Earthquake Fold Growth Above the Puente 19
Hills Blind Thrust Fault, Los Angeles, California:
Implications for Fold kinematics and Seismic Hazard
Abstract 19
2.1 Introduction 21
2.1.1 Puente Hills thrust 22
2.2 Data 27
2.2.1 Seismic reflection data 27
2.2.2 Borehole Observations 30
2.2.2.1 Stratigraphic observations 37
2.2.2.2 Age Control 42
2.2.2.3 Evidence for Folding and Sedimentary 43
Growth
2.2.3 Structural Reconstructions 47
2.2.3.1 Methods 47
2.2.3.2 Shear Restoration of Folding Events 48
2.2.3.3 Location, Geometry, Width, and 55
Evolution of the Forelimb Kink Band
2.2.3.4 Incremental Development of Dips within 59
the Kink Band
v
2.3 Discussion 60
2.3.1 Earthquake-by-Earthquake Fold Growth 62
2.3.2 Using Paleofold Scarp Geometry to Estimate 64
Paleomagnitudes
2.3.3 Implications for Seismic Hazard in Southern 64
California
2.3.4 Incremental Structural Evolution of Fault-Related 67
Folds
2.4 Conclusions 68
Chapter 2 References 72
Chapter 3: Blind Thrust Paleoseismology of the Los Angeles Segment 76
of the Puente Hills Blind-Thrust Fault, Los Angeles,
California: Implications for the Potential of System-Wide
Earthquakes to Occur in a Segmented Blind-Thrust System
3.1 Introduction 76
3.1.1.Puente Blind-Thrust Fault 77
3.2 Results 80
3.2.1 High-Resolution Seismic Reflection Data 81
3.2.2 Borehole Data 84
3.2.3 Cone Penetration Testing Data 85
3.2.4 Stratigraphic Evidence for Paleo-Folding Events 88
3.2.5 Radiocarbon Dating 92
3.3 Discussion 96
3.3.1 Fault Slip Rate Estimates 96
3.3.2 Earthquake-By-Earthquake Fold Growth 99
3.3.3 Paleomagnitude Estimates and Implications for 100
Fault Behavior
3.3.4 Incremental Structural Evolution of Fault- 101
Related Folds
3.3.5 Implications for Seismic Hazard in Southern 102
California
Chapter 3 References 104
Chapter 4: Holocene and Late Pleistocene Slip Rates on the 107
Santa Fe Springs Segment of the Puente Hills Blind-Thrust
Fault
4.1 Introduction 107
4.1.1 Puente Hills Thrust 107
4.2 Results 108
4.2.1 Seismic Reflection Data 110
4.2.2 Borehole Observations 113
vi
4.3. Discussion 120
4.3.1 Fault Slip Rate Estimates 120
4.3.2 Incremental Development of Dips within the 121
Kink Band
4.4 Conclusions 125
Chapter 4 References 127
Chapter 5: Holocene and Latest Pleistocene Slip Rates on the 129
Coyote Hills Segment of the Puente Hills Blind-Thrust
Fault
5.1 Introduction 129
5.1.1 Puente Hills Thrust Fault 131
5.2 Results 134
5.2.1 High-Resolution Seismic Reflection Data 137
5.2.2 Borehole Data 137
5.3 Discussion 145
5.3.1 Fault Slip Rate Estimates 145
5.4 Conclusions 147
Chapter 5 References 148
Chapter 6: Evidence for Large Holocene Earthquakes on the 149
Compton Fault, Los Angeles, California
Abstract 149
6.1 Introduction 150
6.1.1 Compton Blind-Thrust Fault 151
6.2 Results 153
6.2.1 High-Resolution Seismic reflection Data 153
6.2.2 Borehole Data 154
6.2.3. Stratigraphic Evidence for Paleo-Folding Events 163
6.2.4 Structural Reconstructions 166
6.2.4.1 Methods 166
6.2.4.2 Shear Restoration of Folding Events 167
6.3 Discussion 183
6.3.1 Fault Slip Rate Estimates 183
6.3.2 Paleomagnitude Estimates and Implications 185
for Fault Behavior
6.3.3 Incremental Development of Dips within the 188
Kink Band
6.3.4 Implications for Seismic Hazard in Southern 189
California
Chapter 6 References 192
vii
Chapter 7: Evidence for Holocene-Late Pleistocene Activity 196
of the Compton Fault, Lakewood, California:
Implications for System-Wide Earthquakes and Seismic
Hazard
7.1 Introduction 196
7.1.1 Compton Blind-Thrust Fault 197
7.2 Results 199
7.2.1 High-Resolution Seismic Reflection Data 199
7.2.2 Borehole Data 203
7.2.2.1 Sediment Accumulation Rates 205
7.2.3 Cone Penetration Testing Data 212
7.2.4 Stratigraphic Evidence for Paleo-Folding Events 212
7.3 Discussion 214
7.3.1 Fault Slip Rate Estimates 214
7.3.2 Paleomagnitude Estimates and Implications 216
for Fault Behavior
7.3.3 Incremental Development of Dips within the 220
Kink Band
7.3.4 Implications for Seismic Hazard in Southern 221
California
Chapter 7 References 224
Chapter 8: Conclusions 228
8.1 Summary 228
8.2 Spatial and Temporal Variations in Slip Rates 229
8.3 Implications for the Potential of System-Wide Earthquakes 230
to Occur
8.4 Structural Characterization of Sub-Surface Folds 230
Chapter 8 References 232
References 233
Appendices 239
Appendix A: Supplemental Material for Chapter 2 240
Appendix A References 245
Appendix B: Supplemental Material for Chapter 6 246
Appendix B References 253
viii
LIST OF TABLES
Table 2.1: Radiocarbon Ages and Calibrated, Calendric Dates of Samples 39
from the Carfax and Gardenland Transects
Table 2.2: Uplift amounts, age limits, and Estimated Paleomagnitudes 66
For the Santa Fe Springs Segment of the Puente Hills Blind-Thrust Fault
Table 3.1: Calibrated, Calendric Dates and Radiocarbon ages of Samples 89
from the Budlong Avenue Transect
Table 3.2: Uplift Amounts, Age Limits, and Estimated Paleomagnitudes 97
for the Los Angeles Segment of the Puente Hills Blind-Thrust Fault
Table 4.1: Calibrated, Calendric Dates and Radiocarbon Ages of Samples 117
Collected from the Pico Rivera Study Site
Table 5.1: Calibrated, Calendric Dates and Radiocarbon Ages of Samples 144
from the North Raymond Avenue transect
Table 6.1: Calibrated, Calendric Dates and Radiocarbon Ages of Samples 171
from the Stanford Avenue Transect
Table 6.2: Uplift, Age Limits, and EstimatedPaleomagnitudes for the 184
Compton Blind-Thrust Fault
Table 7.1: Calibrated calendric dates and radiocarbon ages of samples 206
from Stanford Avenue transect
Table 7.2: Uplift, Age Limits, and Estimated Paleomagnitudes for the 217
Southern Segment of the Compton Blind-Thrust Fault
ix
LIST OF FIGURES
Figure 1.1: Satellite Image of the Los Angeles Basin with the 3
Location of Major Fault Zones
Figure 2.1: Location of the Gardenland/Carfax Study Site 23
Figure 2.2: Three-Dimensional Block Diagram of the Metropolitan 25
Los Angeles Region
Figure 2.3: Multi-Scale Seismic Reflection Image of the Forelimb 28
Fold Structure
Figure 2.4: Detailed Location Map of the Study Site 32
Figure 2.5: Borehole Results from the Carfax Transect, 34
Gardenland-Greenhurst Transect, and SCE Transect
Figure 2.6: Cross-section of Borehole Results Overlying High- 36
Resolution Seismic-Reflection Hammer Profile
Figure 2.7: Sediment Accumulation Rate Curves for the Carfax 45
and Gardenland Transects
Figure 2.8: Structural Reconstruction of Uplift Events from the 49
Carfax Avenue Transect
Figure 2.9: Structural Reconstruction of Uplift Events from the 53
Gardenland Avenue Transect
Figure 2.10: Fault-Bend Fold Models 70
Figure 3.1: Structure Contour Map of the Puente Hills Blind- 78
Thrust Fault (PHT)
Figure 3.2: Multi-scale Seismic Reflection Profiles of Folding 82
above the Los Angeles Segment of the Puente Hills Blind-Thrust
Fault
Figure 3.3: Borehole Results from the Budlong Avenue Transect 86
Figure 3.4: Alternative Borehole Results from the Budlong Avenue 87
transect
x
Figure 3.5: Sediment Accumulation-Rate Curve for Budlong 93
Avenue Transect
Figure 4.1: Structure Contour Map of the Puente Hills Blind- 109
Thrust Fault (PHT)
Figure 4.2: Multi-scale Seismic Reflection Images of the Backlimb 111
Fold Structure of the Santa Fe Springs Segment of the Puente Hills
Blind-Thrust Fault
Figure 4.3: Detailed Borehole Log from the Backlimb of the Santa 114
Fe Springs Segment of the Puente Hills Blind-Thrust Fault
.
Figure 4.4: High-Resolution Seismic Reflection Profiles Above the 118
Santa Fe Springs Segment of the Puente Hills Blind-Thrust Fault
Figure 4.5: High-Resolution Seismic Reflection Profiles Processed 123
with Different Velocities for Shallow Sediments
Figure 5.1: Structure Contour Map of the Puente Hills Blind- 132
Thrust Fault (PHT)
Figure 5.2: Multi-Scale Seismic Reflection Profiles of Folding 135
Above the Coyote Hills Segment of the Puente Hills Blind-Thrust
Fault
Figure 5.3: High-resolution Seismic Reflection Profiles of Shallow 138
Folding Above the Eastern and Western Coyote Hills Segments of
the Puente Hills Blind-Thrust Fault
Figure 5.4: Alternative Cross-Sections of Borehole Results from 141
the North Raymond Avenue transect
Figure 6.1: Map of the Compton Thrust with Adjacent Structures 152
in the Los Angeles Basin
Figure 6.2: Contour Map of the Study Site Along Stanford 155
Avenue
Figure 6.3: Compton Fault Seismic Reflection Data 158
Figure 6.4: Borehole from the Stanford Avenue Transect 161
xi
Figure 6.5: Sediment Accumulation-Rate Curve for the Stanford 168
Avenue Transect
Figure 6.6: High-Resolution Seismic Reflection Hammer 174
Profile and Borehole
Figure 6.7: Sequential Reconstructions for the Six Observed 178
Uplift Events
Figure 7.1: Map of the Compton Thrust with Adjacent Structures 200
in the Los Angeles Basin
Figure 7.2: Multi-Scale Seismic Reflection Profiles of Folding 202
Above the Southern Segment of the Compton Fault
Figure 7.3: Borehole Results from the Briercrest Avenue Transect 204
Figure 7.4: Sediment Accumulation-Rate Curve from the 209
Briercrest Avenue Transect
xii
ABSTRACT
In order to understand the paleo-earthquake history and structural evolution of
blind-thrust faults and their associated folds, I use a multi-disciplinary methodology to
link blind faulting at seismogenic depths directly to near-surface fault-related folding. My
research focused on two major blind-thrust systems beneath metropolitan Los Angeles,
California; the Puente Hills thrust fault (PHT) and the Compton thrust fault. I acquired
data from multiple study sites along both faults in order to analyze shallow, subsurface
folding for individual thrust ramps and study the potential for system-wide earthquakes to
occur in segmented blind-thrust systems.
To resolve how folds grow in response to slip on the underlying thrust ramps, I
utilized a methodology that combines high-resolution seismic reflection profiles,
borehole excavations, and cone penetration testing to analyze the overlying growth folds
of segmented blind-thrust faults. These shallow data allow me to observe folding at
multiple depths and identify discrete buried fold scarps formed during past earthquakes at
study sites above individual thrust ramps of both the PHT and Compton thrust.
The results of this dissertation provide Holocene-Late Pleistocene slip rates from
study sites above the Los Angeles, Santa Fe Springs, and Coyote Hills, segments of the
PHT. I also determined slip rates for both the northern and southern segments of the
Compton fault. In addition to fault slip rates, this research demonstrates that both the
Puente Hills and Compton blind thrust faults have generated multiple, large-magnitude
(Mw >7) earthquakes during the past 12-14 ka, and that both faults are capable of
generating multi-segment ruptures. Moreover, this study provides insights into the
xiii
detailed kinematics of earthquake-by-earthquake fold growth above the underlying blind
thrust ramps. At all of the study sites, the borehole data show that the folded strata within
the kink bands acquired their dips incrementally, suggesting that fold kinematics involves
components of both kink-band migration and limb rotation. These analyses of the
geometry and evolution of young, shallow growth structures have allowed me to decipher
the previously unknown paleo-earthquake history of the PHT and Compton fault,
providing the basis for effective seismic hazard analysis.
1
CHAPTER 1:
Introduction
1.1 Introduction
The seismic hazard potential of blind-thrust faults was dramatically illustrated by
the 1994 M
w
6.7 Northridge earthquake. Although this earthquake was only of moderate-
magnitude, it caused billions of dollars of damage and still remains one of the costliest
natural disasters in U.S. history. Blind faults are faults that do not cut the surface of the
earth but instead are manifested by near-surface folding. For many moderate-magnitude
earthquakes on blind-thrust faults, geodetic studies generally show broad, diffuse patterns
of co-seismic uplift of the overlying folds centered above the fault rupture. This suggests
a component of distributed folding during moderate-magnitude ruptures [e.g. Stein and
King, 1984; Lin and Stein, 1989; Stein and Ekstrom, 1992]. Many fault-related folding
theories, however, predict that folding is localized at fault tip-lines and bends [e.g. Suppe,
1983; Suppe and Medwedeff, 1990; Erslev, 1991; Allmendinger, 1998]. This component
of localized folding can occur co-seismically, as observed during the 1980 M
w
7.3 El
Asnam, Algeria and 1999 M
w
7.6 Chi Chi, Taiwan earthquakes. In fact, during the Chi-
Chi earthquake, a 20-m wide, 5-m-high, and more than 20-km long fold scarp developed
above a bend in the Chenglupu fault along a previously mapped geologic syncline. This
earthquake established the significant hazard potential of the formation of surface fold
scarps during large blind-thrust ruptures. These fault-related folds provide the only
discernible record of paleo-earthquake history on the underlying thrust ramps and it is the
discrete nature of deformation associated with large-magnitude ruptures that is within the
reach of paleoseismological investigation.
2
Obtaining paleoseismological data for seismic hazard analysis of blind-thrust
systems has proved to be particularly challenging. Such data are critical, however, to
analyze the processes of blind faulting. This dissertation comprises research that used a
multi-disciplinary methodology to analyze shallow, sub-surface folding above individual
fault segments of both the Puente Hills (PHT) and Compton blind-thrust faults. My
research focused on acquiring high-resolution seismic reflection data combined with cone
penetration testing (CPT) and borehole excavations. This integrated approach provided
the tools necessary to fill in missing slip rates for each segment of these two major
hazards beneath metropolitan Los Angeles.
This study was proposed to define how the PHT and Compton fault, and their
associated fault-related folds, evolved through Holocene time and decipher
paleoseimologic data for seismic hazard analysis (Figure 1.1). This methodology,
combined with the collection and dating of Quaternary sediments involved in these folds,
helped to constrain how discrete, shallow subsurface folding records fault slip history for
individual thrust ramps. Moreover, collection of data from multiple study sites facilitated
comparison of near-surface folds between individual fault segments to determine whether
they exhibit similar patterns and styles of deformation.
1.2 Active Blind-Thrust Faulting and Folding in the Los Angeles Basin
The Los Angeles Basin lies near the Southern California coast between the
Transverse and Peninsula Ranges. It consists of Quaternary through upper Miocene
strata more than 5 km thick, which overly middle Miocene through Upper Cretaceous
units [Yerkes et al., 1965] and Mesozoic Catalina schist [Crouch and Suppe, 1993].
3
Figure 1.1 Satellite Image of the Los Angeles Basin with the Location of Major Fault
Zones
The location of the Puente Hills thrust (PHT) and Compton blind-thrust systems as
well as other blind-thrust systems are shown on a satellite image of the Los Angeles
Basin. The cross-hatched yellow lines indicate the location of downtown Los
Angeles. Blue lines indicate the location of the Los Angeles River (west), San Gabriel
River, and Santa Ana River (east). White circles indicate locations of the study sites;
1 - Santa Fe Springs segment of the PHT (Chapter 2), 2 - Los Angeles segment of the
PHT (Chapter 3), 3 - backlimb site above the Santa Fe Springs segment of the
PHT(Chapter 4), 4 - Coyote Hills segment of the PHT (Chapter 5), 5 - northern
segment of the Compton fault (Chapter 6), 6 - southern segment of the Compton
fault (Chapter 7).
4
Figure 1.1, Continued
5
Extensional tectonics began as early as ~24 million years ago and transitioned to
transpressional tectonics ~4 million years ago to produce the northwest-southeast
trending basin geometry [Wright, 1991; Yeats and Beal, 1991; Crouch and Suppe, 1993].
The basin and surrounding region is currently being deformed by several active fault
zones that have produced moderate-size historic earthquakes including the 1933 Long
Beach, 1971 San Fernando, 1987 (M
w
6.0) Whittier-Narrows, and 1994 (M
w
6.7)
Northridge [Hauksson and Jones, 1989; Hauksson, 1990; USGS and SCEC Scientists,
1994]. Geodetic studies suggest a 4.4 to 5 mm/yr rate of basin shortening in a north-
south to northeastern-southwestern direction [Walls et al., 1998; Argus et al., 1999; 2005;
Bawden et al., 2001]. It appears that deformation within the basin is dominated by
decoupled strike-slip and thrust motions [Hauksson, 1990] with competing models to
account for the north-south closing of the basin. Walls et al. [1998] suggested that the
majority of basin shortening is taken up by conjugate strike-slip faulting, while Bawden
et al. [2001] proposed that at least 50% of this rate is taken up by thrust faulting. Some
of this basin shortening, however, is still unresolved. It is possible that as much as half of
the unresolved north-south shortening is taken up by blind-thrust faulting.
1.2.1 Puente Hills Blind-Thrust fault
The Puente Hills thrust (PHT) is a major blind thrust fault, originally identified by
Shaw and Shearer [1999], that extends for >40 km beneath the northern edge of the Los
Angeles metropolitan region from near Beverly Hills east-southeastward beneath
downtown Los Angeles to northern Orange County (Figure 1.1). Shaw et al. [2002]
documented the subsurface geometry of the PHT using petroleum industry seismic
6
reflection and well data. They identified three right-stepping, en-echelon blind-thrust
ramps that dip to the north ~ 25-30
o
and terminate upwards at ~ 3 km depth. These fault
segments are, from west to east, the Los Angeles segment, the Santa Fe Springs segment,
and the Coyote Hills segment. Rupture of a small part of the central segment of the PHT
generated the 1987 M
w
6.0 Whittier Narrows earthquake. Hypocentral relocations for the
1987 event and its aftershocks reveal that the mainshock fault plane was coincident with
the down-dip projection of the Santa Fe Springs segment of the Puente Hills thrust at ~13
km depth [Shaw and Shearer, 1999]. This relatively small earthquake provided direct
evidence that the PHT is active and capable of generating damaging earthquakes directly
beneath the heart of the Los Angeles metropolitan region, home to more than 12 million
people.
Petroleum-industry seismic reflection profiles reveal that large anticlines have
developed above each of the three fault segments during slip on the underlying blind-
thrust ramps [Shaw et al., 2002]. Structural analysis and biostratigraphic age control
from well data reveal a two-stage history of fault motion and fold growth on the central
Santa Fe Springs segment. This consists of an initial period of Pliocene fault-propagation
or tri-shear folding followed by a period of structural quiescence and subsequent early
Quaternary reactivation of the system in the currently active phase of folding. Average
Quaternary slip rates were calculated for individual ramp segments of the PHT based on a
north-dipping fault plane between 20
o
and 40
o
with preferred rates based on a 27
o
dipping
fault. These slip rates for the Santa Fe Springs, Los Angeles, and Coyote Hills segments
are 0.44 to 0.82 mm/yr with a preferred rate of 0.62 mm/yr, 0.60 to 1.13 mm/yr with a
7
preferred rate of 0.85mm/yr, and 0.90 to 1.70 mm/yr with a preferred rate of 1.28 mm/yr,
respectively [Shaw et al., 2002].
1.2.2 Compton Blind-Thrust Fault
The Compton thrust fault was originally identified by Shaw and Suppe [1996]
using petroleum-industry seismic reflection profiles and well data. Based on their
analysis of the folding that occurs above the thrust fault, they showed that the Compton
fault extends northwest-southeast for ~40 km beneath the western edge of the Los
Angeles metropolitan region (Figure 1.1). Petroleum-industry seismic reflection data
define a growth fault-bend fold associated with the gently northeast-dipping thrust fault
ramp, which, combined with well data, reveal compelling evidence for Pliocene and
Pleistocene activity.
The Compton blind-thrust fault is overlain by the Compton-Los Alamitos trend
and extends through the center of the Los Angeles Basin. Petroleum-industry seismic
reflection profiles image a northeast dipping monoclinal fold limb that trends northwest-
southeast. Continuous, coherent reflectors across the fold trend confirm the presence of
Quaternary folding that thins across the crest of the fold. Petroleum-industry seismic
reflection profiles image dipping strata that narrows upward into a well-defined kink
band suggesting at least a component of folding occurs through kink-band migration.
The Compton ramp strikes northwest-southeast and dips to the north between 20
o
to 30
o
parallel to the overlying fold limb. At the base of the Compton fault, > 7 km depth, a
decollement may link the Compton ramp to the thrust ramps in the lower Elysian Park
fault system. Based on the analysis of folding observed in industry seismic reflection
8
profiles, the Compton ramp consists of two distinct northwest trending fault segments
with a central north-northwest trending section representing a lateral ramp connecting the
two main segments. The Pliocene/ Quaternary dip-slip rate is ~1.4+0.4 mm/yr for the
entire Compton thrust ramp [Shaw and Suppe, 1996]. A comprehensive study by Lehle
[2007] analyzed forty industry seismic reflection profiles to provide detailed mapping of
the subsurface structure of the Compton-Los Alamitos fold trend, and concluded that
each fault segment experienced different slip rates during the past ~2.5 million years. The
slip rate for the northern segment of the Compton fault remained fairly consistent
throughout Pliocene/ Quaternary time at 1.48+0.39 mm/yr. However, the southern
segment experienced a marked change in slip rate at ~1 MA from 0.42 mm/yr to 2.69 +
0.67 mm/yr. Moreover, the temporally and spatially discrete slip pulses of each ramp
segment suggests that either the faults merged early in the fault history but were still able
to behave in a segmented fashion, or that the fault remained segmented throughout most
of its early history but was still capable of simultaneous, multi-segment ruptures.
1.3 Methods
1.3.1 High-Resolution Seismic Reflection Data
Petroleum industry seismic reflection data typically does not image the upper ~
200 m of the stratigraphic section which records the most recent deformation associated
with slip on the blind-thrust fault at depth. In order to bridge this data gap a key
component of this study was the acquisition of high-resolution seismic reflection data at
multiple scales. In collaboration with Thomas L. Pratt (USGS) I acquired high-resolution
seismic reflection data at multiple study sites utilizing three different seismic sources.
9
For the study sites above the Los Angeles segment of the PHT (Chapter 3) and northern
segment of the Compton fault (Chapter 6), I used a geometrics seismograph, available
from IRIS, using a 60-channel system with weight-drop source to image 30- to 600-m-
depth range and hammer source for 10- to 100-m-depth range. Geophones were placed
along the side of the street at 5-m intervals for the weight-drop source profile and 2-m
intervals for the hammer profile. Four impacts from a truck-mounted Geometrics PWD
weight-drop source (82 kg) using elastomer technology were recorded at each source
point (5 m apart). For the higher-resolution seismic reflection profile four impacts were
recorded with a 4.5 kg sledgehammer at 1-m intervals. High-resolution seismic reflection
profiles acquired from study sites above the backlimb of the Santa Fe Springs segment
(Chapter 4), Coyote Hills segment of the PHT (Chapter 5), and southern segment of the
Compton fault (Chapter 7), utilized a geometrics seismograph with a 144-channel system
and mini-vibe T7000 source provided by the University of Nevada, Las Vegas. These
profiles were acquired with 60 hertz geophones at 5 m spacing. The high-resolution
seismic reflection data imaged reflectors from 30 to 1000 m depth range. An additional
profile, using the mini-vibe source, was also acquired at the study site above the Los
Angeles segment of the PHT with geophones at 2-m spacing. The high-resolution seismic
reflection data was processed by Thomas L. Pratt (USGS) at the University of
Washington, Seattle using Linux workstations with the Seismic Unix (Colorado School
of Mines) software package. Combining the high-resolution seismic reflection data with
the existing petroleum industry profiles allowed us to observe the styles of fault-related
10
folding above individual ramps of both segmented thrust systems from ~ 6-km depths to
within 10 to 60 m’s of the surface.
1.3.2 Borehole Excavations
Any successful investigation of Holocene folding requires analysis of near-
surface deformation; this shallow folding represents uplift during the most recent events
on the underlying thrust ramp. Borehole excavation not only allows us to sample folded
horizons at depths up to ~70 m, but, unlike typical trenching techniques, it also allows us
to study folds beneath urbanized areas.
In order to analyze the near-surface deformation at our study sites, I drilled
hollow-stem, continuously cored boreholes along a transect parallel to the high-resolution
seismic reflection data across the zone of active folding. The 8- to 10-cm-diameter
boreholes penetrated depths of up to 66 m and provided information on near-surface
deformation of stratigraphic horizons. These shallow excavations also provided the
opportunity to collect samples for radiocarbon dating, thus providing age control for
uplift events as well as slip rates for individual ramp segments. Boreholes were excavated
using either a CME 75, 85, or 95 drill rig. The type of drill rig used depends on the target
depth for borehole excavation. Commercial companies were contracted to drill the
boreholes and extract the cores for analysis. TestAmerica Inc. acquired boreholes for the
study sites above the Santa Fe Springs (both forelimb and backlimb sites; Chapters 2 and
4), Los Angeles (Chapter 3), and Coyote Hills (Chapter 5) segments of the PHT, and both
segments of the Compton fault (Chapters 6 and 7). Additional boreholes were acquired
by ABC drilling for the study site above the southern segment of the Compton fault
11
(Chapter 7). Cores were logged during drilling and grain size, texture, color, and any
structural features recorded during acquisition.
Stratigraphic units were correlated between boreholes by grain size, texture and
color (using a Munsell color chart). Measuring the magnetic susceptibility of collected
cores in 1-inch intervals provided additional correlations between boreholes for both the
Los Angeles segment of the PHT (Chapter 3) and the southern segment of the Compton
fault (Chapter 7). Boreholes excavated at the study sites provided abundant detrital
charcoal and organic-rich soils for sampling and radiocarbon dating. Samples from my
study site above the forelimb of the Santa Fe Springs segment of the PHT (Chapter 2)
were dated at the Accelerator Mass Spectrometry (AMS) laboratory located at the
University of Arizona, Tucson. The remaining samples were sent to Lawrence
Livermore National Laboratories (AMS) for analysis.
When combined with uplift measurements for individual events, radiocarbon ages
facilitated timing of paleo-earthquakes for individual ramp segments. These data were
used to compare timing of uplift events for segmented blind-thrust systems to gain a
better understanding of the potential for system-wide ruptures occurring across
segmented systems. Furthermore, combining measurements of structural relief with age
control from radiocarbon analyses provided unprecedented insight into the rates and
patterns of fault activity and enabled comparison of short-term Holocene slip rates with
the long-term rates derived from petroleum-industry seismic reflection profiles.
12
1.3.3 Cone Penetration Testing
In support of the borehole excavations, cone penetration testing (CPT) provided
additional analysis of the stratigraphy within the zones of deformation encountered at my
study sites. Although cores are not collected directly, CPT analysis combines three main
parameters for soil behavior that are generally consistent with the grain size of
stratigraphic units logged from nearby borehole excavations. The CPT consists of a
penetrometer pushing vertically into the soil, at a relatively slow and constant rate. A
penetrometer comprises a series of rods terminated by a penetrometer tip, which includes
a cone and a cylindrical shaft. The cone measures the penetration resistance of the
penetrometer as well as the local friction on a sleeve located in the cylindrical shaft.
Measurements of the axial force acting on the cone are divided by the projected area to
calculate resistance or cone bearing (Q
c
). The force acting on the friction sleeve within
the cylindrical shaft is divided by the surface area of the friction sleeve to measure sleeve
friction (f
s
). Measurements of Q
c
and F
s
are continuously calculated as the cone pushes
vertically in the ground. Additional pore pressure measurements are collected every 5 cm
by a pore pressure transducer located behind the tip of the cone. The friction ratio (R
f
%)
is then calculated by dividing the sleeve friction by the cone bearing, determined at the
same depth and expressed as a percentage. A total (corrected) cone resistance (q
t
) is
calculated by adding the cone bearing measurement (q
c
) to the pore pressure
measurement (u) multiplied by 1 minus the cone area factor (a), which is defined as the
net area of the cone (A
N
) divided by total area of the cone (A
C
). The interpreted soil
behavior (SBT) is reported for depths of up to 50 feet and is calculated by graphing cone
13
bearing (q
t
) in bars versus friction ratio (R
f
) as a percent after Robertson et al. [1986].
For depths greater than 50 feet SBTn is calculated after Robertson [1990]. These
geotechnical parameters are interpreted by the software based on q
t
, f
s
, and u
2
with 12 soil
behavior zones from sensitive, fine-grained material to sand-gravelly sand and reported
as vertical logs. This method of data acquisition was applied to my study site above the
southern segment of the Compton fault and the Los Angeles segment of the PHT to study
the shallow, sub-surface folding associated with the southern segment of the Compton
fault. The CPT information logs provided additional confidence in correlating
stratigraphic units across the zone of deformation located by high-resolution seismic
reflection and borehole data. All CPT data were collected and analyzed by Gregg
Drilling, Inc.
1.4 Research Implications
Two of the chapters in my dissertation present research results previously
published in international, peer-reviewed journals [Leon et al., 2007 (Chapter 2); Leon et
al., 2009 (Chapter 6)]. Two additional chapters are currently in preparation for
submission [Leon et al., in prep (Chapter 3); Leon et al., in prep (Chapter 7)].
Chapter 2 details the results from a study to investigate the three-dimensional
nature of buried fold scarps above the Santa Fe Springs segment of the Puente Hills
Thrust fault (PHT) and provide supporting paleoseimologic data for seismic hazard
analysis.
Chapter 3 reports on the first paleoseismologic data from the Los Angeles
segment of the PHT. A combination of high-resolution seismic reflection profiles,
14
borehole data, and cone penetration tests (CPT’s) reveal the occurrence of four buried
fold scarps. Analysis of individual fold scarps provided event-by-event uplift, paleo-
magnitudes, and slip rates for the westernmost segment of the PHT. Preliminary
uplift calculations and timing of events suggests the possibility of multi-segment ruptures
involving both the Santa Fe Springs and Los Angeles segments of the PHT.
Preliminary results from a study site above the backlimb fold of the central, Santa
Fe Springs segment of the PHT are detailed in Chapter 4. A combination of petroleum-
industry and high-resolution seismic reflection profiles provides a continuous of image of
reflectors from ~7 km depth to within 45 m of the surface. Additionally, excavation of a
66-m-deep borehole at this location provided samples for both radiocarbon and optically
stimulated luminescence (OSL) dating. Although I do not yet have ages from the OSL
samples, by combining the radiocarbon ages with our calculation of structural relief
across the upper four reflectors imaged in the high-resolution data, I was able to calculate
an independent Holocene/Latest Pleistocene slip rate for the Santa Fe Springs segment of
the PHT.
Chapter 5 presents initial results of both high-resolution seismic reflection and
borehole data above the Coyote Hills segment of the PHT. This research provides a
preliminary Holocene/Late Pleistocene slip rate for the easternmost segment of the PHT.
An analysis of event-by event uplift at a study site above the northern segment of
the Compton fault is presented in Chapter 6. This study provides evidence for six,
temporally discrete, uplift events during Holocene time, and documents a
Holocene/Latest Pleistocene slip rate for Compton fault.
15
Chapter 7 presents the results of a combined study integrating high-resolution
seismic reflection, borehole, and cone penetration testing (CPT) data acquisition for the
southern segment of the Compton fault. This research documents a multi-segment
Holocene study of deformed strata above this large blind-thrust system underlying the
western edge of the Los Angeles Basin.
A summary of the main findings of my research is included in the conclusion
section. Each chapter of my dissertation is designed to stand alone, thus an amount of
redundancy is unavoidable. For this I apologize to the reader.
16
Chapter 1 References
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Argus, D. F., Heflin, M. B., Donnellan, A., Webb, F. H., Dong, D., Hurst, K. J.,
Jefferson, D. C., Lyzenga, G. A., Watkins, M. M., and Zumberge, J. F. (1999),
Shortening and thickening of metropolitan Los Angeles measured and inferred by
using geodesy: Geology, v. 27, n. 8, 703 – 706.
Argus, D. F., Heflin, M. B., Peltzer, G., and Crampe´, F. (2005), Interseismic strain
accumulation and anthropogenic motion in metropolitan Los Angeles, J. Geophys.
Res., 110, B04401, doi: 10.1029/2003JB002934.
Bawden, G. W., Thatcher, W., Stein, R. S., Hudnut, K. W., and Peltzer, G. (2001),
Tectonic contraction across Los Angeles after removal of groundwater pumping
effects: Nature, 412, 712-815.
Crouch, J. A., and Suppe, J. (1993), Late Cenozoic tectonic evolution of the Los Angeles
basin and Inner California Borderland: A model for core-complex-like crustal
extension, Geol. Soc. Am. Bull., v. 105, 1415-1434.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Hauksson, E. (1990), Earthquakes, faulting, and stress in the Los Angeles Basin, J. Geophys.
Res., 95, 15,365-15,394.
Hauksson, E., and Jones, L. (1989),The 1987 Whittier Narrows earthquake sequence in Los
Angeles, southern California: Seismological and tectonics analysis: J. Geophys. Res.,
94, 9569.
Lehle, D. (2007), Geometry and slip history of the Compton thrust fault: Implications for
earthquake hazard assessment, thesis, Harvard Univ., Cambridge, Mass.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Leon, L. A., J. F. Dolan, J. H., Shaw, and T. L. Pratt (2009), Evidence for large Holocene
earthquakes on the Compton thrust fault, Los Angeles, California, J. Geophys.
Res., in press, doi: 10.1029/2008JB006129.
17
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Roberston, P.K. (1990), Soil Classification using the Cone Penetration Test, Canadian
Geotech. Journ., v. 27, 151-158.
Robertson, P.K., Campanella, R. G., Gillespie, D., and Rice, A. (1986), Seismic CPT to
Measure In-Situ Shear Wave Velocity, J. Geotech. Eng. ASCE, v. 112, No. 8,
791-803.
Scientists of the U. S. Geological Survey and the Southern California Earthquake Center
(1994), The magnitude 6.7 Northridge, California, earthquake of 17 January 1994,
Science, 266, 389-397.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Stein, R. S. and Ekström, G. (1992), Seismicity and geometry of a 110-km-long blind
thrust fault. 2. Synthesis of the 1982 – 1985 California earthquake sequence: J.
Geophys. Res., 97(B4), 4865 - 4883.
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
Suppe, J. (1983), Geometry and kinematics of fault-bend folding: Am. J. Sci., 283, p.
684-721.
Suppe, J., and Medwedeff, D. A. (1990), Geometry and kinematics of fault-propagation
folding: Ecologae Geol. Helv., 83, no. 3, p. 409-454.
Walls, C., Rockwell, T., Mueller, K., Bock, Y., Williams, S., Pfanner, J., Dolan, J., and
Peng, F. (1998), Escape tectonics in the Los Angeles metropolitan region and
implications for seismic risk: Nature, 394, 356-360.
18
Wright, T. L. (1991), Structural geology and tectonic evolution of the Los Angeles basin,
California, in Active Margin Basins, edited by K. T. Biddle, AAPG Mem., 52, 35-
134.
Yeats, R. S., and Beall, J. M. (1991), Stratigraphic controls of oil fields in the Los
Angeles basin: A guide to migration history, in Active Margin Basins, edited by
K. T. Biddle, AAPG Bull. Mem., 52, 221-235.
Yerkes, R. F., McCulloh, T. H., Schoellhamer, J. E., and Vedder, J. G. (1965), Geology
of the Los Angeles basin, California – An introduction, U.S. Geol. Surv. Prof.
Pap., 420-A, 57.
19
CHAPTER 2:
Earthquake-by-Earthquake Fold Growth above the Puente Hills Blind Thrust
Fault, Los Angeles, California: Implications for Fold Kinematics and Seismic
Hazard
Abstract
Boreholes and high-resolution seismic reflection data collected across the
forelimb growth triangle above the central segment of the Puente Hills thrust fault (PHT)
beneath Los Angeles, California, provide a detailed record of incremental fold growth
during large earthquakes on this major blind thrust fault. These data document fold
growth within a discrete kink band that narrows upwards from ~460 m at the base of the
Quaternary section (200-250 m depth) to <150 m at 2.5 m depth, with most growth
during the most recent folding event occurring within a zone only ~60 m wide. These
observations, coupled with evidence from petroleum-industry seismic reflection data,
demonstrate that most (>82% at 250 m depth) folding and uplift occur within discrete
kink bands, thereby enabling us to develop a paleoseismic history of the underlying
blind-thrust fault.
The borehole data reveal that the youngest part of the growth triangle in the
uppermost 20 m comprises three stratigraphically discrete growth intervals marked by
southward-thickening sedimentary strata that are separated by intervals in which
sediments do not change thickness across the site. I interpret the intervals of growth as
occurring after the formation of now-buried paleo-fold scarps during three large PHT
earthquakes in the past 8 ka. The intervening intervals of no fold growth record periods of
structural quiescence and deposition at the regional, near-horizontal stream gradient of
20
the study site. Minimum uplift in each of the scarp-forming events, which occurred at
0.2-3.0 ka [Event Y], 3.0-6.3 ka [Event X], and 6.6-8.1 ka [Event W], ranged from ~1.1
to ~1.6 m, indicating minimum thrust displacements of ≥2.5 to 4.5 m. Such large
displacements are consistent with the occurrence of large-magnitude earthquakes (M
w
>7). Cumulative, minimum uplift in the past three events was 3.3 to 4.7 m, suggesting
cumulative thrust displacement of ≥7 to 10.5 m. These values yield a minimum Holocene
slip rate for the PHT of ≥0.9 to 1.6 mm/yr.
The borehole and seismic reflection data demonstrate that the dip of the strata
within the kink band is acquired incrementally, such that older strata that have been
deformed by more earthquakes dip more steeply than younger strata. Specifically, strata
dip 0.4
o
at 4 m depth, 0.7
o
at 20 m depth, 8° at 90 m, 16° at 110m, and 17° at 200 m.
Moreover, structural restorations of the borehole data show that the locus of active
folding (the anticlinal active axial surface) does not extend to the surface in exactly the
same location from earthquake to earthquake. Rather, the axial surfaces migrate from
earthquake to earthquake, reflecting a component of fold growth by kink-band migration.
The incremental increase of bed dip in the growth triangle may reflect some combination
of fold growth by limb rotation in addition to kink-band migration, possibly through a
component of tri-shear or shear fault-bend folding. Alternatively, the component of limb
rotation may result from curved-hinge fault-bend folding, and/or the mechanical response
of loosely consolidated granular sediments in the shallow subsurface to folding at depth.
21
2.1 Introduction
The degree to which folding above active blind thrust faults is localized within
discrete zones has been vigorously debated within the structural geology and earthquake
science communities in recent years. On the one hand, simple elastic half-space models
of several small- to moderate-magnitude blind thrust earthquakes have shown regionally
diffuse uplift patterns that are broadly consistent with geodetic data (e .g ., 1983 M
w
6.5
Coalinga, 1987 M
w
6.0 Whittier Narrows, and 1994 M
w
6.7 Northridge, California,
earthquakes [e.g., Stein and King, 1984; Lin and Stein, 1989; Stein and Ekström, 1992;
Hudnut et al., 1996]). In contrast, most kinematic models of fold development invoke
large components of folding localized along axial surfaces that are pinned to fault bends
and fault tips. These models have been used to develop precise records of the long-term
structural evolution of numerous, thrust-related folds provided by geologic mapping and
petroleum-industry seismic-reflection and well data [e.g. Suppe, 1983; Suppe and
Medwedeff, 1990; Erslev, 1991; Shaw and Suppe, 1996; Allmendinger, 1998]. Thus,
models of coseismic and long-term fold growth generally do not agree. I attempt to
reconcile these differences by examining the expression of folding at a variety of scales
above an active blind-thrust fault.
In this chapter, I use continuously cored borehole data, in conjunction with high-
resolution and petroleum-industry seismic reflection data collected and analyzed during
companion studies [Pratt et al., 2002; Shaw et al., 2002], to document the geometry and
structural evolution of very young folds that have developed above the Puente Hills thrust
fault; a major blind thrust beneath metropolitan Los Angeles, California. Herein, I present
22
new borehole data that, together with the previously published borehole results of Dolan
et al. [2003], allow us to assess the three-dimensional geometry of buried fold scarps and
to demonstrate the reproducibility of our stratigraphic and structural measurements.
Moreover, I present detailed sequential structural restorations of fold growth during
discrete uplift events. I use these data to generate paleo-earthquake magnitudes and slip
rates for the blind thrust fault, and discuss the data in light of their implications for fold
kinematics and seismic hazard assessment.
2.1.1. Puente Hills Thrust Fault
The Puente Hills thrust (PHT) is a major blind thrust fault, originally identified by
Shaw and Shearer [1999], that extends for >40 km beneath the northern part of the Los
Angeles metropolitan region, from near Beverly Hills east-southeastward beneath
downtown Los Angeles to northern Orange County (Figure 2.1). Shaw et al. [2002]
documented the subsurface geometry of the PHT using petroleum industry seismic
reflection and well data. They identified three right-stepping, en-echelon blind thrust
ramps that dip to the north at ~ 25-30
o
and terminate upwards at ~ 3 km depth. These
fault segments are, from west to east, the Los Angeles segment, the Santa Fe Springs
segment, and the Coyote Hills segment. Rupture of a small part of the central segment of
the PHT generated the 1987 M
w
6.0 Whittier Narrows earthquake (Figure 2.2).
Hypocentral relocations for the 1987 event and its aftershocks reveal that the mainshock
fault plane was coincident with the down-dip projection of the Santa Fe Springs segment
of the Puente Hills thrust at ~13 km depth [Shaw and Shearer, 1999]. This relatively
small earthquake provided direct evidence that the PHT is active and capable of
23
Figure 2.1 Location Map of the Santa Fe Springs Segment Study Site.
(A) Map with location of study site (red dot) and (B) cross section showing the
location of the study site (red dot) and its relation to the Puente Hills thrust (PHT).
Inset box shows the area enlarged in (A). The forelimb active axial surface above
the Santa Fe Springs segment of the Puente Hills Thrust (PHT) extends upwards
from the tip of the thrust ramp and deforms sediments excavated by our borehole
transects. White areas denote mapped extent of PHT, with 5-km depth contours on
fault surface (modified from Shaw et al., 2002). LAS shows the Los Angeles segment
of the PHT, SSFS shows the Santa Fe Springs segment, and CHS shows the Coyote
Hills segment of the fault as mapped by Shaw et al. (2002). Note that the irregular
southern half of the blue line on part (A) from B to B’ denotes the portion of the
industry seismic reflection profile collected along the San Gabriel River. MF,
Montebello fault; WF, Whittier fault; LA, downtown Los Angeles area; QT,
Quaternary; Tfu, Pliocene upper Fernando Formation; Tfl, Pliocene lower
Fernando Formation; Tp, Miocene Puente Formation; A, away; T, towards; S.L. sea
level. Map coordinates are Universal Transverse Mercator Zone 11, North
American datum 27. The blue lines in (A) show the location of the cross section.
(Modified after Shaw and Shearer [1999]; Shaw et al. [2002]).
24
Figure 2.1, Continued
25
Figure 2.2 Three-dimensional Block Diagram of the Metropolitan Los Angeles
Region.
Cut-away three-dimensional block diagram illustrating proximity of Puente Hills
thrust fault to the heart of the metropolitan Los Angeles region. Image constructed
by draping a Landsat image over a digital elevation model. Cross section along
same profile as in figure 2.1 (modified from Shaw and Shearer, 1999). Red dot
denotes Carfax - Gardenland study site discussed in text. Green dashed lines show
surface projection of locus of active folding along thrust ramp tipline (from Shaw et
al., 2002). Afershocks are colored and scaled by magnitude; M
w
2 – small, green,
M
w
6 – large, purple. No vertical exaggeration. Figure generated by Andreas Plesch
and John Shaw (Harvard University).
26
generating damaging earthquakes directly beneath the heart of metropolitan Los Angeles,
home to more than 12 million people.
Petroleum-industry seismic reflection profiles reveal that large anticlines have
developed above each of the three fault segments during slip on the underlying blind
thrust ramps [Shaw and Suppe, 1996; Shaw et al., 2002]. Structural analysis and
biostratigraphic age control from well data reveal a two-stage history of fault motion and
fold growth on the central Santa Fe Springs segment. This consists of an initial period of
Pliocene fault-propagation or tri-shear folding followed by a period of structural
quiescence and subsequent early Quaternary reactivation of the system in the currently
active phase of folding [Shaw et al., 2002]. The seismic reflection data reveal narrow,
well-defined forelimb and backlimb kink bands above the Santa Fe Springs ramp, and a
classic fault-bend fold geometry (Figure 2.3). The data show little evidence for
significant Quaternary folding outside of the kink bands [Shaw et al., 2002].
For example, in the petroleum-industry seismic reflection data (e. g., Figure 2.3A)
comparisons of the structural relief across the forelimb kink band versus the total
maximum-possible structural relief across the entire fold, measured relative to the
regional (and almost certainly non-tectonic) - limiting dip of the footwall - indicate that:
(1) for the base Quaternary horizon (yellow line in figure 3A), at least 70% of the folding
(300 m out of total maximum-possible structural relief of 430 m, based on a regional
footwall dip of 2.2°) occurs within the kink band; and (2) for the uppermost Quaternary
horizon imaged in the petroleum-industry data at ~250 m depth (uppermost pink line in
figure 3A), at least 82% of the folding (90 m out of total maximum-possible structural
27
relief of 110 m, based on a regional dip of 0.8°) occurs within the kink band. I emphasize
that these values are minima, as we used the non-tectonic, depositional dip from the
central part of the Los Angeles basin, well to the south of the PHT folding, as the limiting
dip in our calculations. The data permits the possibility that essentially all folding occurs
within the discrete kink bands.
The most recent deformation associated with slip on the PHT thrust ramps is
recorded in the youngest sediments at the top of the growth triangle. The petroleum
industry reflection data, however, typically do not image strata within the uppermost 200
– 300 m. In order to bridge the resulting data gap from the shallowest industry data at
~250 m depth to the surface, we acquired, during a companion study [Pratt et al., 2002],
high resolution seismic reflection data at two different scales (Figures 2.3B and C).
2.2 Data
2.2.1 Seismic Reflection Data
The study site is located in the City of Bellflower, about 25 km southeast of
downtown Los Angeles, along the distal, low-gradient floodplain of the San Gabriel
River, a major south-flowing drainage (Figure 2.1). Pratt et al. [2002] collected the two
north-south high resolution seismic reflection profiles along the same alignment ~100 m
to the west of, and parallel to, the active channel of the San Gabriel River, along the east
curb of Carfax Avenue (Figures 2.3B and C). This alignment is ~90 m west of, and
parallel to, the petroleum industry seismic reflection profile shown in Figure 2.3A. The
seismic reflection profiles are oriented nearly perpendicular to the active anticlinal axial
28
Figure 2.3 Multi-scale Seismic Reflection Image of the Forelimb Fold Structure.
Multi-scale seismic reflection images of the forelimb fold structure showing upward-
narrowing zone of active folding (growth triangle) delimited by sharply defined
axial surfaces (Shaw and Shearer, 1999; Pratt et al., 2002). These overlapping
profiles provide a complete image of forelimb folding above the Santa Fe Springs
segment of the Puente Hills thrust fault from the top of the thrust ramp at 3 km
depth to the surface. Red lines represent fault plane reflections, solid, colored lines
in (A) and dashed colored lines in (B) and (C) represent reflectors, and dashed black
lines represent axial surfaces.
29
Figure 2.3, Continued
30
surface associated with the PHT forelimb growth triangle, which extends approximately
east-west beneath the study site [Shaw et al., 2002].
Pratt et al. [2002] used a Mini-Sosie source [Barbier, 1983; Stephenson et al.,
1992] to image the 50 to 500 m depth range (Figure 3B) and a sledge hammer source to
image the 10 to 100 m depth range (Figure 2.3A [Pratt et el., 2002]). Three prominent
reflectors located at ~ 280 m, ~ 110 m, and ~ 90 m depth are discernible on the Mini-
Sosie profile. These three reflectors are folded within an upward-narrowing zone 300 to
400 m wide, coincident with the upward continuation of the kink band identified at depth
by Shaw and Shearer [1999] on the deeper-penetration industry profile. The hammer
profile exhibits several prominent, gently south-dipping reflectors between 10 and 25 m
depth located at the upward projection of the growth triangle observed in the Mini-Sosie
profile. The shallowest reflection at ~10 – 12 m depth appears to correlate with the top of
a laterally extensive sand unit (Unit 40 sand; see discussion below). The high-resolution
and petroleum-industry seismic reflection data thus provide complete, overlapping
coverage of the forelimb growth triangle above the Santa Fe Springs segment of the PHT
from the top of the thrust ramp at ~3 km depth to 10 m depth.
2.2.2 Borehole Observations
The shallowest reflectors imaged on the high-resolution hammer profile are well
within the reach of standard geotechnical boreholes. In order to document the details of
recent fold growth above the PHT, I drilled three parallel, north-south transects of
continuously cored boreholes to depths of 11 to 50 m across the locus of recent folding
identified in high-resolution seismic reflection data (Figures 2.3B and C). The Carfax
31
site is ideally suited for the borehole investigation because of its location on the distal,
near-flat floodplain of San Gabriel River. The site is 40 to 70 km downstream of the main
sediment sources, and sediment deposited there during Holocene time is relatively fine-
grained - typically sand, silt, and clay (see discussion below). The river flows almost due
south at this location, approximately perpendicular to the surface projection of the active
anticlinal fold axial surface observed on the deep-penetration petroleum-industry seismic
reflection data [Shaw et al., 2002]. The three north-south borehole transects are thus
oriented approximately perpendicular to the upward projection of the active axial surface
of folding above the Puente Hills blind thrust fault.
The continuously cored boreholes allow me to document the three-dimensional
subsurface geometry and ages of the youngest sediments folded above the central
segment of the PHT. From east to west, the three borehole transects were drilled along
the Southern California Edison power line right-of-way (SCE transect), Carfax Avenue
(Carfax transect), and Gardenland Avenue and Greenhurst Street (Gardenland transect).
The 450-m-long Carfax transect, approximately ~100 m east of the modern San
Gabriel River, comprised 16 boreholes drilled to ~11 to 50 m depths (Figures 2.4 and
2.5). Initially interpreted by Dolan et al. [2003], this borehole transect was excavated
along the same alignment as the high-resolution hammer seismic reflection profile
(Figure 2.3C). These boreholes thus provide direct ground truth on the youngest
reflectors imaged by seismic reflection data (Figure 2.6). Seven of these boreholes (18
through 24) were continuously sampled hollow-stem boreholes, whereas eight of the
boreholes (10 through 17) were large-diameter (70 cm) bucket-auger holes. The bucket
32
Figure 2.4 Detailed Location Map of the Study Site.
Street map of Carfax Avenue, Gardenland Avenue, Greenhurst Street, and
Southern California Edison right-of-way showing the spatial relationship between
the three transects and the San Gabriel River. Dashed black lines denote
approximate property boundaries, irregular thin black lines are topographic
contours (in feet, from U.S.G.S 7.5’ Whittier Quadrangle [1981]). Dashed gray lines
show the up-dip projection of the axial surfaces as determined from petroleum
industry seismic reflection profiles. Thick purple line on the small inset map shows
the extent of the high-resolution seismic reflection line and it’s relative location to
the borehole transects. Blue line represents the location of the SCE borehole
transect, green line the Carfax Avenue borehole transect, and pink line the
Gardenland/Greenhust street borehole transect. Black dots identify the locations of
each borehole.
33
Figure 2.4, Continued
34
Figure 2.5 Borehole Results from the Carfax Avenue Transect, Gardenland-
Greenhurst Transect, and SCE Transect.
Cross-section showing major stratigraphic units (8x vertical exaggeration). Thin
vertical lines denote boreholes. Numbers show key stratigraphic units discussed in
the text. Thin red lines mark the tops of major sand and gravel filled units. Double-
headed red vertical arrows along the left side of the figure show the stratigraphic
range within which discrete uplift events occurred (see text for discussion). Solid
green lines between red arrows on left side of figure show stratigraphic intervals
that do not change thickness across the transect, reflecting periods of structural
quiescence. Proposed correlations between Carfax boreholes and a water well
(1589S) located 175 m north of the transect are shown by dashed stratigraphic
contacts along the right side of the figure. Contacts at 12, 17, and 20 m depth are
correlative between borehole 22 and the water well.
35
Figure 2.5, Continued
36 Figure 2.6 Cross-Section of Borehole Results Overlying the High-Resolution Seismic-Reflection Profile.
High-resolution seismic reflection profile from hammer source with major stratigraphic units from the
Carfax borehole transect (2x vertical exaggeration). Note strong reflectors at depths of ~10 m
(shallowest interpretable reflector, dashed blue line) and ~12 m (dashed green line) (northern end of the
transect) that dip southward. These reflectors appear to correlate well with the top and base of the
Unit 40 sand. The strength of the prominent reflector at ~12-15 m depth (dashed green line) is
probably a function of the strong acoustic impedance contrast between the overlying water-saturated
Unit 40 sand and the underlying section of thick clays and silts (Units 46 and 47).
37
augers were sampled downhole directly by a geologist to the depth of the shallowest
aquifer. Below this depth, one foot bucket drives were used for sampling. The newly
acquired Gardenland borehole transect, which consisted of seven continuously cored
hollow-stem boreholes, was drilled parallel to and ~100 m west of, the Carfax Avenue
transect (Figure 2.5). The SCE transect, which was much shorter (80 m) than the other
two transects, was collected ~50 m east of and parallel to the northern part of the Carfax
transect. The SCE transect consisted of seven, large-diameter (70 cm) bucket-auger holes
excavated to 12 meters in depth (Figure 2.5). These results are recorded in the data
repository.
2.2.2.1 Stratigraphic Observations
The Carfax, Gardenland, and SCE borehole transects reveal similar stratigraphic
sections that comprise laterally continuous, friable sands and gravels, which we interpret
as channel deposits, interbedded with silts and clays, which I interpret as overbank
deposits (Figure 2.5). More than a dozen distinctive sedimentary units can be correlated
between the borehole transects across the entire study site. The uppermost 20 meters of
the section consists of fine-to medium-grained sands one to several meters thick,
separated by centimeter- to meter-scale units composed of cohesive silt and clay. At ~20
m depth there is a marked change downward to a much coarser-grained, gravel-
dominated section.
Major sand units are denoted by intervals of 10, from youngest to oldest (e.g., 10, 20,
30, 40, and 50). Six laterally continuous units of cohesive, silty-sand to clay (units 12,
15, 25, 35, 45, and 47) have also been identified across the 450 m length of the borehole
38
transects (Figure 2.5). Specifically, these units exhibit identical stratigraphic sequences,
with each unit defined by its distinctive grain size, color, and texture. One particularly
useful marker bed is unit 46, an organic-rich, black, clay-silt that records an extended
period of marsh development over the entire site. In many of the boreholes, particularly
along the Carfax Avenue transect, Unit 46 contains numerous white micro-gastropods (<
1 mm in diameter). These gastropods are a unique and easily recognized correlative
feature across both the Carfax and Gardenland transects.
Three of the major sands units (10, 20, and 40) are both laterally continuous and
present in all three transects. Sand unit 30 also is present in all three transects, but
pinches out in the Carfax transect between boreholes 13 and 23. Locally, the sand- and
gravel-filled channels have eroded down into, and cut out parts of underlying units. One
of the most pronounced examples of this is the thinning of unit 45 southward from
borehole 13 in the Carfax transect as a result of erosion by the unit 40 channel sand. A
laterally continuous buried soil (Unit 12), is discernable along the entire length of the
Carfax Avenue transect. This soil, however, is only intermittently developed in the
Gardenland transect.
A borehole log of a water well (1589S) located ~ 175 m north of my northernmost
borehole (borehole 21) on Carfax Avenue shows there is no northward change in depth of
two major stratigraphic contacts that I can confidently identify on the driller’s log (Figure
2.5). Specifically, at a depth of 12.8 m (42 ft) in borehole 1589S, there is a sharp contact
between sands above and a bluish-gray-colored clay below. This clay section persists to
16.7 m (55 ft) depth, below which there is a thick section of coarse-grained sands and
39
Table 2.1 Radiocarbon Ages and Calibrated, Calendric Dates of Samples from the
Carfax and Gardenland Transects.
Ccl=charcoal fragment; bulk=standard bulk soil date; bulk-h=bulk soil humic date.
Calendric age is reported as 2 sigma (95% confidence limit) age range. All samples
were dated at the University of Arizona Accelerator Mass Spectrometry (AMS)
Laboratory, except two samples, 18-7.3 bulk soil and humic, which were dated by
Lawrence Livermore National Laboratories and sample 1101, which was dated by
Beta Analytic. PDF number is the individual probability distribution function for
each sample shown in Figure 2.7.
40
Table 2.1, Continued
PDF
#
Sample
#
Bore
hole
Sample
type
True
depth
(m)
Projected
depth to
bh 19 (m)
Unit
14
C
Age BP
13
C
measure
-ment
Calendric age
1 2010 18 ccl 23.5 23.7 55 9274 +71 -27.9 8650-8300BC
2 1813 18 ccl 19.5 19.5 47 8754 +72 -26 8200-7550BC
3 1903 19 ccl 20.2 20.2 47 8747 +73 -27.9 8200-7550BC
4 2108 21 ccl 16.6 21.0 47 8291 +80 -27.1 7520-7120BC
5 1808 18 ccl 17.3 18.9 47 8287 +68 -24.0 7520-7130BC
6 1401Bb 14 bulk 16.6 18.9 47 8096 +78 -25.4 7350-6750BC
7 1807 18 ccl 17.2 18.1 47 8009 +63 -24.7 7080-6690BC
8 1401Ba 14 bulk - h 16.6 18.9 47 7500+100 -27.1
6530-6100BC
9 2102 21 ccl 14.5 17.7 46 7479 +59 -24.9 6440-6230BC
10 2103 21 ccl 14.5 17.7 46 7240+280 -28.4 6700-5500BC
11 1302 13 bulk 15.3 17.9 46 7390 +79 -26.9 6420-6080BC
12 1400B 14 bulk 15.5 17.8 46 7350 +60 6370-6070BC
13 1301 13 bulk 15.3 17.9 46 7160 +81 -26.9 6230-5880BC
14 2007 20 ccl 17.0 17.5 45 7165 +59 -26.9 6210-5970BC
15 2303 23 ccl 15.5 17.4 45 7265 +69 -26.4 6260-6000BC
16 1806 18 ccl 16.5 17.4 45 6819 +61 -25.0 5840-5620BC
17 1804 18 ccl 9.6 10.7 35 5922 +51 -26.9 4940-4690BC
18 1704 17 ccl 8.4 10.1 35 6138 +62 -25.0 5230-4900BC
19 1101 11 ccl 7.5 9.5 30/35 5430 +40 4360-4220BC
20 1700 17 ccl 7.6 9.5 30/35 5375 +52 -24.8 4340-4050BC
21 2300 23 ccl 7.8 9.3 30/35 5500+110 -25.0 4600-4000BC
22 1803 18 ccl 5.9 5.8 20/25 3078 +43 -25.0 1440-1250BC
23 18-7.3h 18 bulk - h 2.2 2.0 12 990 +40 980-1160AD
24 18-
7.3bs
18 bulk 2.2 2.0 12 2650+100 1050-400BC
25 1800 18 ccl 2.1 1.7 11 372 +31 -28.9 1440-1530AD
26 10325 103 ccl 19.9 20.6 50 8,469+80 -25 7610-7320BC
27 10522 105 ccl 19.1 19.4 47 8,985+51 -29.39 8300-8160BC
or8130-
7960BC
28 10324 103 ccl 19.7 20.4 47 8,738+49 -27.45 7960-7600BC
29 10517 105 ccl 17.9 19.3 47 8,356+52 -26.16 7550-7300BC
or 7220-
7190BC
30 10515 105 ccl 16.9 18.2 46 7,958+46 -26.69 7050-6690BC
31 10319 103 ccl 17.1 17.9 46 7,265+48 -23.68 6230-6030BC
32 10412 104 ccl 17.1 16.9 45 7,076+64 -26.15 6070-5800BC
33 10315 103 bulk 16.2 16.9 45 7,052+47 -26.36 6030-5830BC
34 10512 105 ccl 15.5 16.9 45 7,050+46 -27.43 6020-5830BC
35 10413 104 ccl 17.25 16.8 45 7,793+46 -25.31 6700-6480BC
36 10410 104 ccl 13.6 13 40 7,377+48 -23.2 6380-6090BC
37 1019 101 ccl 8.4 10.5 35 6,399+43 -23.73 5480-5310BC
38 1039 103 ccl 9.3 10.4 35 5,757+66 -25.3 4770-4450BC
39 1016 101 ccl 6.9 9 25 5,665+41 -24.32 4610-4360BC
40 1014 101 ccl 4.2 6.2 20 2,076+35 -24.49 200BC-10AD
41 1024 102 ccl 3.15 3.15 15 2,108+36 -25.15 310-350BC or
40-210BC
41
Table 2.1, Continued
PDF
#
Sample
#
Bore
hole
Sample
type
True
depth
(m)
Projected
depth to
bh 19 (m)
Unit
14
C
Age BP
13
C
measure-
ment
Calendric age
42 1043 104 ccl 2.5 2.3 15 428 +29 -26.97 1420-1510AD
or 1600-
1620AD
43 1042 104 ccl 2.4 2.1 15 359 +24 -26.52 1450-1550AD
or 1550-
1640AD
42
gravels. Both of these contacts are at the same depths in my Carfax boreholes 22 and 21,
the northernmost holes in our transect.
2.2.2.2 Age Control
Age control for the Carfax and Gardenland boreholes is provided by forty-three
radiocarbon ages from charcoal fragments (36 samples) and bulk-soil samples, including
both standard (6 samples) and humic (1 sample) dates (Table 2.1). The radiocarbon ages
are from sample positions that span the depth range from 1.5 to 40 m. The ages of these
samples were calibrated using Oxcal v.3.1 [Bronk Ramsey, 1995; Bronk Ramsey, 2001
(using atmospheric data from Reimer et al., 2004); Bronk Ramsey et al., 2001], which
uses a Bayesian statistical analysis and stratigraphic ordering to calculate calendric dates
[Bronk Ramsey, 2001]. All calendric dates are reported as two-sigma (95% confidence
limit) age ranges. The radiocarbon ages of two wood fragments (samples 1809 and 1910)
acquired at ~32 m depth in Carfax boreholes 18 and 19 were much older than all other
samples, and appear to have been reworked.
Units 45, 46, and 47 contained an abundance of dateable material and therefore
have the greatest number of dated samples and the best-constrained ages. Sample ages
from both the Carfax and Gardenland transects are generally in correct stratigraphic
order, indicating minimal reworking. Table 1 shows the depth at which each sample was
collected, as well as the depth of the sample projected along bedding to its correlative
stratigraphic level at the southern and northern ends of each transect. Normalized
stratigraphic depths are based on the depths of each stratigraphic unit in borehole 19 for
the Carfax Avenue transect and borehole 102 for the Gardenland transect.
43
The age data demonstrate that sediment accumulation has been relatively constant
over the past 10,000 years at a rate of ~2 mm/yr (Figure 2.7). The sediment
accumulation rate curves for both the Carfax and Gardenland transects show remarkably
similar depth-versus-age relationships throughout Holocene time, demonstrating that
deposition was continuous across the site, as might be expected in this low-gradient,
distal floodplain setting. Moreover, the numerous radiocarbon dates from correlative
stratigraphic units in both transects indicate that these units were deposited
synchronously across the site. There is no evidence for any discernible time-
transgressive deposition across the site, except for the development of the Unit 12
paleosol in the Carfax transect (see discussion below). At ~2-3 ka the two sediment
accumulation rate curves diverge slightly, with faster sediment accumulation along the
Carfax Avenue transect (Figure 2.7). This is followed by a brief hiatus in sediment
accumulation between ~1000 to 3000 years ago. During this same period, sediment
accumulation was more continuous along the Gardenland transect.
2.2.2.3 Evidence for Folding and Sedimentary Growth
Several observations reveal the details of recent folding of sediments beneath the
study site. Most basically, the tops of all of the sedimentary units are deeper at the
southern end of the transect than at the northern end, with the structural relief of horizons
increasing with depth (Figure 2.5). This relief reflects southward thickening of
sedimentary sequences that also increases with depth. The sedimentary thickening (or
“growth”) is localized in an upward-narrowing zone located at the updip projection of the
active axial surface imaged on the petroleum industry, mini-sosie, and hammer seismic
44
reflection data (Figure 2.3) [Pratt et al., 2002; Shaw et al., 2002]. The zone of south-
tilted strata is ~250 m wide at 15 m depth, whereas it is less than 145 m wide at 2.5 m
depth. Strata to the north and south of the zone of folding are nearly flat, [Shaw and
Shearer, 1999; Pratt et al., 2002; Shaw et al., 2002]. Finally, the southward sedimentary
thickening is restricted to discrete stratigraphic intervals that are separated by intervals in
which the sediments do not change thickness along the length of the transect.
A key consideration for interpreting the deformation in this low-gradient fluvial
floodplain setting is that although cohesive silts and clays might partially drape any
existing topographic slopes (e. g., fold scarps), sands and gravels will not do so [e.g.,
Blatt et al., 1980]. Rather, the sands and gravels will buttress, or onlap, any existing
topography. Moreover, the tops of any sand- or gravel-filled channels will be deposited
at the local stream gradient. The pre-development stream gradient of the site and the
regional slope are both extremely gentle (0.15°). Outside of the zone of active folding,
the stream gradient is likely to have been this gentle throughout the Holocene, given the
nearly-flat regional topography and sub-horizontal reflectors observed to the north and
south of the zone of active folding [Shaw and Shearer, 1999; Shaw et al., 2002; Pratt et
al., 2002]. The tops of the laterally extensive sand and gravel channels exposed at the
study site thus provide paleo-near-horizontal indicators.
45
Figure 2.7 Sediment Accumulation Rate Curve for the Carfax and Gardenland
Transects.
Sediment accumulation curve for Carfax (gray) and Gardenland (green) transects
with results of OxCal v.3.1 (Bronk Ramsey, 1995; Bronk Ramsey, 2001 [using
atmospheric data from Reimer et al., 2004]; Bronk Ramsey et al., 2001) analyses of
calendric radiocarbon dates sorted by depth for each transect. True depth has been
corrected for folding in individual events by moving sample up or down relative to
the nearest stratigraphic horizon. Depths were projected to borehole 19 in the
Carfax Transect and borehole 102 in the Gardenland Transect. Probability density
functions are black for the Carfax Transect, red for the Gardenland Transect and
pink for samples that appear to have been reworked. Numbers next to probability
density functions represent sample numbers in Table 2.1.
46
Figure 2.7, Continued
47
2.2.3 Structural Reconstructions
2.2.3.1 Methods
By unfolding the strata exposed in the boreholes to their near-horizontal
depositional geometry [Novoa et al., 2000], I have reconstructed the evolution of the
folded sediments. This process, known as inclined shear restoration, is a general method
of sequentially restoring growth folds that does not imply a specific mechanism of fold
growth. Rather, inclined shear methods can resolve growth histories of folds that develop
by kink-band migration or limb-rotations mechanisms, or combinations of these and
members [Novoa et al., 2000]. The critical points are that the restorations are sequential
(providing path, not just initial and final states), and that the balance of folding
mechanisms are specified by the geometries of the folded growth horizons, which
generally distinguish between kink-band migration and limb rotation mechanisms. The
various fault-related folding theories (e.g., fault-bend folding, trishear) all specify the
kinematics of fold growth (i.e., if the folds grow by kink-band migration, limb-rotation,
or some specific combination of these). Thus, inclined shear restorations, by determining
folding mechanisms, are considered a means of determining what fault-related folding
theories might be most apt at describing a structure. These reconstructions also allow
incremental measurements of sediment growth and the development of structural relief
across the fold in three-dimensions (Figures 2.8 and 2.9). In these restorations, I use the
63
o
N dip of the active axial surface revealed by seismic reflection data as the direction of
shear (Figure 2.3A; [Shaw et al., 2002]).
48
2.2.3.2 Shear Restoration of Folding Events
The shear restorations for both the Carfax Avenue (Figure 2.8) and Gardenland/
Greenhurst Street transects (Figure 2.9) reveal remarkably similar structural histories.
Specifically, restoration of both transects demonstrates that there have been three
stratigraphically discrete uplift events during the past ~8000 years, each of which were
followed by sedimentary growth and subsequent restoration of the regional, near-flat
stream gradient. These uplift events, labeled Y, X, and W, from youngest to oldest
(following the terminology of Dolan et al., 2003), occurred between 0.2 to 2.1 ka, 3.0 to
5.8 ka, and 6.6 to 8.1 ka, respectively.
In such a situation, when a fold scarp forms in response to slip on the blind-thrust
fault at depth, the river will immediately begin to adjust its bed in order to return to the
pre-earthquake, equilibrium stream gradient. This may be achieved by either deposition
of sediment on the downstream side of the scarp, leading to growth of a thickened
sedimentary package until the pre-folded stream gradient is re-established, and/or erosion
of the upstream, upthrown side of the scarp. Thus, our uplift measurements in these
events are minima, as they do not account for any possible erosion that might have
occurred on the uplifted block as the stream was re-establishing its gradient. Also, any
compaction of fine-grained sediments on the downstream, downthrown side of the scarp,
although likely to be very minor at these shallow depths, would make our uplift
measurements minima.
As defined by the minimum paleoscarp heights, uplift in Event Y was ≥ 1.1 m
+0.2 in the Carfax transect and ≥ 1.6 m + 0.4 in the Gardenland transect. For Event X,
49
Figure 2.8 Structural Reconstruction of Uplift Events from the Carfax Avenue
Transect.
Sequential reconstructions (A through E; from youngest to oldest) of fold growth
during uplift Events Y, X, and W in the Carfax Avenue Transect (8x vertical
exaggeration). Boreholes shown as thin, black vertical lines. The major
stratigraphic horizons have been sequentially restored to the paleo-steam gradient
in the direction of shear using the methodology of Novoa et al. (2000) (Figure 2.5C)
(eight times vertical exaggeration). Anticlinal axial surfaces are represented by
inclined green lines and synclinal axial surfaces are represented by inclined yellow
lines for each event. Note that the apparent steepness of the axial surfaces is due to
the vertical exaggeration. The actual dip of the axial surface used in our
reconstruction is 63
o
, which was measured directly from industry and high-
resolution seismic reflection data (Shaw et al., 2002; Pratt et al., 2002). See text for
discussion. We used the tops of several major sand units as restoration horizons, as
the intervals overlying these sands do not change thickness along the transect,
indicating deposition at the equilibrium stream gradient. The scoured base of sand
units is not an appropriate indicator of paleo-stream gradient due to the erosional
nature of these contacts. The graph above each reconstruction shows increasing
sedimentary thickness from north to south (blue line) versus increasing depth of top
of major stratigraphic horizons (pink line). (A) Present day configuration of major
stratigraphic units. Gray numbers denote boreholes. Sedimentary thickening from
north to south occurs within Units 10 and 11 (blue line), indicating that the
50
Figure 2.8, Continued
sedimentary growth within these units records onlap of a now-buried, south-facing
fold scarp that developed during Event Y. This folding event occurred sometime
after the deposition of Unit 12 and before the completion of deposition of Unit 10.
Pink line represents the increase in depth of the top of Unit 12 to the south. (B)
Restoration of Event Y. Based on the results of Dolan et al., (2003) the top of Unit
12 has been restored to the paleo-stream gradient. The lateral continuity and lack
of growth within Units 12 and 15 (blue line) indicate that these sediments were
deposited at the near-horizontal stream gradient and record a period of structural
quiescence. The top of Unit 20 does not increase in depth to the south (pink line).
(C) Top of Unit 20 has been restored to the paleo-stream gradient. Sedimentary
growth occurs within Units 20 and 25 (blue line), indicating that these laterally
continuous units were also deposited above a south-facing paleo-scarp that formed
during Event X, after deposition of Unit 30 and before the completion of deposition
of Unit 20. Pink line represents the increase in depth of the top of Unit 30 to the
south. (D) Restoration of Event X. The absence of growth within the Unit 30 to 35
section (blue line) supports our inference that the top of the Unit 40 channel sand
was deposited at the near-horizontal paleo-stream gradient (pink line). (E) Top of
Unit 40 has been restored to the paleo-stream gradient. The Unit 40/45 interval
records onlap of an additional south-facing fold-scarp that developed during Event
W. Event W occurred sometime after the deposition of Unit 46 and before the
completion of deposition of Unit 40. Blue line represents the southward thickening
51
Figure 2.8, Continued
of Units 40/45 and pink line represents the increase in depth of the top of Unit 46 to
the south. (F) Restoration of Event W. Top of Unit 46 has been restored to the
paleo-stream gradient. The coarse grain size and lateral continuity of Unit 50 (pink
line) suggest that the top of this gravel channel was deposited at the near horizontal
paleo-stream gradient. This inference is supported by the laterally constant
thickness of Units 46 and 47 (blue line), indicating that these sediments were
deposited nearly horizontally. Units 46 and 47 record a period of structural
quiescence during which no discernible uplift occurred.
52
Figure 2.8, Continued
53
Figure 2.9 Structural Reconstruction of Uplift Events from the Gardenland Avenue
Transect.
Sequential reconstructions (A through E; from youngest to oldest) of fold growth
during uplift Events Y, X, and W in the Gardenland Avenue/Greenhurst Street
transect (Figure 2.5A). Same format as figure 2.8. (A) Present-day configuration of
major stratigraphic units. Sedimentary thickening from north to south occurs
within Unit 10 (pink line), indicating that the growth within these units records
onlap of a now-buried, south-facing fold scarp that developed during Event Y,
sometime after the deposition of Unit 15 and before the completion of deposition of
Unit 10. The Unit 12 paleosol is only intermittently developed in the Gardenland
transect and therefore cannot be used as a restoration horizon. Pink line represents
the southward increase in depth of the top of Unit 30. (B) (i) and (ii) demonstrate
that almost all growth above the Unit 20 sands occurs within the Unit 10 sand (top
blue line), with no evidence for consistent southward thickening in Unit 12 – 15
(bottom blue line). This is consistent with the observations in the Carfax Transect,
where Dolan et al. (2003) interpret the ground surface during Event Y at the Unit 12
paleosol. (C) through (F) same as in figure 2.8.
54
Figure 2.9, Continued
55
minimum uplift was ≥ 1.1 m + 0.2 in the Carfax transect and ≥ 1.1 m + 0.1 in the
Gardenland transect. For Event W, minimum uplift measured in the Carfax and
Gardenland transects was ≥ 1.6 m + 0.4 and ≥ 1.3 m + 0.3, respectively. Our estimate of
uplift for Events Y, X, and W differ slightly from those published by Dolan et al. [2003]
because of my re-evaluation of, and slight adjustments to, stratigraphic contacts made in
light of our analysis of the newly acquired Gardenland borehole transect. The intervals
of sedimentary growth following these three uplift events are separated from one another
by stratigraphic intervals that do not exhibit changes in thickness across the transects.
Specifically, units 12 - 15, units 30 – 35, and units 46 – 47 do not exhibit any systematic
thickness changes along the length of the transects (Figures 8 and 9). I interpret these
intervals as recording periods of structural quiescence characterized by deposition at the
regional, near-horizontal stream gradient.
2.2.3.3 Location, Geometry, Width, and Evolution of the Forelimb Kink Band
Based on the shear restorations of the Carfax and Gardenland boreholes, we have
defined, for each uplift event, the active anticlinal and inactive synclinal axial surface
locations, as well as the width and dip of the dip panel for the growth triangle. In figures
8 and 9 we also show, at the same scale, the amount of sedimentary growth of each
horizon. The units used to define sedimentary growth for Events X and W are consistent
between the Carfax and Gardenland transects. They differ, however, for Event Y, the
most recent event, because of the different degree of development of paleosol Unit 12 in
the two transects. In the Carfax transect, the Unit 12 paleosol is well-developed and
traceable across the entire length of the transect. This allowed Dolan et al. [2003] to
56
provide precise constraints on the exact depth of the Event Y horizon. Specifically,
because there is no growth along the Carfax transect between the top of Unit 20 and the
top of Unit 12, they inferred that Event Y occurred after stabilization of the Unit 12
paleo-surface. Unfortunately, as noted above, the Unit 12 paleosol is either intermittently
developed or poorly preserved in the Gardenland transect and therefore cannot be used in
the event reconstructions. Rather, I have used the next shallowest horizon - the top of the
Unit 20 sand - as the basis for our reconstruction of uplift during Event Y. In order to
more precisely establish where most post-Unit 20 sedimentary growth has occurred in the
Gardenland Transect we also determine the amount of growth within Unit 10, the most
recent sand, and the growth within Unit 12/15 fine-grained interval. Because of the
intermittent development of Unit 12 in the Gardenland transect, this paleo-ground surface
cannot be used to restore the top of Unit 20 for Event Y.
The anticlinal axial surface for the Event W is located between boreholes 24 and
15 in the Carfax transect, with a synclinal axial surface located near borehole 20, yielding
a growth triangle width of ≤ 161 m (Figure 2.8). Most growth for this transect, however,
appears to be within a zone ≤137 m wide between boreholes 15 and 20. For the
Gardenland transect, the anticlinal axial surface for Event W appears to be located at, or
north of, my northernmost borehole (100B), with all the sedimentary growth within our
transect occurring between boreholes 100 and 105 (Figure 2.9). This dip panel is thus at
least 86 m wide. However, because I may not have captured the entire width of the Event
W fold scarp in our Gardenland transect, the dip panel may be wider than 86 m in this
57
location. This may account for the differences in minimum scarp height between the
Carfax and Gardenland transects (1.6 m + 0.4 and 1.1+ 0.2 m respectively).
For Event X, the active anticlinal axial surface is located at, or just south of,
borehole 20 in the Carfax transect (Figure 2.8), and borehole 103 in the Gardenland
transect (Figure 2.9). Due to the lack of continuity of unit 30 in boreholes 15, 14, and 20
in the Carfax transect, and in borehole 101 in the Gardenland transect, as well as the
location of growth with units 20 and 25, Event X is not well constrained when calculated
from unit 20 – 25 growth. This is due to the fact that Units 25 and 30 have locally been
eroded in both the Carfax and Gardenland transects. For Event X in the Carfax transect,
growth within units 20 and 25 extends southward at least as far as the southernmost
borehole (25). Thickening also extends south of the southern-most borehole in
Gardenland (104) for this event. The kink band for Event X is thus between ~180 and
195 m wide in both the Carfax and Gardenland transects. Shear reconstruction of unit 20,
however, indicates that the paleo-scarp defined by the Unit 30 and Unit 40 sands was
located at, or near, borehole 20 in the Carfax transect and borehole 103 in the Gardenland
transect. Most growth, therefore, occurred within a narrow (<72 m wide) zone in the
Carfax transect between boreholes 20 and 19.
The active anticlinal axial surface for Event Y is located just south of borehole 14
in the Carfax transect, with a dip panel width of ~190 m (Figure 2.8). Most of the
growth, however, appears to be localized between boreholes 14 and 18, a zone ~65 m
wide. In the Gardenland transect, the active anticlinal axial surface is located at, or south
of, borehole 101 (Figure 2.9). Although minor growth appears to continue south of our
58
southernmost borehole (104), most growth occurred between boreholes 101 and 103, a
distance of ≤ 127 m.
The shear restorations indicate that the axial surfaces for Events Y, X, and W do
not occur in the exact same location from event to event. Rather, the locus of folding
migrates between individual uplift events. The anticlinal axial surface for Event W is
located ~225 meters north of the southernmost borehole (25) in the Carfax transect, close
to the location of the anticlinal axial surface for the most recent event, Event Y, and at
least 300 m north and possibly even further north in the Gardenland transect (104) if we
have not captured the entire scarp. In Event X, the growth of the section between the top
of Unit 20 and the top of Unit 30 has a northern limit at borehole 20 in the Carfax
transect and borehole 103 in the Carfax transect. This is ~160 meters south of the
anticlinal axial surface for the subsequent event in the Carfax transect and ~130 meters
south in the Gardenland transect. As noted above, it is possible that I have not captured
the entire width of the dip panel for this event, as thickening occurs at least as far south as
our southernmost boreholes and may continue beyond this location.
The active anticlinal axial surface for Event Y is ~65 m south of the northernmost
boreholes (21/22) in the Carfax transect and ~85 m south of the northernmost borehole
(100A/100B) in the Gardenland transect. Dolan et al. [2003], in their description of the
Carfax Avenue borehole results, noted an older event, which they termed Event V. Event
V was not observed in the Gardenland boreholes because those boreholes did not extend
deep enough to document this stratigraphic interval. It is worth noting, however, that in
59
the Carfax transect, the active anticlinal axial surface for Event V lies almost directly
beneath the active axial surface for Event W.
2.2.3.4 Incremental Development of Dips within the Kink Band
The combined borehole and seismic reflection data indicate that strata folded
within the kink band acquire their dips incrementally. For example, analysis of deep
penetration petroleum industry seismic reflection data indicates that all the strata in the
upward-narrowing forelimb growth triangle dip ~23° between the top of the thrust ramp
at ~3 km and the shallowest imaged reflectors at 250 m (Figure 2.3A; [Shaw and
Shearer, 1999; Shaw et al., 2002]). In contrast, the high-resolution reflection data reveal
a progressive decrease upwards in dip at depths of 200 to ~90 m. Specifically, prominent
reflectors in the Mini-Sosie profile dip ~17° at 200 m, ~ 16° at 110 meters and 8° at 90
meters. The sharp change in dip between 110 to 90 m depth in the Mini-Sosie profile
appears to be an unconformity, and suggests that a significant amount of the section is
missing, probably as a result of a hiatus in deposition [Pratt et al., 2002].
This incremental acquisition of dip extends to the scale of individual earthquakes,
as revealed by the structural reconstructions of events Y, X, and W. Specifically, these
data show that the strata in individual buried fold scarps dip only ~0.7° for unit 50 and ~
0.6° to 0.4
o
for units 40, 30, and 20. Following Event Y, the most recent event, unit 20
acquired a dip of ~0.4
o
, while units 46, 47, and 50 dip ~0.7
o
, almost doubling the dip of
strata observed at this site. After Event X, units 46 and 47 acquired a southward dip of
~0.6
o
, at which time units 30 and 40 acquired a dip of ~ 0.5°. Following Event W, units
46 and 47 acquired a southward dip of ~ 0.7° Dips calculated from the cross sectional
60
geometry of fold scarps along both the Carfax and Gardenland transects are gentle but, at
a depth of 20 m, are about four times greater than the near-horizontal, present-day
topographic gradient at the study site (~0.15
o
along the Carfax transect and ~0.14
o
along
the Gardenland transect).
2.3 Discussion
The borehole and seismic reflection data from the Carfax site provide continuous
imagery of the forelimb growth triangle associated with slip on the Santa Fe Springs
segment of the PHT, from the top of the thrust ramp at 3 km depth to the surface. This
combined data set demonstrates that folding is localized within a discrete kink band that
narrows upward to a width of <150 m at 2.5 m depth, with most folding at that depth
occurring within a zone only ~60 m wide during the most recent event. Moreover, these
complete images of the growth triangle demonstrate that folding is continuous across the
active axial surface associated with the tip of the thrust ramp, thus obviating the
possibility that localized deformation is related to secondary faults developed at stress
concentrations at the ramp tipline [e. g., Lin and Stein, 1989]. These observations
confirm that the forelimb of the Santa Fe Springs anticline is growing, at least in part, by
kink-band migration in a manner broadly consistent with growth fault-bend fold theory
[Suppe et al., 1992; Shaw and Suppe, 1994; Shaw et al., 2002]. Kink-band migration is
implied by the localization of deformation along the fold’s anticlinal axial surface, and
the upward narrowing of the fold limb within the syntectonic growth section. This
evidence for discrete fold growth above a blind thrust ramp differs from the broad and
diffuse patterns of deformation observed geodetically and modeled by dislocations in
61
elastic half-space for several moderate-magnitude blind thrust earthquakes, including the
1987 M
w
6.0 Whittier Narrows earthquake and the 1983 Coalinga (M 6.5) earthquake
[Stein and King, 1984; Lin and Stein, 1989; Stein and Ekström, 1992]. In contrast,
discrete folding has been observed to occur co-seismically, specifically as a result of
synclinal fault-bend folding associated with the 1999 Chi Chi (M
w
7.7) event in Taiwan
[e.g., Chen et al., 2007]. This implies that the amount of discrete versus distributed
deformation may vary for different types of blind-thrust systems, or perhaps as functions
of earthquake magnitude, patterns of co-seismic slip, and fault geometry.
The similarity of stratigraphic and structural observations and age control from
the new Gardenland borehole data and the earlier Carfax Avenue borehole results
presented in Dolan et al. [2003] demonstrate the reproducibility of the results.
Specifically, these two borehole transects provide repeatable measurements of scarp
height, depth and age of uplift events, and the location of anticlinal active and synclinal
inactive axial surfaces for each uplift event. Moreover, the three-dimensional control on
the sub-surface geometry of buried fold scarps provided by these combined data sets
demonstrates that the buried scarps are of tectonic, rather than fluvial origin. For
example, the anticlinal active axial surfaces for both Events Y and X strike approximately
west to west-southwest, sub-parallel to the regional strike of the axial surface as defined
by the deep industry seismic reflection profiles [Shaw et al., 2002], but nearly
perpendicular to the southward-flowing San Gabriel River adjacent to the study site. The
strike of Event W is less well defined because we may not have captured the
62
northernmost part of the fold on the Gardenland transect, but it appears to strike
northwesterly.
2.3.1 Earthquake by Earthquake Fold Growth
I infer that the three stratigraphically discrete uplift events revealed by the
incremental shear restorations each represents deformation in a single, large-magnitude
earthquake on the PHT. It is also possible, however, that each of these uplift events
comprises more than one earthquake. If each uplift event does record multiple, smaller-
magnitude earthquakes, then these must have occurred during relatively brief clusters, on
the order of ≤1500 years for Event W (and for Event V, as reported in Dolan et al.,
2003). I consider this possibility unlikely, and favor the single-event interpretation of
each of the stratigraphically discrete uplift events, particularly given the observation that
discrete co-seismic folding generally does not occur in moderate magnitude blind-thrust
earthquakes, but rather is documented in very large earthquakes that rupture to fault tips
and across fault bends [e.g., Chen et al., 2007]. Moreover, the temporal discreteness of
these uplift events obviates the possibility that the folds grow in response to quasi-
continuous fault creep on the PHT ramp. I cannot, however, rule out the possibility that
some component of folding occurs during punctuated periods of aseismic fault slip, either
as afterslip following large earthquakes or during discrete periods of interseismic fold
growth. But given the occurrence of the 1987 (M
w
6.0) Whittier Narrows earthquake on
the Santa Fe Springs segment of the PHT, coupled with the absence of any geodetic or
geological evidence of interseismic fault creep, we think that this last possibility is
unlikely.
63
I estimate minimum displacements on the Santa Fe Springs segment thrust ramp
in each of these earthquakes by dividing the minimum measurements of uplift by the sine
of the well-constrained 27°±2° north dip of the Santa Fe Springs Segment thrust ramp
documented by Shaw and Shearer [1999] and Shaw et al. [2002]. Recent discrete
element models demonstrate that uplift provides a robust proxy for fault slip at depth
[Benesh et al., 2007]. This methodology assumes that no folding occurs outside of the
narrow kink band documented in the borehole results. This observation implies that all
estimates of uplift and fault displacement are minima. As noted above, however, analysis
of seismic reflection profiles suggest that most (>82% at 250 depth) and perhaps almost
all, fold growth does occur within the kink band. Thus, my minimum estimates are
probably close to the actual values for folding at the study site. Complicating this result
is the fact that the Carfax site lies near the western end of the central, Santa Fe Springs
segment of the PHT, within a zone where slip at depth is partially transferred westward
onto the Los Angeles segment of the PHT. Shaw et al. [2002] used calculations of total
cumulative displacement on the Santa Fe Springs segment based on kink band width and
structural relief of the base Quaternary reflector observed in industry seismic profiles to
show that cumulative slip at the Carfax site is only 65% of the maximum cumulative slip
near the center of the Santa Fe Springs segment. Moreover, the south-dipping
Montebello Hills back thrust partitions off some slip from the deep PHT ramp beneath
the Carfax site. Thus, in a long-term sense, our slip estimates from the borehole data
probably record ≤65% of total long-term slip on the deep part of the Santa Fe Springs
segment thrust ramp.
64
2.3.2 Using Paleo-fold Scarp Geometry to Estimate Paleo-magnitudes
I use our estimates of displacement per event to determine a conservative range of
paleo-magnitudes based on the displacement-versus-moment magnitude (M
w
) regressions
of Wells and Coppersmith [1994; Table 2.2]. I used both average displacement-versus-
M
w
(M
w
= 6.93+0.82*log [average displacement]) and maximum displacement-versus-
M
w
regressions (M
w
= 6.69+0.74*log [maximum displacement]) to estimate a range of
possible magnitudes of paleoearthquakes at the Carfax site. The assumption that the
minimum displacements that we measure represent average displacements yields paleo-
magnitudes of ~7.3 + 0.1 for Event Y, ~7.2 + 0.1 for Event X, and ~7.35 + 0.1 for Event
W. If we assume that our minimum displacement estimates record maximum
displacement during each of these three earthquakes, the corresponding paleo-magnitude
estimates are ~7.05 + 0.1 for Event Y, ~7.0 + 0.1 for Event X, and ~7.05 + 0.1 for Event
W. I note, however, that it is unlikely that all of our measurements represent maximum
displacements along the Santa Fe Springs segment of the PHT, given that only 65% of
the total fault slip is manifested as folding at our study site.
2.3.3 Implications for Seismic Hazard in Southern California
The occurrence of large-magnitude earthquakes on the PHT has obvious and
significant implications for seismic hazard assessment, given the location of this fault
directly beneath the heart of metropolitan Los Angeles. Indeed, recent loss estimates for
large PHT earthquakes based on the findings of Dolan et al. [2003] and Shaw et al.
[2002] suggest that the occurrence of such an event would represent one of the worst
seismic disasters in the United States [Field et al., 2005]. Of particular concern is the
65
possibility that any up-dip source directivity would funnel energy directly into the 10-km-
deep Los Angeles basin. Moreover, such large-magnitude earthquakes on the PHT would
involve permanent displacements of several meters directly beneath the downtown Los
Angeles high rise district. It is worth reiterating, however, that my data indicate that such
large-magnitude earthquakes recur infrequently, with inter-event times measurable in
thousands, rather than hundreds, of years. The cumulative minimum displacement of 8.8
m + 1.8 m measured in the three uplift events, combined with the 6.6-8.1 ka age of the
oldest event (Event W), provides a minimum Holocene slip rate for the Santa Fe Springs
segment of the PHT of between 0.9 – 1.6 mm/yr. This minimum slip rate is similar to the
minimum 1.1 – 1.6 mm/yr minimum rate determined from a slightly longer record of
earthquakes published earlier for the Carfax transect, which included an additional, older
earthquake (Event V) [Dolan et al., 2003]. These minimum slip rates suggest that the
Puente Hills Thrust fault accommodates at least 20-35% of the 4-5 mm/yr of north-south,
interseismic shortening across the northern metropolitan region measured by geodesy
[e.g., Walls et al., 1998; Argus et al., 1999; 2005; Bawden et al., 2001]. If I consider the
fact that, in a long-term sense, only 65% of the slip on the central part of the Santa Fe
Springs segment is manifested by folding at the Carfax site, then a likely minimum slip
rate for the segment of ~1.4-2.4 mm/yr can be derived by multiplying our minimum rates
by a factor of 1.5. These revised rate estimates would suggest that as much as 30-60% of
the total current interseismic north-south shortening is accommodated by slip on the
PHT.
66
Table 2.2 Uplift Amounts, Age Limits, and Paleo-magnitudes for the Santa Fe
Springs Segment of the Puente Hills Blind-Thrust Fault
Event Age (ka) Growth section Uplift (m) Slip (m) M
w
Y
0.2 – 3.0
0.0 – 2.1
Surface to top of Unit
12 (Carfax)
Surface to top of Unit
15 (Gardenland)
1.1 + 0.2
1.6 + 0.4
2.4 + 0.5
3.5 + 0.9
7.2 + 0.2
7.4 + 0.1
X
3.0 – 6.3
2.1 – 5.8
Units 20 and
Carfax
Gardenland
1.1 + 0.2
1.1 + 0.1
2.4 + 0.4
2.4 + 0.3
7.2 + 0.1
7.2 + 0.1
W
6.6 – 8.2
6.4 – 8.1
Top of Unit 40 to top
of Unit 46
Carfax
Gardenland
1.6 + 0.2
1.3 + 0.3
3.5 + 0.9
2.9 + 0.6
7.4 + 0.1
7.3 + 0.1
Uplift amounts, age limits, and estimated moment magnitudes (M
w
) for
paleoearthquakes W through Y, as determined from Carfax and Gardenland
borehole results. See text for discussion. Preferred M
w
assumes that the measured
slip was the average slip during paleoearthquakes, using regressions of Wells and
Coppersmith (1994).
67
2.3.4 Incremental Structural Evolution of Fault-related Folds
The borehole data described above document the details of folding within the
upward tip of the forelimb growth triangle, allowing us to determine the incremental,
earthquake-by-earthquake development of the fold. Two basic observations stand out.
First, the folded strata within the forelimb growth triangle acquire their dip progressively
through time, such that deeper strata, which have experienced more PHT earthquakes, dip
more steeply than younger, less deformed strata near the top of the growth triangle.
Second, the active axial surface (locus of active folding) in each folding event does not
occur in the same location from earthquake to earthquake. Rather, the axial surface
migrates through time by as much as several hundred meters. This migration may simply
reflect a component of kink-band migration [Suppe et al., 1992], or more complex
processes related to changes in dip and width of the axial surface between different
earthquakes.
The progressive downward increase in bed dip genuinely reflects an aspect of fold
kinematics distinct from classic growth fault-bend fold theories. This change in limb dips
may reflect some component of fold growth by limb rotation in addition to kink-band
migration. These hybrid kinematics are manifest in several types of fault-related folds,
including trishear fault-propagation folding [Erslev, 1991; Allmendinger, 1998] and shear
fault-bend folding [Suppe and Connors, 2004; Shaw et al., 2005]. Alternatively, the
component of limb rotation may reflect curvature of the fold hinge, as is described by
curved-hinge fault-bend folding theories (Figure 2.10) [Suppe et al., 1997; Novoa et al.,
2000]. Finally, the progressive change in bed dips may reflect the mechanical response
68
of loosely consolidated, granular sediments in the shallow subsurface to folding at depth.
Recent work by Benesh et al., [2007] demonstrates that, even in cases of folding of pre-
tectonic layers exclusively by kink-band migration, shallow growth sediments acquire
their dips progressively. This is due to the finite width of the axial surface zone and its
rotation and migration with increasing deformation. In any event, the spatially and
temporally discrete folding observed in our subsurface excavations demonstrates that
blind-thrust faulting can leave a discernable record of paleo-earthquakes, and that folding
occurs predominantly by some discrete deformation mechanism (or mechanisms). Fold
kinematics may involve components of kink-band migration and perhaps limb rotation,
the latter a reflection of the finite width of the axial surfaces and fold hinge, as well as the
governing deformation mechanism of loosely consolidated near-surface sediments. In
general, this means that subsurface growth fold geometries, folding mechanisms, and
paleo-earthquake histories are best constrained by observations at a range of scales, from
the scale of entire thrust fault-fold system provided by seismicity and petroleum-industry
seismic reflection and well data, through the intermediate scale of fold growth within the
upper few hundred meters observed on high-resolution seismic data, to the scale of
individual earthquakes revealed by detailed paleoseismological investigations.
2.4 Conclusions
Continuously cored boreholes and seismic reflection data provide a complete,
overlapping record of folding above the central Santa Fe Springs segment of the Puente
Hills blind thrust fault, from the top of the thrust ramp at 3 km depth to the surface. These
data demonstrate that folding is continuous at all scales, obviating the possibility that
69
secondary faults contribute significantly to deformation in the forelimb of the Santa Fe
Springs anticline. Moreover, the borehole and reflection observations show that most
folding (>82% at 250 m depth) occurs within a discrete, upward-narrowing kink band
delimited by active (anticlinal) and inactive (synclinal) axial surfaces. The incremental
acquisition of dip within the growth triangle may reflect some combination of fold
growth by limb rotation in addition to kink-band migration. Possible mechanisms that
could produce these hybrid kinematics include tri-shear fault-propagation folding [Erslev,
1991; Allmendinger, 1998] and shear fault-bend folding [Suppe et al., 2004; Shaw et al.,
2005]. Alternatively, the component of limb rotation may result from curved-hinge fault-
bend folding [Suppe et al., 1992; Novoa et al., 2000], and/or the mechanical response of
loosely consolidated, granular sediments in the shallow subsurface to folding at depth.
The buried fold scarps revealed by the three-dimensional network of boreholes
collected across the locus of active folding record incremental fold growth during PHT
earthquakes. The new borehole data from the Gardenland transect, in conjunction with
data collected earlier from the parallel Carfax transect 100 m to the east, demonstrate the
reproducibility of measurements of scarp height, depth and age of each uplift event, and
the location of anticlinal active and synclinal inactive axial surfaces for each uplift event.
This reproducibility indicates that paleoseismological data can be obtained from blind
thrust faults with a precision similar to that of surface rupturing faults. The large
displacements that I infer for each of the three scarp-forming earthquakes are consistent
with large-magnitude (Mw >7) ruptures of the PHT ramp, which probably involved
70
Figure 2.10 Fault-Bend Fold Models.
Fault-bend fold models illustrating the patterns of growth structures that develop
with angular (A) and curved (B) fold hinges. The model with an angular fault bend
and discrete axial surface yields an upward-narrowing fold limb of constant dip in
growth strata (Suppe et al., 1992). In this model, growth folding occurs
instantaneously as strata pass through the anticlinal active axial surface. In the
curved hinge model (after Suppe et al., 1997), growth strata acquire their dips
progressively as they pass through an active axial surface zone of finite width,
bounded by active entry (a) and exit (a') axial surfaces. The passive, inactive axial
surface zone bounded by axial surfaces b and b' marks the strata that were initially
deposited along the active, axial surface and that have been subsequently translated
above the thrust ramp by fault displacement. Colors designed to schematically
represent stratigraphic units shown in figure 2.5.
71
rupture of the entire PHT. Recurrence of such a large earthquake directly beneath the
heart of the Los Angeles metropolitan region would generate enormous damage. Indeed,
loss estimates in such events by Field et al. [2005] indicate that a major Puente Hills
blind thrust earthquake may represent, in a deterministic sense, potentially the worst
seismic disaster in United States history.
72
Chapter 2 References
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Argus, D. F., Heflin, M. B., Donnellan, A., Webb, F. H., Dong, D., Hurst, K. J.,
Jefferson, D. C., Lyzenga, G. A., Watkins, M. M., and Zumberge, J. F. (1999),
Shortening and thickening of metropolitan Los Angeles measured and inferred by
using geodesy: Geology, v. 27, n. 8, 703 – 706.
Argus, D. F., Heflin, M. B., Peltzer, G., and Crampe´, F. (2005), Interseismic strain
accumulation and anthropogenic motion in metropolitan Los Angeles, J. Geophys.
Res., 110, B04401, doi: 10.1029/2003JB002934.
Barbier, M. G. (1983), The mini-sosie method: International Human Resources
Development Corporation, Boston, MA, 186 pp.
Bawden, G. W., Thatcher, W., Stein, R. S., Hudnut, K. W., and Peltzer, G. (2001),
Tectonic contraction across Los Angeles after removal of groundwater pumping
effects: Nature, 412, 712-815.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
73
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Field, E. H., H. A. Seligson, N. Gupta, V. Gupta, T. H. Jordan, and K. W. Campbell
(2005), Loss estimates for a Puente Hills blind-thrust earthquake in Los Angeles,
California, Earthquake Spectra, 21, 329-338.
Field., E. H. (2005), Collaborative SCEC/USGS efforts to improve seismic-hazard analysis;
RELM and OpenSHA: Proceedings of the 5th U.S.-Japan natural resources meeting and
Parkfield, California fieldtrip, Open-File Report - U. S. Geological Survey, Report: OF
2005-1131, pp. 18. Resource Location: http://pubs.usgs.gov/of/2005/1131/
Hudnut, K. W., Shen, Z., Murray, M., McClusky, S., King, R., Herring, T., Hager, B., Feng,
Y., Fang, P., Donnellan, A., and Bock, Y.(1996), Co-seismic displacements of the 1994
Northridge, California, earthquake: Bull. Seismol. Soc. Am., v. 86, No. 1B, p. S19 – S36.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Pratt, T. L., J. H. Shaw, J. F. Dolan, S. A. Christofferson, R. A. Williams, J. K. Odum,
and A. Plesch (2002), Shallow folding imaged above the Puente Hills blind-thrust
fault, Los Angeles, California, Geophys. Res. Lett., 29, 18-1 - 18-4.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Scientists of the U. S. Geological Survey and the Southern California Earthquake Center
(1994), The magnitude 6.7 Northridge, California, earthquake of 17 January 1994,
Science, 266, 389-397.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
74
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Stein, R. S. and Ekström, G. (1992), Seismicity and geometry of a 110-km-long blind
thrust fault. 2. Synthesis of the 1982 – 1985 California earthquake sequence: J.
Geophys. Res., 97(B4), 4865 - 4883.
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
Stephenson, W. J., Odum, J. Shedlock, K. M., Pratt, T. L., and Williams, R. A. (1992),
Mini-sosie high-resolution seismic method aids hazards studies, Eos Trans AGU,
73, 473-476.
Suppe, J. (1983), Geometry and kinematics of fault-bend folding: Am. J. Sci., 283, p.
684-721.
Suppe, J., and Medwedeff, D. A. (1990), Geometry and kinematics of fault-propagation
folding: Ecologae Geol. Helv., 83, no. 3, p. 409-454.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J. and Medwedeff, D. A. (1990), Geometry and kinematics of fault-propagation
folding: Ecologae geol. Helv., v. 83/3, 409-454.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
75
Walls, C., Rockwell, T., Mueller, K., Bock, Y., Williams, S., Pfanner, J., Dolan, J., and
Peng, F. (1998), Escape tectonics in the Los Angeles metropolitan region and
implications for seismic risk: Nature, 394, 356-360.
Wells, D. L., and K. J. Coppersmith (1994), New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface displacement,
Bull. Seismol. Soc. Am., 84, 974-1002.
76
CHAPTER 3:
Blind Thrust Paleoseismology of the Los Angeles Segment of the Puente Hills Blind-
Thrust Fault, Los Angeles, California: Implications for System-Wide
Earthquakes to Occur in a Segmented Blind-Thrust System
3.1 Introduction
The recent occurrence of several destructive thrust earthquakes (e.g., 1980 M
s
7.3
El Asnam, Algeria, 1987 M
w
6.0 Whittier-Narrows, California, 1994 M
w
6.7 Northridge,
California, 1999 M
w
7.6 Chi-Chi, Taiwan, 2005 M
w
7.5 Kashmir, Pakistan, and 2008 M
w
7.9 Wenchuan, China) demonstrate the substantial seismic risk these structures pose to an
increasingly urbanized population. The acquisition of detailed paleoseismological data on
the ages and displacements of ancient earthquakes that they have generated, are the keys
to understanding the rate at which these faults store and release seismic energy.
Although such data are routinely documented for many faults around the world, the
absence of paleoseismologic data from blind-thrust faults creates a major gap in our
ability to include this class of fault into probabilistic seismic hazard models. Previous
paleoseismological studies have documented Holocene slip rates and paleo-magnitudes
for uplift events involving the central Santa Fe Springs segment of the Puente Hills thrust
(PHT) [Dolan et al., 2003; Leon et al., 2007 (Chapter 2)]. The absence of paleo-
earthquake age and displacements from the western and eastern segments of the Puente
Hills thrust (PHT), however, creates a major gap in our understanding of not only the
seismic hazards related to slip on the individual thrust ramps but also the collective
behavior of this segmented blind-thrust system.
In an attempt to bridge the data gap and provide paleoseismological data for
seismic hazard analysis, I must study the near-surface folded strata overlying blind-thrust
77
ramps at seismogenic depths. Moreover, by acquiring paleoseimologic data from multiple
study sites above individual segments of blind-thrust systems, I can determine whether
these segmented blind-thrusts are capable of rupturing in system-wide, large-magnitude
(>M
w
7.0) earthquakes.
In this chapter, I use a combination of petroleum-industry seismic reflection
profiles [Shaw et al., 2002], high-resolution seismic reflection data, continuously cored
boreholes, and cone penetration tests to document the geometry and structural evolution
of shallow fault-related folds that have developed above the Los Angeles segment of the
PHT. I document a series of buried fold scarps that I tentatively interpret as uplift events
formed by slip on the underlying thrust ramp. I use these data to generate preliminary
paleo-earthquake magnitudes and slip rates for seismic hazard assessment.
3.1.1 Puente Hills Thrust Fault
The Puente Hills thrust (PHT) is a major blind-thrust fault that extends for >40
km along strike in the northern Los Angeles basin from near Beverly Hills east to Brea in
northern Orange County (Figure 3.1). The PHT was originally identified by Shaw and
Shearer [1999], who defined the central Santa Fe Springs segment with seismic reflection
profiles and well data. Using precisely relocated seismicity the authors proposed that
rupture of a small part of the central segment of the PHT generated the 1987 M
w
6.0
Whittier Narrows earthquake. Hypocentral relocations for the 1987 event and its
aftershocks reveal that the mainshock fault plane was coincident with the down-dip
projection of the Santa Fe Springs segment of the Puente Hills thrust at ~13 km depth
[Shaw and Shearer, 1999]. This moderate-magnitude earthquake provided direct
78
Figure 3.1 Structure Contour Map of the Puente Hills Blind-Thrust Fault
Structure contour map of the three en echelon segments of the Puente Hills blind-
thrust fault (PHT) showing the location of the 1987 Whittier Narrows (M
w
6.0)
earthquake sequence [Hauksson and Jones, 1989] as relocated by Shaw and Shearer,
[1999]. A0 - A2 marks the trace of the industry seismic-reflection profile shown in
Figure 3.2A. B0 - B1 marks the trace of both the weight-drop source profile (Figure
3.2B) and the the mini-vibe source profile (Figure 3.2C). The inset shows the
location of the PHT and 1994 Northridge (M
w
6.7) earthquake. Major state and
interstate highways are show for reference. Map coordinates are UTM Zone 11,
NAD27 datum (modified after Shaw et al., [2002]).
79
evidence that the PHT is active and capable of producing small, ruptures confined to
individual segments.
Shaw et al. [2002] documented the size and three-dimensional subsurface
geometry of the PHT using petroleum-industry seismic reflection profiles and well data.
They identified three right- stepping, en-echelon blind-thrust ramps that dip to the north ~
25
o
-30
o
and terminate upwards at ~ 3 km depth. These fault segments are, from west to
east, the Los Angeles segment, the Santa Fe Springs segment, and the Coyote Hills
segment. Through analysis of the mapped fault geometry and slip profiles, the authors
also described the systematic behavior of the segments. The en echelon fault ramps are
characterized by displacement profiles in which slip is gradually transferred from one
segment to another across segment boundaries with overlapping ramps. This suggests
that each fault ramp of the PHT could either rupture independently or in larger, multi-
segment (M
w
7.1) earthquakes.
The western, Los Angeles segment of the PHT extends from the western margin
of the Montebello Hills, beneath downtown Los Angeles, to near Beverly Hills (Figure
3.1). The industry seismic reflection data reveal narrow, well-defined, forelimb kink
bands with south-dipping strata consistent with classic fault-bend fold geometry (Figure
3.2). The kink-band is superimposed on a gently south-dipping monocline that forms the
northern boundary of the central Los Angeles Basin. The upper ~300 meters of the
petroleum-industry seismic reflection data (Figure 3.2A), however, is poorly imaged and
therefore precise definition of the age of the kink band above the Quaternary sequence
boundary is difficult to establish. The Quaternary strata do thin across the structure,
80
suggesting that the kink-band of the fault-related fold that formed above the Los Angeles
segment of the PHT developed at the same time as the Santa Fe Springs structure. To
bridge the data gap between the shallowest reflectors imaged in the petroleum industry
seismic reflection profile and the surface, high-resolution seismic reflection data at two
different scales were acquired (Figures 3.2B and 3.2C) to provide a continuous image of
folding from the tip of the thrust ramp at ~3 km depth to ~ 60 m depth. I also excavated
a transect of continuously cored boreholes and cone penetration tests (CPT’s) to
document a preliminary Holocene-latest Pleistocene slip rate and paleo-earthquake
history of the Los Angeles segment of the PHT. In this chapter I discuss the potential for
the occurrence of multi-segment ruptures and the implications of these results for seismic
hazard assessment.
3.2 Results
The study site is located along Budlong Avenue in the City of Los Angeles, ~ 6
km southwest of downtown Los Angeles, along the distal, aggradational, low-gradient
floodplain of the Los Angeles River (Figure 3.1). The river currently flows almost due
south at this location, approximately perpendicular to the surface projection of the active,
anticlinal axial surface of the fault-related fold observed in the deep-penetration
petroleum industry seismic-reflection data (Figure 3.2A). In the past, however, the Los
Angeles River periodically drained to the west, flowing westward just north of the study
site.
81
3.2.1 High-Resolution Seismic Reflection Data
I collected two north-south high-resolution seismic reflection profiles (Figures
3.2B and 3.2C) along Budlong Avenue ~2.5 km west of, and parallel to, the petroleum
industry profile acquired along Avalon Avenue shown in Figure 3.2A. The study site is
near the western edge of the Los Angeles segment, which strikes roughly N60
o
W at this
location. I acquired a 1640-m-long high-resolution seismic reflection profile utilizing a
weight-drop source to image the ~100- to 400-m depth range, thus overlapping with the
petroleum-industry data shown in Figure 3.2A (Figure 3.2B). Specifically, I used a
geometrics seismograph, available from IRIS, with a 60-channel system. Geophones
were placed along the side of the street at 5-m intervals. Four impacts from a truck-
mounted Geometrics PWD weight-drop source (82 kg) using elastomer technology were
recorded at each source point (5 m apart). An additional 670 m profile was acquired with
a mini-vibe source to image the 60- to 200-m-depth range to overlap with the weight-
drop source profile shown in Figure 3.2B. For this higher-resolution seismic reflection
profile I used a geometrics seismograph with a 144-channel system and mini-vibe T7000
source provided by the University of Nevada, Las Vegas. Data processing was routine
for seismic reflection profiles and included velocity analysis, residual statics corrections,
pre- and post-stack bandpass filter and deconvolution, Stolt time migration, and time-to-
depth conversion.
The industry and high-resolution seismic reflection data image south-dipping
strata within a forelimb kink band bounded by an active, anticlinal and inactive, synclinal
axial surface. A number of dipping reflectors between 100- to 400-m-depths are clearly
82
Figure 3.2 Multi-scale Seismic Reflection Profiles of Folding above the Los Angeles
Segment of the Puente Hills Blind-Thrust Fault.
(A) Petroleum-industry seismic reflection image of the forelimb fold structure
showing an upward-narrowing zone of folding (growth triangle) delimited by
sharply defined axial surfaces [Shaw et al., 2002]. Dashed black lines represent
active, antinclinal and inactive, synclinal axial surfaces (A0 -A2 Figure 3.1).
(B) Weight-drop source seismic reflection profile (migrated; B0 -B1 Figure 3.1).
Vertical yellow lines represent borehole locations and numbers in red represent
specific boreholes. (C) Mini-vibe source seismic reflection profile (migrated; 8x
vertical exaggeration). Vertical yellow lines indicate borehole locations and
numbers in red represent specific boreholes.
83
Figure 3.2 Continued
84
discernible on the weight-drop source seismic reflection profile. These reflectors are
folded within an upward-narrowing zone of south-dipping reflectors ~350 m wide at 400
m depth decreasing in width to < 150 m wide at ~100 m depth. These reflectors are
consistent with the upward continuation of the kink band identified in the deeper-
penetration industry profile (Figure 3.2A). The mini-vibe-source seismic reflection
profile (Figure 3.2C) also images several prominent, south-dipping reflectors between
~60- to 100-m depths located at the upward projection of the growth triangle observed in
the weight-drop profile (Figure 3.2B). The combined industry and high-resolution
seismic reflection data provide an uninterrupted image of the full depth extent of the fold
above the Los Angeles segment of the PHT, from ~3 km depth to within 60 m of the
surface, allowing us to accurately site boreholes and cone penetration tests across the
locus of most recently active folding (Figures 3.3 and 3.4).
3.2.2 Borehole Data
I excavated a 730-m-long, north-south transect of eleven, 17- to 34-m-deep,
continuously cored boreholes along the east curb of Budlong Avenue. The borehole
transect is located across the updip projection of the active, anticlinal axial surface
identified in both high-resolution seismic reflection profiles (Figures 3.2B and 3.2C).
The stratigraphy consists of a laterally extensive sequence of fine- to coarse-grained
sand- and gravel-filled channels interbedded with cohesive, fine- grained silt and clay
overbank units deposited during Holocene-latest Pleistocene time (Figures 3.3 and 3.4).
The sands represent the highest-energy deposits in the Holocene part of the section, and
in almost all cases (except the shallowest part of the stratigraphic section), these sands are
85
overlain by very fine-grained (clay/ silt) deposits that record low-energy, slack-water
deposition. It is therefore unlikely that the tops of the sands have been modified by
erosion. The uppermost 10- to 16 m of the section consists of fine- to medium-grained
sands separated by cohesive silt and clay units. There is a marked change downward to a
much coarser-grained, gravel-dominated section at 12- to 17- m depth (Unit 60). Unit 60
is a distinctive, laterally extensive coarse-grained sand and gravel unit that is continuous
across the entire transect. The units within the upper 16 m, however, exhibit lateral facies
changes, possibly due to different sediment sources or source directions. In general,
these units exhibit stratigraphic sequences defined by grain size, color, texture, and
magnetic susceptibility, allowing preliminary correlations between units across the
borehole transect. I provide two alternative preliminary cross-sections (Figures 3.3 and
3.4) to show two potential sets of stratigraphic reconstructions.
3.2.3 Cone Penetration Testing Data
In addition to the 11 borehole excavations, I acquired six cone penetration tests
(CPTs) to provide details of the stratigraphy within the zone of deformation encountered
at our study site (Figures 3.3 and 3.4). Although cores are not collected directly, CPT
analysis measures and combines three main parameters for soil behavior that are
generally consistent with the grain size of stratigraphic units logged from nearby borehole
excavations. The interpreted soil behavior (SBT) is calculated after Robertson et al.
[1986] and reported for depths of up to 50 feet. For depths greater than 50 feet SBTn
(modified SBT) is calculated after Robertson [1990]. These geotechnical parameters are
86
Figure 3.3 Borehole Results from the Budlong Avenue Transect
Cross-section of major stratigraphic units (8x vertical exaggeration). Colors denote different
sedimentary units. Vertical black lines indicate the borehole locations. Vertical blue lines indicate
CPT locations. Double-headed red vertical arrows along the left side of the figure show the
stratigraphic ranges of intervals of sedimentary thickening across the transect, with the uplift in
each event shown in red to the left of each arrow. Double-headed green vertical arrows show
intervals of no sedimentary growth. Green horizontal line represents the regional gradient of the
study site.
87
Figure 3.4 Alternative Borehole Results from the Budlong Avenue Transect
Alternative cross-section of major stratigraphic units (8x vertical exaggeration). Colors denote
different sedimentary units. Vertical black lines indicate the borehole locations. Vertical blue lines
indicate CPT locations. Green horizontal line represents the regional gradient of the study site.
88
interpreted by the software as 12 “soil behavior zones” that range from sensitive, fine-
grained material to sand-gravelly sand. These sedimentary characteristics are reported as
vertical logs for each CPT hole. The CPT logs provide additional confidence in
correlating stratigraphic units across the zone of deformation located by high-resolution
seismic reflection and continuously cored borehole data.
3.2.4 Stratigraphic Evidence for Paleo-Folding Events
In the Budlong Avenue transect, I found evidence for two main growth intervals
separated by a unit that does not change thickness across the transect. My preliminary
interpretation (Figure 3.3) suggests the occurrence of at least four uplift events identified
by the southward stratigraphic thickening (or “growth”) that occurs across the fold limb.
I present an alternative interpretation in Figure 3.4 because there is potential ambiguity
in my preliminary stratigraphic horizons in the upper 13 m because of lateral facies
changes within these units.
I tentatively interpret the upper and lower growth intervals in Figure 3.3 as now-
buried, south-facing fold scarps that were subsequently onlapped by the next-youngest
stratigraphic unit. This reflects the relative uplift of the hanging wall during folding
and the subsequent, post-uplift deposition that restores the original near-horizontal stream
gradient (0.50
o
S at the study site). Sedimentary growth is localized in an upward-
narrowing zone located at the updip projection of the active, anticlinal axial surface
imaged on the petroleum industry, weight-drop, and mini-vibe seismic reflection profiles
(Figure 3.2). The uplift events, labeled Y, X, W, and V, from youngest to oldest
(following the terminology of Dolan et al., [2003] for the Santa Fe Springs segment of
89
Table 3.1 Calibrated, Calendric Dates and Radiocarbon Ages of Samples Collected
from the Budlong Avenue Transect
All analyses were performed at Lawrence Livermore National Laboratory in
Livermore, California. Samples were calibrated using Oxford Radiocarbon
Accelerator Unit (ORAU) calibration program OXCAL v4.11 [Bronk Ramsey, 1995;
2001; using atmospheric data from Reimer et al., 2004 (IntCal07); Bronk Ramsey
and van der Plicht, 2001]. For sample type; ccl=charcoal fragment; bulk=bulk soil.
Calendric age is reported as 2 sigma (95% confidence limit) age range. Bh indicates
the borehole that the sample was collected from. Calendric age is listed as calibrated
years before present (Cal BP).
90
Table 3.1, Continued
PDF
#
Sample # bh
Sample
type
True
depth
(m)
Projected
depth to
bh 8 (m)
14
C Age BP
13
C Calendric
age (Cal BP)
%
proba-
bility
1 PHTLA51 5 bulk 1.58 1.60 4840 + 40 -25 5653 – 5575
& 5546 –
5475
57.2
38.2
2 PHTLA45 4 bulk 1.60 1.60 6400 + 45 -25 7422 – 7262 95.4
3 PHTLA11 1 bulk 1.68 1.68 4980 + 35 -25 15880 – 5826
& 5721 –
5610
11.2
84.2
4 PHTLA21 2 ccl 5.08 4.78 2660 + 35 -25 2845 – 2742 95.4
5 PHTLA41 4 ccl 4.73 4.78 2745 + 35 -25 2925 – 2763 95.4
6 PHTLA61 6 bulk 4.98 4.78 2770 + 35 -25 2951 – 2784 95.4
7 PHTLA34 3 bulk 4.01 5.0 3360 + 35 -25 3690 – 3552
& 3539 –
3482
81.5
13.9
8 PHTLA16 1 bulk 2.19 5.04 3070 + 40 -25 3380 – 3206
& 3185 –
3165
92.9
2.5
9 PHTLA10
1
10 bulk 5.08 5.08 3115 + 35 -25 3440 – 3429
& 3405 –
3247
1.4
94.0
10 PHTLA81 8 bulk 5.18 5.18 2970 + 35 -25 3261 – 3004 95.4
11 PHTLA55 5 bulk 3.13 5.21 3610 + 40 -25 4080 – 4035
& 4000 –
3830
6.9
88.5
12 PHTLA31 3 bulk 4.32 5.08 4410 + 45 -25 5277 – 5167
& 5216 –
5108 &
5071 – 4861
16.7
2.0
76.7
13 PHTLA72 7 bulk 4.88 6.10 4780 + 35 -25 5595 – 5465
& 5359 –
5354 & 5348
– 5333
92.6
0.5
2.4
14 PHTLA71 7 bulk 4.88 6.15 5085 + 35 -25 5913 – 5746 95.4
15 PHTLA32 3 bulk 5.06 6.20 7230 + 40 -25 8161 – 7970 95.4
16 PHTLA10
2
10 bulk 6.27 6.27 4905 + 50 -25 5744 – 5582
& 5502 –
5490
94.2
1.2
17 PHTLA91 9 bulk 5.33 7.93 4685 + 40 -25 5579 – 5532
& 5481 –
5316
12.9
82.5
18 PHTLA12 1 bulk 5.69 7.93 4780 + 35 -25 5595 – 5465
& 5359 –
5354 & 5348
– 5333
92.6
0.5
2.4
19 PHTLA42 4 ccl 8.13 7.93 11260 + 180 -25 13462 –
12861
95.4
20 PHTLA33 3 ccl 7.93 7.93 10240 + 160 -25 12622 –
12445 &
12413 –
11340
7.0
88.4
21 PHTLA53 5 ccl 8.54 11.00 18850 + 90 -25 22554 –
22210
95.4
91
Table 3.1, Continued
PDF
#
Sample # bh
Sample
type
True
depth
(m)
Projected
depth to
bh 8 (m)
14
C Age BP
13
C Calendric
age (Cal BP)
%
proba-
bility
22 PHTLA13 1 ccl 8.05 12.96 23980 + 570 -25 25215 –
22910
95.4
23 PHTLA22 2 bulk 15.52 15.85 9980 + 40 -25 11679 –
11319 &
11619 -
11625
95.4
24 PHTLA10
3
10 bulk 15.93 16.03 9465 + 35 -25 11060 –
11035
95.4
25 PHTLA82 10 bulk 16.16 16.16 10265 + 40 -25 12227 –
12216 &
12165 –
11822
0.5
94.9
26 PHTLA54 8 bulk 11.28 17.53 13780 + 210 -25 17065 –
15765
95.4
27 PHTLA92 1 bulk 23.93 ~54.00 49400 +
3900
-25 69826 –
44154
95.4
92
the PHT), occurred between 0 to 3.3 ka, 2.8 to 5.6 ka, 5.6 to 10.8 ka, and 5.6 to 17.2 ka,
respectively (Table 3.2).
As defined by the minimum paleo-fold-scarp heights, uplift in Event Y was ≥ 1.1
+ 0.1/-0.2 m. For Event X, minimum uplift is ≥ 1.3 + 0.2/- 0.1 m. For Event W,
minimum uplift measured is ≥ 1.7 m + 0.1, and for Event V, minimum uplift measured is
> 2.3 + 0.1 m. The upper and lower intervals of sedimentary growth are separated from
one another by stratigraphic intervals that do not exhibit changes in thickness across the
transect. Specifically, Units 20, 35, and 45 do not exhibit any systematic thickness
changes along the length of the transect (Figure 3.3). I interpret these constant-thickness
units as recording periods of structural quiescence characterized by deposition at the
regional near-horizontal stream gradient.
3.2.5 Radiocarbon Dating
Accelerator Mass Spectrometer (AMS) radiocarbon (
14
C) analyses of 27 samples
(20 bulk soil and 7 charcoal; Table 3.1) indicate that sediment accumulation has been
relatively continuous at ~1.4 mm/yr over the past ~13 ka (Figure 3.5).
Radiocarbon dates were calibrated using Oxcal v.4.11 [Bronk Ramsey, 1995;
Bronk Ramsey, 2001 (using atmospheric data from Reimer et al., [2004] (IntCal04);
Bronk Ramsey et al., 2001), which uses a Bayesian statistical analysis and stratigraphic
ordering to calculate calendric dates [Bronk Ramsey, 2001]. All calendric dates are
reported as two-sigma (95% confidence limit) age ranges. Table 3.1 shows the depth at
which each sample was collected, as well as the depth of the sample projected along
bedding to its correlative stratigraphic level in borehole PHTLA8. A scarcity of charcoal
93
Figure 3.5 Sediment Accumulation-Rate Curve for Budlong Avenue Transect
Sediment accumulation-rate curve for Budlong Avenue transect, with calibrated,
calendric radiocarbon dates sorted by depth. Radiocarbon dates were calibrated
using the Oxford Radiocarbon Accelerator Unit (ORAU) calibration program
OXCAL v4.1.1. (Bronk Ramsey, 1995; Bronk Ramsey, 2001[Using atmospheric
data from Reimer et al., 2004 (IntCal 04)]; Bronk Ramsey et al., 2001). True depth
was corrected for folding by projecting all samples to a common reference point
(PHTLA8), by moving each sample along the nearest stratigraphic horizons to its
projected depth in borehole PHTLA8. Numbers next to the probability functions
represent sample numbers in Table 3.1. Sediment accumulation-rate curve is gray.
Anomalously old samples are circled. Uncertainties in ages of uplift events (Y
through V) are marked by horizontal black lines, whereas depth uncertainties are
marked by vertical black lines. BH# indicates borehole location for each sample.
94
Figure 3.5, Continued
95
Figure 3.5, Continued
PDF Sample #
14
C Date BH# PDF Sample #
14
C Date BH#
1 PHTLA51-BS 4840+40BP PHTLA5 15 PHTLA32-BS 7230+40BS PHTLA3
2 PHTLA45-BS 6400+45BP PHTLA4 16 PHTLA102-BS 4905+50BP PHTLA10
3 PHTLA11-BS 4980+35BP PHTLA1 17 PHTLA91-BS 4685+40BP PHTLA9
4 PHTLA21 2660+35BP PHTLA2 18 PHTLA12-BS 4780+35BP PHTLA1
5 PHTLA41 2745+35BP PHTLA4 19 PHTLA42 11260+180BP PHTLA4
6 PHTLA61-BS 2770+35BP PHTLA6 20 PHTLA33 10240+160BP PHTLA3
7 PHTLA34-BS 3360+35BP PHTLA3 21 PHTLA53 18850+90BP PHTLA5
8 PHTLA16-BS 3070+40BP PHTLA1 22 PHTLA13 23980+570BP PHTLA1
9 PHTLA101-BS 3115+35BP PHTLA10 23 PHTLA22-BS 9980+40BP PHTLA2
10 PHTLA81-BS 2970+35BP PHTLA8 24 PHTLA103-BS 9465+35BP PHTLA10
11 PHTLA55-BS 3610+40BP PHTLA5 25 PHTLA82-BS 10265+40BP PHTLA8
12 PHTLA31-BS 4410+45BP PHTLA3 26 PHTLA54-BS 13780+210BP PHTLA5
13 PHTLA72-BS 4780+35BP PHTLA7 27 PHTLA92-BS 40580+420BP PHTLA8
14 PHTLA71-BS 5085+35BP PHTLA7
96
and limited organic-rich layers restricted our ability to date the entire section. Organic
rich Units 25 and 55 therefore yielded the majority of calibrated ages. In the case where
older samples overlie younger samples, I assume that samples include reworked material
that was deposited during deposition of the soil (e.g., samples PHTLA51, PHTLA45,
PHTLA11, and PHTLA16).
3.3 Discussion
In this chapter, the combined seismic reflection profiles, and continuously cored
borehole and CPT data document the structural evolution and geometry of the forelimb
growth triangle developed above the Los Angeles segment of the Puente Hills blind-
thrust fault. These data reveal that folding is continuous throughout the entire depth
range, from the tip of the thrust ramp at ~3 km to the surface, localized within a discrete
kink band between an active, anticlinal and inactive, synclinal axial surface. Moreover,
the localization of deformation along the fold’s anticlinal axial surface and the upward
narrowing of the fold limb within the syntectonic growth section suggests that the fold is
growing, at least in part, by kink-band migration in a manner broadly consistent with
growth fault-bend fold theory [Suppe et al., 1992; Shaw and Suppe, 1994; Shaw et al.,
2002].
3.3.1 Fault Slip Rate Estimates
I calculate fault slip estimates for the Los Angeles segment of the PHT and uplift
in individual events by dividing the structural relief above specific stratigraphic horizons
across the fold by the sine of the 27°±2° N of the thrust ramp documented by Shaw et al.
[2002]. Recent discrete-element models demonstrate that uplift measured at my study site
97
Table 3.2 Uplift Amounts, Age Limits, and Estimated Paleo-magnitudes for the Los Angeles Segment of the
Puente Hills Blind-Thrust Fault.
Event Age (ka) Growth section Uplift
(m)
Slip (m) M
w
(All-
slip-type
avg.
displace-
ment)
M
w
(All-
slip-type
max.
displace-
ment)
M
w
(Thrust-
fault-only
avg.
displace-
ment)
M
w
(Thrust-
fault-only
max.
displace-
ment)
Y 0 – 3.3 ka
Top of Unit 10
to top of Unit
20
1.1 +
0.1/-0.2
2.4 + 0.9 7.25 + 0.2
7.0 + 0.3
6.7 -0.2/+0.5 6.7 -0.2/+0.4
X 2.8 – 3.6
ka
Top of Unit 25
to top of Unit
35
1.3 +
0.2/-0.1
3.7 +1.1/-
0.7
7.3 + 0.2
7.0 + 0.3
6.7 -0.2/+0.5 6.7 -0.2/+0.4
W 5.6 – 10.8
ka
Top of Unit 40
to top of Unit
45
1.7 +
0.1
2.5 +1.3/-
1.0
7.4 + 0.3
7.1 + 0.3
6.7 -0.2/+0.5 6.8 -0.2/+0.4
V 5.6 – 17.2
ka
Top of Unit 50
to top of Unit
60
2.3 +
0.1
5.1 +0.2 7.5 + 0.3
7.2 + 0.3
6.7 -0.2/+0.5 6.8 -0.2/+0.4
Uplift amounts, age limits, and estimated moment magnitudes (M
w
) for paleoearthquakes V through Y, as
determined from the Budlong Avenue borehole results. See text for discussion. Preferred M
w
assumes that the
measured slip was the average or maximum slip during paleoearthquakes, using regressions of Wells and
Coppersmith (1994).
98
provides a robust proxy for fault slip at depth [Benesh et al., 2007]. This methodology
assumes that no folding occurs outside of the narrow kink band documented in the
borehole results, which in turn renders all of the estimates of uplift and fault displacement
as minima. Dividing the structural relief at the top of Unit 60 (the most distinctive and
robust stratigraphic horizon at the study site) by the sine of 27°±2° indicates cumulative
thrust displacement at this horizon of 14.3 + 1.1 m. Dividing the cumulative thrust
displacement by a calibrated radiocarbon sample (17,150-15,750 Cal YBP) collected
from the top of Unit 60 (PHTLA54) yields a minimum Holocene-latest Pleistocene slip
rate of 0.90 + 0.10 mm/yr, similar to the long-term slip rate of 0.8 to 0.9 mm/yr
calculated by Shaw et al., [2002]. My rate is a minimum because: (1) the uplift
measurements fail to account for any potential erosion of the hanging wall; (2) it assumes
that all slip on the deep PHT Los Angeles ramp is manifested as folding within the kink
band. As noted above, however, I think that erosion of the hanging wall uplift in each of
the folding events is likely to be minimal because of the low energy deposition recorded
by the fine-grained deposits overlying most of the event horizons.
Complicating this result is the fact that the Budlong site lies near the western end
of the Los Angeles segment of the PHT. Shaw et al. [2002] used calculations of total
cumulative displacement on the Los Angeles segment based on kink band width and
structural relief of the base Quaternary reflector observed in industry seismic profiles to
show that cumulative slip at the Budlong site is only ~50% of the maximum cumulative
slip near the center of the Los Angeles segment. Thus, in a long-term sense, my slip
estimates from the borehole data may record ≤50% of total long-term slip on the deep
99
part of the Los Angeles segment thrust ramp. Alternatively, due to the sub-parallel trend
of the PHT and Compton fault at the study site and the close proximity of their
overlying folds (they are separated by < 2.5 km in the shallow sub-surface at this
location), strain may be partitioned between the two thrust ramps, thereby reducing the
amount of slip generated by both the Los Angeles segment of the PHT and northern
segment of the Compton fault. This may also explain why the slip rates calculated for the
southern segment of the Compton fault in Lakewood, California (Chapter 7) appear to be
faster than our northern segment slip rate (Chapter 6) by as much as 1 mm/yr.
3.3.2 Earthquake-by-Earthquake Fold Growth
I infer that the four uplift events represent deformation during single, large-
magnitude (>M
w
7.0) earthquakes on the PHT. It is also possible however, that each of
these uplift events comprises more than one earthquake. I favor the single-event
interpretation for uplift events Y, X, and W, however, particularly in light of the
observation that discrete, co-seismic folding is documented in very large earthquakes that
rupture to fault tips and across fault bends [e.g., Chen et al., 2007], but generally does not
occur in moderate-magnitude blind-thrust earthquakes [e.g., Lin and Stein, 1989].
Moreover, the temporal discreteness of these uplift events obviates the possibility that the
folds grow in response to quasi-continuous fault creep on the PHT ramp. The possibility
that some component of folding occurs during punctuated periods of aseismic fault slip,
either as afterslip following large earthquakes or during discrete periods of interseismic
fold growth, however, cannot be ruled out. Finally, due to the large amount of uplift and
lack of readily identifiable, stratigraphically discrete intervals without growth, I cannot
100
rule out the possibility that that the uplift measured for Event V (2.3 + 0.1 m) occurred
during multiple events that were clustered in time.
3.3.3 Paleo-magnitude Estimates and Implications for Fault Behavior
I estimate a conservative range of magnitudes for these paleo-earthquakes by
comparing my measured minimum displacements to global regressions of slip vs.
magnitude [Wells and Coppersmith, 1994]. If I make the simplifying assumption that my
measured displacements represent average displacement in each paleo-earthquake, the
magnitude estimates for PHT thrust earthquakes range from M
w
7.25 + 0.2 to M
w
7.5 +
0.4 (Table 3.2). Alternatively, if I assume the extreme limiting case that all of my
measured displacements record the maximum displacements in each earthquake, the
minimum magnitude estimates range from M
w
7.0 + 0.3 to M
w
7.2 + 0.3. If I use the
thrust-fault-only regressions of Wells and Coppersmith [1994], rather than their all-slip-
type regressions, and each of our displacement measurements represents an average
displacement, the magnitude estimates are M
w
6.7 -0.2/+0.5. If my measurements record
maximum displacement, then the thrust-fault-only magnitude estimates range from M
w
6.7 -0.2/+0.4 to M
w
6.8 -0.2/+0.4. The thrust regressions, however, have many fewer
data points than the regressions based on all-slip-type data, and the uncertainty in the
mean is therefore larger for thrust-slip-only regression [Wells and Coppersmith, 1994].
In summary, the uncertainties associated with Wells and Coppersmith [1994] regressions
convert into uncertainties on the order of 0.1 to 0.3 in magnitude estimates for all-slip-
type regressions, with larger uncertainties of 0.1 to 0.5 in magnitude estimates for thrust-
only regressions due to the fewer number of available data for thrust-fault events. Based
101
on regressions of rupture area to magnitude (rather than slip), rupture of all three
segments of the PHT could generate an earthquake of ~M
w
7.1 [Shaw et al., 2002], in line
with the smaller end of the magnitude estimates based on my measured displacements.
3.3.4 Incremental Structural Evolution of Fault-related Folds
The borehole data and high-resolution seismic reflection profiles reveal that
development of folding within the forelimb growth triangle occurs incrementally. The
folded strata within the forelimb growth triangle acquire their dip progressively through
time, such that deeper strata, which have experienced more PHT earthquakes, dip more
steeply than younger, less-deformed strata near the top of the growth triangle. This
progressive downward increase in bed dip is observed at all my study sites and reflects an
aspect of fold kinematics that is distinct from classic growth fault-bend fold theories.
This change in limb dips may reflect some component of fold growth by limb rotation in
addition to kink-band migration. These hybrid kinematics are manifest in several types
of fault-related folds, including trishear fault-propagation folding [Erslev, 1991;
Allmendinger, 1998] and shear fault-bend folding [Suppe and Connors, 2004; Shaw et al.,
2005]. Alternatively, the component of limb rotation may reflect curvature of the fold
hinge, as is described by curved-hinge fault-bend folding theories [Suppe et al., 1997;
Novoa et al., 2000]. Finally, the progressive change in bed dips may reflect the
mechanical response of loosely consolidated, granular sediments in the shallow
subsurface to folding at depth. Recent work by Benesh et al., [2007] demonstrates that,
even in cases of folding of pre-tectonic layers exclusively by kink-band migration,
102
shallow growth sediments acquire their dips progressively. This is due to the finite width
of the axial surface zone and its rotation and migration with increasing deformation.
3.3.5 Implications for Seismic Hazard in Southern California
The occurrence of large-magnitude earthquakes on the PHT has obvious and
significant implications for seismic hazard assessment, given the location of this fault
directly beneath downtown Los Angeles. Indeed, recent loss estimates for large PHT
earthquakes based on the findings of Dolan et al., [2003] suggest that the occurrence of
such an event would represent one of the worst seismic disasters in the United States
[Field et al., 2005]. Of particular concern is the possibility that any up-dip source
directivity would funnel energy directly into the 10-km-deep Los Angeles basin.
Moreover, such large-magnitude earthquakes on the PHT would involve permanent
displacements of several meters directly beneath the downtown Los Angeles high-rise
district. Moreover, providing paleoseismologic data for the Los Angeles segment of the
PHT provides the ability to assess the potential for multi-segment ruptures of this
segmented blind-thrust system.
The results of this study combined with the previously published paleoseimologic
data for the Santa Fe Springs segment of the PHT [Dolan et al., 2003; Chapter 2] provide
slip rates and paleoseismologic data from two major segments of the PHT, allowing me
to assess the degree to which these segments fail together, or separately. Although these
combined studies cannot uniquely confirm that a co-seismic folding event observed at the
study site above the Santa Fe Springs segment is the same co-seismic folding event
observed at the Budlong Avenue site above the Los Angeles segment, our preliminary
103
findings indicate that both the timing constraints and amounts of displacement that
occurred in each event at each site suggests the data are permissive of large-magnitude,
multi-segment ruptures. For example, the displacement calculated for Event Y at the
study site above the Santa Fe Springs segment of the PHT is 2.4 + 0.4 m for the Carfax
transect and 3.5 + 0.9 m for the Gardenland transect [Leon et al., 2007 (Chapter 2)]. The
displacement calculated for Event Y at the Budlong study site above the Los Angeles
segment is 2.4 + 0.9 m. Moreover, the timing of Events at both study sites are permissive
of multi-segment ruptures. Event Y occurred between 0.2 and 3.0 ka at the study site
above the Santa Fe Springs segment [Leon et al., 2007 (Chapter 2)], and at the Los
Angeles study site Event Y occurred between 0 and 3.3 ka. Additional radiocarbon
analyses from Lawrence Livermore National Laboratories should significantly reduce the
uncertainties in the timing of events for the Budlong Avenue study site. These results,
however, particularly evidence for large displacements in the paleo-earthquakes that
generated the fold scarps, are most consistent with the idea of large-magnitude, multi-
segment ruptures.
104
Chapter 3 References
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Allmendinger, R. W., and Shaw, J. H. (2000), Estimation of fault propagation distance
from fold shape: Implications for earthquake hazard assessment: Geology, v. 28,
n. 12, 1099-1102.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Field, E. H., H. A. Seligson, N. Gupta, V. Gupta, T. H. Jordan, and K. W. Campbell
(2005), Loss estimates for a Puente Hills blind-thrust earthquake in Los Angeles,
California, Earthquake Spectra, 21, 329-338.
Field., E. H. (2005), Collaborative SCEC/USGS efforts to improve seismic-hazard analysis;
RELM and OpenSHA: Proceedings of the 5th U.S.-Japan natural resources meeting and
Parkfield, California fieldtrip, Open-File Report - U. S. Geological Survey, Report: OF
2005-1131, pp. 18. Resource Location: http://pubs.usgs.gov/of/2005/1131/
105
Hauksson, E., and Jones, L. (1989), The 1987 Whittier Narrows earthquake sequence in Los
Angeles, southern California: Seismological and tectonics analysis: J. Geophys. Res.,
94, 9569.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Roberston, P.K. (1990), Soil Classification using the Cone Penetration Test, Canadian
Geotech. Journ., v. 27, p 151-158.
Robertson, P.K., Campanella, R. G., Gillespie, D., and Rice, A. (1986), Seismic CPT to
Measure In-Situ Shear Wave Velocity, J. Geotech. Eng. ASCE, v. 112, No. 8, p
791-803.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
106
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
Wells, D. L., and K. J. Coppersmith (1994), New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface displacement,
Bull. Seismol. Soc. Am., 84, 974-1002.
107
CHAPTER 4:
Holocene and Late Pleistocene Slip Rates for the Santa Fe Springs Segment of the
Puente Hills Blind-Thrust Fault
4.1 Introduction
Recent moderate-magnitude (M
w
<7.0) blind-thrust earthquakes (e.g., 1987 M
w
6.0
Whittier-Narrows and 1994 M
w
6.7 Northridge) demonstrate the substantial seismic risk
posed by blind-thrust faults to metropolitan Los Angeles. Unlike most faults, which
rupture to the surface in large earthquakes, deformation associated with slip on blind
faults is manifested as near-surface folding, rather than faulting. In the case of very large
earthquakes on blind-thrust faults, discrete coseismic fold scarps have developed above
bends or tips of the propagating faults [e.g., King and Vita-Finzi, 1981; Stein and King,
1984; Stein and Yeats, 1989; Lin and Stein, 1989; Lai et al., 2006; Chen et al., 2007],
providing potential targets for studies of blind-thrust slip rates and paleo-earthquake
history.
4.1.1 Puente Hills Thrust Fault
The Puente Hills blind-thrust fault (PHT) extends for >40 km beneath the
northern edge of the Los Angeles Basin from near Beverly Hills, east-southeastward
beneath downtown Los Angeles to northern Orange County (Figure 4.1). The PHT was
originally identified by Shaw and Shearer [1999] using petroleum industry seismic
reflection profiles and earthquake relocations from the 1987 Whittier Narrows (M
w
6.0)
earthquake [Hauksson and Jones, 1989]. Shaw et al., [2002] documented the subsurface
geometry of PHT using industry seismic profiles and well data. They identified three
right-stepping, en echelon thrust ramps that dip to the north 25
o
-30
o
and terminate upward
108
at ~3 km. These fault segments are, from west to east, Los Angeles, Santa Fe Springs,
and Coyote Hills (Figure 4.1), each of which is capable of generating M
w
6.5 earthquakes
individually, or a M
w
7.0 earthquake if they rupture simultaneously [Shaw and Shearer,
1999]. Petroleum Industry seismic reflection profiles show that each fault segment is
overlain by fault-related folds with Quaternary sediments thinning across the crests
consistent with Quaternary fault motion. Deformation associated with slip on the
underlying thrust ramps is localized within discrete kink bands that progressively narrow
upwards into syntectonic (growth) strata.
To bridge this data gap, I acquired a high-resolution seismic reflection profile and
66-m-deep borehole to document the geometry and structural evolution of shallow-
subsurface folding above the backlimb of the central Santa Fe Springs segment of the
PHT. I use these data to calculate structural relief across the backlimb growth triangle
and generate preliminary intermediate-term (10
4
yrs) slip rates for the PHT.
4.2 Results
Over the past decade, the development of multi-disciplinary methodologies for
extracting paleoseismological data from blind-thrust faults has yielded slip rates and the
ability to document the size and timing of individual paleoearthquakes [e.g. , Dolan et al.,
2003; Sugiyama et al., 2003; Ishiyama et al., 2004; 2007; Leon et al., 2007 (Chapter 2)].
The Holocene-Latest Pleistocene slip rates provided by recent paleoseismologic studies,
combined with longer-term geologic rates constrained by the analysis of Pliocene to early
Pleistocene growth structure [Shaw et al., 2002] provide slip rates on short and long-time
scales. The slip rates for intermediate time scales, however, are still missing, and
109
Figure 4.1 Structure Contour Map of the Puente Hills Blind-Thrust Fault
Structure contour map of the three en echelon segments of the Puente Hills blind-
thrust fault (PHT) showing the location of the 1987 Whittier Narrows (Mw 6.0)
earthquake sequence [Hauksson and Jones, 1989] as relocated by Shaw and Shearer,
[1999], modified after Shaw et al., [2002]. A0 - A3 marks the trace of the seismic
reflection profile shown in Figure 4.2A. B0 - B1 marks the trace of the mini-vibe
source profile (Figure 4.2B). C marks the trace of the mini-sosie source profile
acquired along Carfax Avenue across the forelimb that overlies the Santa Fe
Springs segment of the PHT (Figure 4.2A). The inset shows the location of the PHT
and 1994 Northridge (M
w
6.7) earthquake. Major state and interstate highways are
shown for reference, Map coordinates are UTM Zone 11, NAD27 datum.
110
information on the consistency of slip rates over different time scales directly affects how
we assess regional earthquake hazards. To successfully analyze the activity of blind-
thrust faults requires a multidisciplinary methodology to relate near-surface deformation
to the causative fault at seismogenic depths. I use a combination of petroleum-industry
seismic reflection profile analysis and high-resolution seismic reflection data acquisition
with borehole excavations to observe the entire depth range of folding.
4.2.1 Seismic Reflection Data
The study site is located in the City of Pico Rivera, ~17 km southeast of downtown Los
Angeles along the Los Angeles County flood control area west of the San Gabriel River, a major
south-flowing drainage (Figure 4.1). I collected a high-resolution seismic reflection profile above
the backlimb of the central, Santa Fe Springs segment of the PHT (Figure 4.2B; B0 to B1 Figure
4.1). This location is ~ 8 km upstream for the forelimb study site in Bellflower, California [Pratt
et al., 2002; Shaw et al., 2002; Dolan et al., 2003; Leon et al., 2007 (Chapter 2)] and parallel to
the industry profile also collected along the San Gabriel River (Figure 4.2A). At this location, the
seismic reflection profiles are oriented nearly perpendicular to the active, synclinal axial surface
associated with the PHT backlimb growth triangle, which extends approximately east-west
beneath the study site [Shaw et al., 2002]. A 144-channel seismic system with mini-vibe source
was utilized to image reflectors between ~45 to 600 m. Geophones were placed at 5 m intervals
along the side of the flood control access road on which the source was used. I recorded 45
seconds of data in the 160-to-240 hertz frequency range.
The high-resolution seismic reflection data reveal a panel of north-dipping
reflectors located at the updip projection of the active, synclinal axial surface observed in
the industry profile [Shaw et al., 2002; Figure 4.2]. Deformation is localized within a
111
Figure 4.2 Multi-scale Seismic Reflection Images of the Backlimb Fold Structure of
the Santa Fe Springs Segment of the Puente Hills Blind-Thrust Fault
Multi-scale seismic reflection images of the backlimb fold structure showing an
upward-narrowing zone of active folding (growth triangle) delimited by sharply
defined axial surfaces. These overlapping profiles provide a complete image of
backlimb folding above the Santa Fe Springs segment of the Puente Hills thrust
(PHT) from the top of the thrust ramp at 3 km depth to the surface. (A) Petroleum-
industry seismic reflection profile [Shaw and Shearer, 1999; Pratt et al., 2002]; red
lines represent fault plane reflections, solid colored lines in (A) and (B) represent
reflectors and black dashed lines represent axial surfaces. (B) High-resolution
seismic reflection profile (migrated; no vertical exaggeration) acquired with a mini-
vibe source along the San Gabriel River across the backlimb fold image in part (A).
Reflectors imaged from ~600 m to ~45 m within the surface. Deformed strata
corresponding to reflections in the shallow subsurface define late Quaternary
activity on the PHT. Deformation is localized along the active, synclinal axial
surface indicated by dashed black line in parts (A) and (B). The upper yellow and
pink reflectors in part (A) are also shown on the high-resolution profile in part (B).
Vertical yellow line indicates the location of a 66-m-deep borehole excavated next to
the San Gabriel River, Pico Rivera, along the seismic reflection profile (Figure 4.3).
The yellow reflector marks the base of the Quaternary, Qt, Tfu, Pliocene Upper
Fernando Formation; Tp, Miocene Puente Formation.
112
Figure 4.2, Continued
113
kink-band that narrows towards the surface and is bounded by well-defined active,
(synclinal) and inactive (anticlinal) axial surfaces. The high-resolution data overlap with
the upper part of the industry data, and what appear to be the same prominent reflectors
are evident in both the industry and high-resolution profiles (Figure 4.2; upper pink and
yellow lines). Structural relief above the base Quaternary reflector (Figure 4.2; yellow
line) indicates that growth has occurred during deposition of the upper Quaternary strata.
The kink band decreases in width from ~395 m at the Quaternary reflector (yellow line,
Figure 4.2) to ~315 m at 125- to 240-m-depths for the shallower prominent reflector
above the base Quaternary (pink line, Figures 4.2). This “pink” reflector exhibits ~ 112
m of structural relief across the deformed zone. The width of the kink band for the
shallowest reflector visible on the high-resolution seismic reflection profile (purple line
on Figure 2B) is ~145 m, with ~20 of structural relief across the zone of deformation.
4.2.2 Borehole Observations
The shallowest reflectors imaged on the high-resolution mini-vibe profile are well
within the reach of standard geotechnical boreholes. In order to constrain the longer-term
slip rate for the PHT and provide a point of comparison for the slip rate calculated from
the study site above the forelimb of the central PHT segment, I drilled one deeper,
continuously-cored borehole to a depth of 66 m. I collected four very small charcoal
samples at depths of 35 m to 43 m, and 21 samples for optically stimulated luminescence
(OSL) dating.
The borehole reveals a stratigraphic section dominated by sand-and-gravel
channel deposits (Figure 4.3). The uppermost 35 m consists of medium to coarse-grained
114
Figure 4.3 Detailed Borehole Log from the Backlimb of the Santa Fe Springs
Segment of the Puente Hills Blind-Thrust Fault
Detailed borehole log from a 66-m-deep borehole acquired from the Pico Rivera site
above the backlimb of the Santa Fe Springs segment of the PHT (A2 in Figure 4.1).
This location is parallel to the seismic reflection acquisition shown in Figures 4.2
and 4.3. Colors to the left of the log indicate lithological changes in grain size
observed in the cores. Colors to the right reflect Munsell soil color. Red dots specify
the depth of charcoal samples collected from the cores (Table 4.1). White, vertical
lines indicate the depth range of samples collected for optically stimulated
luminescence dating (OSL) at Utah State University.
115
Figure 4.3 Continued
116
sand to gravel deposits. At 35 m there is a downward transition to a ~5-m-thick, very
fine-grained sand/ silt layer, separated from another fine-grained unit at 43 m by coarse-
grained sand and gravel. The consistency of the coarse-grained channel deposits
suggests that the location of the San Gabriel River channel has probably remained stable
throughout the Holocene time.
Age control is provided by four, very small charcoal fragments (Table 4.1). Due
to the coarse-grained stratigraphic section evident in the cores and the scarcity of high-
quality charcoal samples, 21 samples were collected for OSL dating. Nine samples were
sent to Utah State University for processing in September 2008 and results are still
pending. Of the four charcoal fragments, three samples yielded only minimum
radiocarbon dates (Table 4.1). One sample, (PHTPR-2) yielded a true radiocarbon age of
14,290+1460 years BP. This age was calibrated using Oxcal v.4.1.1 [Bronk Ramsey,
1995; Bronk Ramsey, 2001 (using atmospheric data from Reimer et al., 2004;
[IntCal04]); Bronk Ramsey et al., 2001], which uses a Bayesian statistical analysis and
stratigraphic ordering to calculate calendric dates [Bronk Ramsey, 2001]. The 13,624-
21,357 calendric date is reported as a two-sigma (95% confidence limit) age range.
Sediment accumulation, based on the depth and calibrated date of the charcoal sample
collected from the crest of the fold imaged in the high-resolution seismic reflection
profile, has occurred at a rate of ~1.7 to 2.7 mm/yr over the past 21 ka. Based on the
structural relief calculated for the shallowest reflector, the sediment accumulation rate
increases to ~2.6 to 4.1 mm/yr north of the fold panel.
117
Table 4.1 Calibrated, Calendric Dates and Radiocarbon Ages of Samples from the
Pico Rivera Study Site.
Sample
number
bh
Sample
type
True
depth
(m)
14
C Age BP
13
C Calendric age
(Cal BP)
%
proba-
bility
PHTPR-1 1 ccl 35.03 >18100 -25
PHTPR-2 1 ccl 36.36 14290 +
1460
-25 21357 – 13624 95.4
PHTPR-3 1 ccl 36.84 >21700 -25
PHTPR-4 1 ccl 43.07 >16900 -25
Calibrated, calendric dates and radiocarbon ages of samples collected from the deep
borehole at the Los Angeles County flood control study site, Pico Rivera, California.
All samples were calibrated using Oxford Radiocarbon Accelerator Unit (ORAU)
calibration program OXCAL v4.11 [Bronk Ramsey, 1995; 2001; using atmospheric
data from Reimer et al., 2004 [IntCal04]; Bronk Ramsey and van der Plicht, 2001].
For sample type; ccl=charcoal fragment; bulk=bulk soil. Calendric age is reported
as 2 sigma (95% confidence limit) age range. Bh indicates the borehole that the
sample was collected from. Calendric age is listed as calibrated years before present
(Cal BP).
118
Figure 4.4 High-Resolution Seismic Reflection Profiles above the Backlimb of the
Santa Fe Springs Segment of the Puente Hills Blind-Thrust Fault.
High-resolution seismic reflection profiles of shallow, subsurface folding above the
Santa Fe Springs segment of the Puente Hills blind-thrust fault (PHT). (A) Seismic
reflection profile acquired with a mini-sosie source, across the forelimb fold
structure showing an upward-narrowing zone of active folding (growth triangle)
bounded by sharply defined axial surfaces (dashed black lines). Data was acquired
along Carfax Avenue, Bellflower located by a C on Figure 4.1. Deformation is
localized along the anticlinal-axial surface (a). Reflectors imaged from ~400 - to 40-
m-depths. Forelimb growth triangle width decreases from ~200 m at 200 m depth to
<100 m at 50 m depth [Pratt et al., 2002]. (B) Seismic reflection profile acquired
with a mini-vibe source across the backlimb fold with deformation localized along
the active, synclinal axial surface (a). Reflectors imaged from ~600 m to ~45 m from
the surface. Backlimb growth triangle width decreases from ~625 m at 200 m depth
to <200 m at 50 m depth.
119
Figure 4.4 Continued
120
4.3 Discussion
The high-resolution seismic reflection data from the Pico Rivera site demonstrates
the continuity of discrete folding into the shallow sub-surface above the backlimb of the
central, Santa Fe Springs segment of the PHT, consistent with shallow profiles acquired
above the forelimb fold structure (Figure 4.4). The combined profiles provide a near-
continuous image of both forelimb and backlimb growth triangles associated with slip on
the Santa Fe Springs segment of the PHT, from the tip of the fault ramp at ~3 km depth to
within ~40 m of the surface. This combined data set demonstrates that deformation is
localized across the active, axial surfaces, largely consistent with kink band migration
folding mechanisms [Suppe et al., 1992; Shaw and Suppe, 1994; Shaw et al., 2002].
4.3.1 Fault Slip Rate Estimates
I combine the structural relief across the shallow backlimb fold imaged in the
high-resolution seismic reflection profile with the dip of the underlying ramp to calculate
slip on the underlying thrust fault. The dip of Santa Fe Springs segment is defined by
fault-plane reflections and reflection truncations imaged in petroleum industry seismic
reflection profiles as 27+2
o
(Figure 4.2A) [Shaw and Shearer, 1999; Shaw et al., 2002]. I
divide the structural relief estimated from the seismic profiles by the sine of the dip of
thrust ramp to derive a cumulative displacement. Structural relief calculated for the
shallowest reflector is 20 m. Assuming rigid-block translation of the hanging wall yields
41.2 to 47.3 m of slip on the blind-thrust ramp. By dividing displacement by the
calibrated radiocarbon date sampled above the shallowest reflector, I derive a preliminary
Holocene-Late Pleistocene slip rate of 1.9 to 3.0 mm/yr. This rate is significantly faster
121
than the slip rate calculated for the forelimb of the Santa Fe Springs segment (0.9 to 1.6
mm/yr; Dolan et al, [2003]; Leon et al., [2007]).
Uncertainties in the velocity structure used to depth process the seismic reflection
data lead to uncertainties in the depths of reflectors and thus uncertainties in the slip rate.
Velocities used to process Figure 4.2 were relatively slow, leading to shallow reflector
depths. By increasing the velocities, the depths of reflectors increase considerably
(Figure 4.5). For example, the ~37 m depth of shallowest reflector in the slow-velocity
profile (4.5A) is located at ~67 m in the high-velocity profile (4.5C). This source of error
could thus decrease the slip rate by at least 0.5 mm/yr, to 1.5 to 2.5 mm/yr. Additional
analysis of the high-resolution seismic reflection data will provide tighter constraints on
the velocity structure used to depth-process the seismic section, and thus reduce
uncertainties in slip-rate calculations. Moreover, the results from our OSL samples
should provide better timing constraints and allow us to determine a more precise long-
term slip rate for the Santa Fe Springs segment of the PHT. Comparison of fault slip for
both the forelimb and backlimb study sites (Pico Rivera and Bellflower, respectively)
may provide insight into the amount of slip consumed up-dip of the backlimb study site
by slip on minor hangingwall back-thrusts and/or distributed deformation within the
hanging wall.
4.3.2 Incremental Development of Dips within the Kink Band
The seismic reflection data indicate that strata folded within the kink band acquire
their dips incrementally. For example, analysis of deep penetration petroleum industry
seismic reflection data indicates that all the strata in the upward-narrowing forelimb
122
growth triangle dip ~23° between the top of the thrust ramp at ~3 km and the shallowest
imaged reflectors at 250 m (Figure 4.2A; [Shaw and Shearer, 1999; Shaw et al., 2002]).
As in the high-resolution seismic reflection data from the forelimb of the Santa Fe
Springs segment (Figure 4.4A), the high-resolution seismic profiles acquired from the
Pico Rivera backlimb site reveal an incremental downward increase in dip ~37 to 240 m
depths. The shallowest reflector (at 37 to 70 m depths) dips to the north ~8
o
, with deeper
reflectors increasing in dip to ~11
o
(48 m depth), 13
o
(65 m depth), and 15
o
(128 m
depth). The prominent reflector evident in both the high-resolution and industry profiles
at 125 – 240 m depths (pink line) dips to the north ~23
o
.
The progressive downward increase in bed dip genuinely reflects an aspect of fold
kinematics distinct from classic growth fault-bend fold theories. This change in limb dips
may reflect some component of fold growth by limb rotation in addition to kink-band
migration. These hybrid kinematics are manifested in several types of fault-related folds,
including trishear fault-propagation folding [Erslev, 1991; Allmendinger, 1998] and shear
fault-bend folding [Suppe and Connors, 2004; Shaw et al., 2005]. Alternatively, the
component of limb rotation may reflect curvature of the fold hinge, as is described by
curved-hinge fault-bend folding theories [Suppe et al., 1997; Novoa et al., 2000].
Finally, the progressive change in bed dips may reflect the mechanical response of
loosely consolidated, granular sediments in the shallow subsurface to folding at depth.
Recent work by Benesh et al. [2007], for example, demonstrates that, even in cases of
folding of pre-tectonic layers exclusively by kink-band migration, shallow growth
sediments
123
Figure 4.5 High-Resolution Seismic Reflection Profiles Processed with Different
Velocities for Shallow Sediments
(A) Seismic profile from Figure 4.2, depth-migrated with (A) slow velocity depth
conversion, (B) intermediate-velocity depth conversion, and (C) fast-velocity depth
conversion (migrated; 2x vertical exaggeration). Data processing was routine for
seismic reflection profiles and included velocity analysis, residual statics corrections,
pre- and post-stack bandpass filter and deconvolution, Stolt time migration, and
time-to-depth conversion. Profiles were processed by Thomas L. Pratt at the
University of Washington, Seattle with Seismic Unix software.
124
Figure 4.5, Continued
125
acquire their dips progressively. This is due to the finite width of the axial surface zone
and its rotation and migration with increasing deformation.
4.4 Conclusions
In this chapter I combine high-resolution seismic reflection profile acquired at the
updip projection of the active axial surface located in industry seismic profiles with data
from a 66-m-deep continuously cored borehole to provide preliminary calculations of
structural relief across the backlimb fold structure of the Santa Fe Springs segment of the
PHT. Deformation above the base-Quaternary reflector is localized across the active,
synclinal axial surface with an upward-narrowing growth triangle consistent with late
Quaternary activity of the PHT. The kink band width decreases from ~390 m at the base-
Quaternary reflector to ~137 m for the shallowest reflector imaged on the seismic
reflection profile (~37 m). Prominent reflectors in the high-resolution seismic reflection
profile reveal the incremental increase in dip from 8
o
at ~37 m to 23
o
at 240 m depth.
I derive a preliminary Holocene-late Pleistocene slip rate of 1.9 to 3.0 mm/yr for the
Santa Fe Springs segment of the PHT. This rate is higher than the 0.9 to 1.6 mm/yr
previously published Holocene slip rate from a study site above the forelimb of the
central PHT segment [Dolan et al., 2003; Leon et al., 2007 (Chapter 2)], likely as a result
of some slip being consumed by hangingwall deformation that occurs updip of the
backlimb study site.
Additional analysis of the shallow velocity structure of the Pico Rivera site
combined with tighter timing constraints provided by OSL dates, should provide a
intermediate slip rate for central PHT segment and may provide insight into the amount
126
of deformation consumed by hanging wall deformation or minor backthrusts updip of the
backlimb study site.
127
Chapter 4 References
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Hauksson, E., and Jones, L. (1989), The 1987 Whittier Narrows earthquake sequence in Los
Angeles, southern California: Seismological and tectonics analysis: J. Geophys. Res.,
94, 9569.
Ishiyama, T., K. Mueller, M. Togo, A. Okada, and K. Takemura (2004), Geomorphology,
kinematic history, and earthquake behavior of the active Kuwana wedge thrust
anticline, central Japan, J. Geophys. Res., 109, B12408, doi:
10.1029/2003JB002547.
Ishiyama, T., K. Mueller, H. Sato, and M. Togo (2007), Coseismic fault-related fold
model, growth structure, and the historic multisegment blind thrust earthquake on
the basement-involved Yoro thrust, central Japan, J. Geophys. Res., 112, B03S07,
doi: 10.1029/2006JB004377.
128
King, G. C. P., and C. Vita-Finzi (1981), Active folding in the Algerian earthquake of 10
October 1980, Nature, 292, 22-26.
Lai, K., Y. Chen, J. Hung, J. Suppe, L. Yue, and Y. Chen (2006), Surface deformation
related to kink-folding above an active fault: Evidence from geomorphic features
and co-seismic slips, Quat. Int., 147, 44-54.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Pratt, T. L., J. H. Shaw, J. F. Dolan, S. A. Christofferson, R. A. Williams, J. K. Odum,
and A. Plesch (2002), Shallow folding imaged above the Puente Hills blind-thrust
fault, Los Angeles, California, Geophys. Res. Lett., 29, 18-1 - 18-4.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
129
Stein, R. S., and R. S. Yeats (1989), Hidden earthquakes, Sci. Am., 260, 48-57.
Sugiyama, Y., K. Mizuno, F. Nanayama, T. Sugai, H. Yokota, T. Hosoya, K. Miura, K.
Takemura, and N. Kitada (2003), Study of blind thrust faults underlying Tokyo
and Osaka urban areas using a combination of high-resolution seismic reflection
profiling and continuous coring, Annals of Geophys., 46, 5.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
130
CHAPTER 5:
Holocene-Latest Pleistocene Slip Rates for the Coyote Hills Segment of the Puente
Hills Blind-Thrust Fault
5.1 Introduction
The occurrence of several destructive earthquakes in the past few decades (e.g.,
1980 M
s
7.3 El Asnam, Algeria, 1987 M
w
6.0 Whittier-Narrows, California, 1994 M
w
6.7
Northridge, California) has focused attention on faults that do not cut the earth’s surface,
(termed “blind” faults) but instead are manifested by young near-surface folding. The
damage caused by these earthquakes highlights the need to better understand the seismic
hazards posed by these structures to an increasingly urbanized global population. The
absence of paleoseimologic data from blind-thrust faults, however, creates a major gap in
our ability to include this class of fault into probabilistic seismic hazard models. To
bridge this data gap I must study the youngest folded strata to obtain the information
necessary to understand the recent seismic behavior of these faults. Such data are critical
if I wish to understand how folds grow in response to slip on blind-thrust ramps at depth
and provide paleoseimologic data for hazard analysis.
Structural and paleoseismologic analysis of the Santa Fe Springs and Los Angeles
segment study sites has provided Holocene-Latest Pleistocene slip rates, slip-per-event
data, and paleo-magnitudes for the central and western segments of the Puente Hills
blind-thrust fault (PHT). A similar study of young folding above the Coyote Hills
segment would, therefore, facilitate analysis of shallow-subsurface folds above the entire
segmented blind-thrust system. In order to achieve this goal, I herein combine existing
petroleum-industry data from the western Coyote Hills segment of the PHT [Shaw et al.,
131
2002] with a high-resolution seismic reflection profile and continuously-cored boreholes
above the eastern Coyote Hills segment. These data document the geometry of young
folds that have developed above the forelimb of the Coyote Hills segment of the Puente
Hills thrust fault (PHT). I provide alternative stratigraphic correlations of shallow, sub-
surface folding in order to generate preliminary slip rates for the eastern Coyote Hills
segment, and discuss the data with respect to their implications for fold kinematics and
seismic hazard assessment.
5.1.1 Puente Hills Thrust Fault
The Puente Hills thrust (PHT) is a major blind thrust fault, originally identified by
Shaw and Shearer [1999], that extends for >40 km beneath the northern part of the Los
Angeles metropolitan region, from near Beverly Hills east-southeastward beneath
downtown Los Angeles to northern Orange County (Figure 5.1). Displacements on the
PHT have led to a series of anticlines that developed above each of the fault segments
during slip on the underlying thrust ramps in the northern Los Angeles Basin [Shaw and
Suppe, 1996; Shaw et al., 2002]. All of these folds are bounded on their southern
margins by narrow forelimbs or kink-bands with folded, south-dipping Pliocene strata
overlain by Pleistocene and Quaternary units that thin across the crest of the folds. Shaw
et al. [2002] documented the subsurface geometry of the PHT using petroleum industry
seismic reflection and well data. They identified three right-stepping, en-echelon blind
thrust ramps that dip to the north at ~ 25-30
o
and terminate upwards at ~ 3 km depth.
These fault segments are, from west to east, the Los Angeles segment, the Santa Fe
Springs segment, and the Coyote Hills segment. In contrast to the subtle surface
132
Figure 5.1 Structure Contour Map of the Puente Hills Blind-Thrust Fault
Structure contour map of the three en echelon segments of the Puente Hills blind-
thrust fault (PHT) showing the location of the 1987 Whittier Narrows (M
w
6.0)
earthquake sequence [Hauksson and Jones, 1989] as relocated by Shaw and Shearer,
[1999], modified after Shaw et al., [2002]. A0
- A2 marks the trace of the seismic
reflection profile shown in Figure 5.2A. B0 - B1 marks the trace of the mini-vibe
source profile (Figure 5.2B and 5.3B). T marks the trace of the high-resolution
seismic reflection profile acquired along Trojan Way in La Mirada (Figure 5.3A).
The inset shows the location of the PHT and 1994 Northridge (M
w
6.7) earthquake.
Major state and interstate highways are show for reference Map coordinates are
UTM Zone 11, NAD27 datum.
133
expression of the Santa Fe Springs and Los Angeles segments of the PHT, the
easternmost folds form the Coyote Hills – a prominent topographic uplift - and provide
structural closure for the east and west Coyote Hills oil fields [Yerkes, 1972]. Direct
evidence that the PHT is active and capable of causing damaging earthquakes was
provided by the 1987 M
w
6.0 Whittier-Narrows earthquake, which was generated by
rupture of ~10% of the central, Santa Fe Springs segment of the PHT [Shaw and Shearer,
1999].
The Coyote Hills segment of the PHT extends from the eastern edge of Los
Angeles County, beneath northern Orange County (Figure 5.1). Based on petroleum-
industry seismic reflection profiles, it is separated by a shallow offset into the eastern and
western Coyote Hills segments, which may crosscut one another or merge into a common
basal ramp at ~5 km [Shaw et al., 2002]. The western Coyote Hills anticline, which is
broad and open, similar to the Santa Fe Springs structure, is bounded on its southern
margin by a discrete forelimb kink band that exhibits parallel, south-dipping strata. The
western Coyote Hills thrust ramp dips to the north between 25
o
to 30
o
, as defined by fault
plane reflections and reflection truncations in seismic profiles between 2- and 5-km
depths. The strike of this segment is roughly east-west, similar to the Santa Fe Springs
segment of the PHT. In contrast, the eastern Coyote Hills segment mimics the strike of
the eastern Coyote Hills anticline, which strikes about N60
o
E [Wright, 1991]. An older
petroleum-industry seismic reflection survey images the eastern Coyote Hills anticline,
but does not provide a direct image of the fault. The older data do, however, image a
steep kink band on the southern margin of the fold. The syncline imaged in the industry
134
data corresponds with the surface fold scarp, consistent with slip being entirely consumed
in the kink band by some type of fault-propagation folding processes. Alternatively, the
thrust ramp may shallow into an upper detachment and backthrust forming a structural
wedge [Shaw et al., 2002]. Extrapolating the PHT beneath the eastern Coyote Hills
anticline from the western Coyote Hills seismic reflection data yields a simple, planar
surface for the fault that dips to the north-northwest at about 25-30
o
.
The most recent deformation associated with slip on the PHT thrust ramps is
recorded in the youngest sediments at the top of the growth triangle. The petroleum
industry reflection data, however, typically do not image strata within the uppermost 200
– 300 m. In order to bridge the resulting data gap from the shallowest industry data at
~200 m depth to the surface, I acquired high-resolution seismic reflection data (Figure
5.2B) and a series of continuously-cored boreholes (Figure 5.3) along a north-south
transect, approximately perpendicular to the trace of the active, synclinal axial surface
above the forelimb of the eastern Coyote Hills segment of the PHT.
5.2 Results
The study site is located along North Raymond Avenue in the City of Fullerton,
Orange County, along the distal floodplain of the Santa Ana River, a major south-flowing
drainage (Figure 5.1). The site is ~2.5 km northwest of the current river channel. The
north-south seismic reflection profiles and boreholes are oriented perpendicular to the
prominent fold scarp of the east Coyote Hills, which rises 72 m in elevation from East
Chapman Avenue (B0 in Figure 5.2) to Linda Lane at the crest of the hill (B1 in Figure
5.2).
135
Figure 5.2 Multi-Scale Seismic Reflection Profiles of Folding above the Coyote Hills
Segment of the Puente Hills Blind-Thrust Fault
(A) Seismic reflection image of the forelimb fold structure showing upward
narrowing zone of folding (growth triangle) delimited by sharply defined axial
surfaces [Shaw et al., 2002]. Dashed black lines represent active, synclinal and
inactive, anticlinal axial surface (A0 - A2 Figure 5.1). (B) Mini-vibe-source seismic
reflection profile in two way travel time (B0 - B1 Figure 5.1). Vertical yellow lines
show borehole locations. Dashed black lines represent the active synclinal axial
surface. Blue line represents prominent reflector at ~100- to 225-m-depth.
136
Figure 5.2, Continued
137
5.2.1 High-Resolution Seismic Reflection Data
I collected a north-south high-resolution seismic reflection profile (Figure 5.2B)
parallel to, and ~10 km east of, the petroleum industry profile (A – A2) acquired along a
north-south transect above the western Coyote Hills segment shown in Figure 5.2A. The
seismic reflection profiles are oriented approximately perpendicular to the active,
synclinal axial surface associated with the PHT forelimb growth triangle, which extends
approximately east-west beneath the Fullerton study site [Shaw et al., 2002].
I used a mini-vibe source to image the ~100 to 400 m depth range (Figure 5.2B
shown in two way travel time [twtt]). Prominent, shallowly dipping (<5
o
) reflectors
located at the south end of the transect are discernible on the mini-vibe profile. A
prominent south-dipping reflector is also discernible between 100- to 225-m depths (blue
line Figure 5.2B). In the Raymond Avenue high-resolution seismic reflection profile
dipping reflectors are localized between an active, synclinal axial surface and an inactive,
anticlinal axial surface (Figure 5.2B and 5.3B). A high-resolution seismic reflection
profile acquired along Trojan way above the western Coyote Hills segment of the PHT
also indicates localization of deformed reflectors along an active, synclinal axial surface
(Figure 5.3A). These reflections are consistent with the upward continuation of the kink
band identified at depth by Shaw et al. [2002] on the deeper-penetration industry profile
for the western Coyote Hills segment of the PHT (Figure 5.2A).
5.2.2 Borehole Data
In order to document the details of recent fold growth above the eastern Coyote
Hills segment of the PHT, I excavated a north-south transect of eight, 25- to 35-m-deep,
138
Figure 5.3 High-Resolution Seismic Reflection Profiles of Shallow Folding above the
Eastern and Western Coyote Hills Segments of the Puente Hills Blind-Thrust Fault
(A) Mini-sosie source seismic reflection profile of the shallow forelimb fold
structure acquired along Trojan Way, La Mirada (T; Figure 5.1 [Shaw et al., 2002]).
Deformed strata corresponding to shallow sub-surface reflections are localized
along the active, synclinal axial surface (a; black dashed line), which corresponds to
the active, synclinal axial surface indicated in Figure 5.2A.The profile above the
seismic reflection profile represents the surface elevation of the study site (Pratt et
al., 2002). (B) Mini-vibe-source seismic reflection profile in two way travel time (B0
- B1 Figure 5.1). Vertical yellow lines show borehole locations. Dashed black lines
represent the active synclinal axial surface.
139
Figure 5.3, Continued
140
continuously cored boreholes across the zone of young folding imaged in the high-
resolution seismic reflection profiles (Figures 5.2B). This transect extends from the
gently south-sloping alluvial surface upward along the south-dipping slope of the Coyote
Hills. Excavation of boreholes allows me to document the geometry of shallow folding,
as well as to sample the folded sediments for age control. Sediments in the cores consist
predominantly of fine-to-coarse-grained sand and gravel units interbedded with cohesive,
fine-grained silts and clays (Figure 5.4).
I identified seven distinctive stratigraphic units based upon grain size, texture, and
Munsell soil color. Major units are denoted by intervals of 10, from youngest to oldest
(e.g.,10, 20, 30, and 40, and 50). Five units are traceable continuously across the entire
length of the borehole transect. Unit 10 consists of cohesive, sandy-silt to clay, whereas
Units 30 through 50 are predominantly friable silty-sand to gravel channel deposits. Unit
20 is not present in PHTCH5, PHTCH6, and PHTCH7. Although each Unit exhibits
similar stratigraphic sequences defined by its distinctive grain size, color, and texture, I
am unable to make a unique correlation between boreholes PHTCH4 and PHTCH2 from
the base of Unit 20 through the top of Unit 50 within. The shallow Unit 20 sand and
gravel is very similar in color and texture to the lower Unit 50 sand and gravel present in
borehole PHTCH2. I have therefore provided alternative cross-sections showing differing
possible stratigraphic correlations for these units. I also identified two, laterally
continuous, organic-rich clay and silt layers, which are numbered by the associated
overlying major unit. Specifically, the organic-rich soil beneath Unit 20 is identified as
Unit 21, and the soil beneath Unit 40 is identified as Unit 41. There is an additional
141
Figure 5.4 Alternative Cross-Sections of Borehole Results from the North Raymond
Avenue Transect
Cross-section of major stratigraphic units (5x vertical exaggeration). Vertical black
lines indicate the borehole locations. Colors denote different sedimentary units.
Green line at the top is the ground surface at the study site. (A) Cross-section 1;
stratigraphic correlation of sedimentary units where the tops of Units 30 through 50
are deeper than in cross-section 2 (part B) (see text for discussion). (B) Cross-
section 2; shallow stratigraphic correlation of Units 30 through 50 in boreholes
PHTCH8 and PHTCH2 increases the measured structural relief across the fold.
142
Figure 5.4, Continued
143
organic-rich soil between Units 30 and 40 that is not laterally continuous and may have
been eroded during deposition of the overlying sand unit. This cohesive unit (Unit 31) is
not present in boreholes PHTCH1, PHTCH3, PHTCH6, and PHTCH8.
The organic-rich clay and silt layers provided abundant detrital charcoal and bulk-
soil samples. Of the 24 charcoal samples collected, six samples were sent to Lawrence
Livermore National Laboratories for radiocarbon dating to constrain the depositional ages
of the strata exposed in the cores. Accelerator Mass Spectrometer (AMS) radiocarbon
(
14
C) analyses of two samples provided reliable radiocarbon dates, with four samples
providing only minimum dates (Table 5.1). Samples were calibrated using the Oxford
Radiocarbon Accelerator Unit (ORAU) calibration program OXCAL v4.11 [Bronk
Ramsey, 1995; 2001; using atmospheric data from Reimer et al., 2004 (IntCal 04); Bronk
Ramsey and van der Plicht, 2001]. The calibrated date for sample PHTCH2-4 indicates
an average sediment accumulation rate over the past ~25 ka of ~1.0 to 1.5 mm/yr. Sample
PHTCH3-3 also indicates a sediment accumulation rate of ~1.0 to 1.5 mm/yr. This
sediment accumulation rate is similar to rates calculated for our Santa Fe Springs and Los
Angeles segment study sites (Chapters 2 and 3, respectively).
The borehole data reveal evidence for a discrete, upward-narrowing zone of
folding localized at the updip projection of the synclinal axial surface identified in the
high-resolution seismic reflection data (Figure 5.2B). Specifically, the synclinal axial
surface projects to the surface at a point between boreholes PHTCH3 and PHTCH5. I
identify 13.2 + 0.5 m of structural relief between the north and south ends of the transect
caused by growth in Units 5 through 41 in alternative cross-section 1 (Figure 5.4A). In
144
Table 5.1 Calibrated, Calendric Dates and Radiocarbon Ages of Samples from the
North Raymond Avenue Transect
Sample # bh
Sample
type
True
depth
(m)
Projected
depth to
bh 3 (m)
14
C
Age
BP
13
C Calendric
age (Cal
BP)
%
proba-
bility
PHTCH2-1 2 ccl 10.37 >31600-25
PHTCH2-4 2 ccl 17.19 29.42 21260
+ 1110
-25>23085 95.4
PHTCH3-3 3 ccl 24.27 24.27 16580
+ 1780
-25 25315 -
25133 &
25024 -
16421
0.6
94.8
PHTCH4-4 4 ccl 15.32 >23000 -25
PHTCH6-1 6 ccl 20.15 >35800-25
PHTCH6-2 6 ccl 27.95 >30000-25
Samples were calibrated using Oxford Radiocarbon Accelerator Unit (ORAU)
calibration program OXCAL v4.11 [Bronk Ramsey, 1995; 2001; using atmospheric
data from Reimer et al., 2004 [IntCal 04]; Bronk Ramsey and van der Plicht, 2001].
For sample type; ccl=charcoal fragment. Calendric age is reported as 2 sigma (95%
confidence limit) age range. Bh indicates the borehole the sample was collected.
Calendric age is listed as calibrated years before present (Cal BP).
145
alternative cross-section 2, I observe 29.4 + 1.4 m of structural relief across the borehole
transect (Figure 5.4B). The borehole data demonstrate that all strata dip parallel to the
street surface at the southern end of the transect (in boreholes PHTCH1, PHTCH3, and
PHTCH5) with Units 10, 30, and 40 thickening to the south (Figure 5.4).
The zone of deformation identified in the borehole transect also corresponds with
the marked inflection point at the southern base of the Coyote Hills. This is consistent
with the observations from the seismic reflection data that the south-facing slope of the
Coyote Hills at the study site is a fold scarp that corresponds with the forelimb of the
eastern Coyote Hills fold.
5.3 Discussion
The high-resolution seismic reflection and borehole data from the North Raymond
Avenue site provide a near-continuous image of the shallow forelimb growth triangle
associated with slip on the eastern Coyote Hills segment of the PHT, from ~400 m depth
to the surface. This combined data set demonstrates that folding is continuous and
localized across the active, synclinal axial surface (Figures 5.2 and 5.4). These
observations confirm that the forelimb of the eastern Coyote Hills anticline is growing, at
least in part, by kink-band migration in a manner broadly consistent with growth fault-
bend fold theory [Suppe et al., 1992; Shaw and Suppe, 1994; Shaw et al., 2002].
5.3.1 Fault Slip Rate Estimates
To determine the reverse slip rate of the fault, I combine the measured minimum
uplifts with the dip of the causative Coyote Hills segment of the PHT. The fault dip is
defined by the dipping fault plane reflectors (27+2
o
) imaged in industry seismic reflection
146
profiles for the western Coyote Hills segment and corrected for apparent dip [Shaw et al.,
2002]. The data defining the relationship between the fold and fault were extrapolated
from petroleum-industry seismic reflection data for the eastern Coyote Hills segment by
positioning the fault at the base of the forelimb kink band. The displacements and rates
discussed below are based on the assumption that no deep slip on the thrust ramp is
accommodated by distributed hangingwall deformation. Thus all of these rates should be
considered minima.
The alternative stratigraphic correlations (Figure 5.4) provide different
calculations of structural relief for the eastern Coyote Hills segment. Dividing by the
sine of the dip of the underlying thrust ramp (27+2
o
), I calculate total slip as 29 +3.4/-2.8
m for alternative cross-section A (Figure 5.4A). A charcoal sample from the top of the
organic-rich cohesive section (Unit 41) above the basal Unit 50 sand provides a
calibrated radiocarbon age of 25,024 to 16,421 Cal YBP. The slip rate for alternative
cross-section A is therefore 1.0 to 2.0 mm/yr. An additional charcoal sample from the
top of Unit 50 yielded a minimum calibrated age of 23,085 YBP, which would suggest a
slip rate of ~1.35 mm/yr. But as this is a maximum age, this would yield a maximum slip
rate. These slip rates are all similar to the long-term slip rate calculated by Shaw et al.,
[2002] of 0.9 to 1.7 mm/yr with a preferred rate of 1.3 +0.4 mm/yr. This suggests the
possibility that the slip rate for the Coyote Hills segment of the PHT has not varied
significantly during Quaternary time. In version B of the cross-section (Figure 5.4B),
total growth is 29.4 m. Dividing by the sine of the dip of the Coyote Hills thrust ramp,
yields total slip of ~65 m. Dividing by the calibrated 16.4 – 25.0 ka date from a charcoal
147
sample collected for borehole PHTCH3 yields a Holocene-late Pleistocene slip rate of 2.6
to 3.9 mm/yr. This slip rate is considerably faster than the calculated long-term slip rate
of Shaw et al. [2002]. Additional, more closely spaced boreholes, combined with further
radiocarbon dating, should reveal unique stratigraphic correlations which will provide a
more precise Holocene-late Pleistocene slip rate (and possibly paleo-earthquake ages and
displacements) for the eastern Coyote Hills segment of the PHT. Moreover, acquiring
additional data would help to identify buried paleo-fold scarps formed by uplift events on
the Coyote Hills segment of the PHT and allow analysis of the entire segmented blind-
thrust system.
5.4 Conclusions
Continuously cored boreholes and high-resolution seismic reflection data provide
a record of shallow, sub-surface folding above the eastern Coyote Hills segment of the
Puente Hills blind thrust fault, from ~400 m depth to the surface. These data demonstrate
that folding is continuous at Holocene-late Pleistocene time scales and provide a record
of sedimentary growth related to uplift of the East Coyote Hills anticline. I calculate a
Holocene/Late Pleistocene slip rate of 1.0 to 2.0 mm/yr, with a preferred slip rate of 1.3
mm/yr. This preferred rate is similar to the 1.3 +0.4 mm/yr long-term average slip rate
calculated by Shaw et al., [2002] for the western Coyote Hills segment.
148
Chapter 5 References
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Hauksson, E., and Jones, L. (1989),The 1987 Whittier Narrows earthquake sequence in Los
Angeles, southern California: Seismological and tectonics analysis: J. Geophys. Res.,
94, 9569.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Wright, T. L. (1991), Structural geology and tectonic evolution of the Los Angeles basin,
California, in Active Margin Basins, edited by K. T. Biddle, AAPG Mem., 52, 35-
134.
Yerkes, R. F., McCulloh, T. H., Schoellhamer, J. E., and Vedder, J. G. (1965), Geology
of the Los Angeles basin, California – An introduction, U.S. Geol. Surv. Prof.
Pap., 420-A, 57.
149
CHAPTER 6:
Evidence for Large Holocene Earthquakes on the Compton Thrust Fault, Los
Angeles, California
Abstract
I demonstrate that the Compton blind-thrust fault is active and has generated at
least six large-magnitude earthquakes (M
w
7.0-7.4) during the past 14,000 years. This
large, concealed fault underlies the Los Angeles metropolitan area, and thus poses one of
the largest deterministic seismic risk in the United States. I employ a methodology that
uses a combination of high-resolution seismic reflection profiles and borehole
excavations to link blind faulting at seismogenic depths directly to near-surface fault-
related folding. Deformed Holocene strata record recent activity on the Compton thrust
and are marked by discrete sequences that thicken repeatedly across a series of buried
fold scarps. I interpret the intervals of growth as occurring after the formation of now-
buried paleo-fold scarps that formed during uplift events on the underlying Compton
thrust ramp. Minimum uplift in each of the scarp-forming events, which occurred at 0.7-
1.75 ka [Event 1], 0.7-3.4 ka or 1.9-3.4 [Event 2], 5.6-7.2 ka [Event 3], 5.4-8.4 ka [Event
4], 10.3-12.5 ka [Event 5], and 10.3-13.7 ka [Event 6], ranged from ~0.6 to ~1.9 m,
indicating minimum thrust displacements of ≥1.3 to 4.2 m. Such large displacements are
consistent with the occurrence of large-magnitude earthquakes (M
w
>7). This multi-
disciplinary methodology provides a means of defining the recent seismic behavior, and
therefore the hazard, for blind-thrust faults that underlie other major metropolitan regions
around the world.
150
6.1 Introduction
The recent occurrence of several highly destructive earthquakes on thrust faults
(e.g., 1994 M
w
6.7 Northridge, California, 1999 M
w
7.6 Chi-Chi, Taiwan, 2005 M
w
7.5
Kashmir, Pakistan, and 2008 M
w
7.9 Wenchuan, China) highlights the need to better
understand the seismic hazards posed by these structures to an increasingly urbanized
global population. The keys to this understanding are the recognition of the rate at which
these faults store and release seismic energy, and the acquisition of detailed
paleoseismological data on the ages and displacements of ancient earthquakes that they
have generated. Although such data are routinely documented for many faults around the
world, one class of faults, the so-called “blind thrust faults”, in which the fault plane does
not extend all the way to the Earth’s surface, has proved particularly difficult to study.
Unlike most faults, which rupture to the surface in large earthquakes, near-surface
deformation above blind thrust faults is accommodated by folding, rather than faulting. In
many thrust fault earthquakes, co-seismic fold growth has been documented [King and
Vita-Finzi, 1981; Stein and King, 1984; Stein and Yeats, 1989; Lin and Stein, 1989; Lai et
al., 2006; Chen et al., 2007; Streig et al., 2007] and in very large thrust fault earthquakes,
discrete coseismic fold scarps have developed above bends and at the tips of propagating
faults [Stein and Yeats, 1989; Ishiyama et al., 2004; 2007; Lai et al., 2006; Chen et al.,
2007; Streig et al., 2007]. The youngest folded strata thus contain the information
necessary to understand the recent seismic behavior of these faults. However, extracting
paleoseismological data from blind thrusts has proved to be challenging, although the
development over the past several years of new inter-disciplinary methodologies for
151
documenting the sizes and ages of past events is beginning to yield a new understanding
of the behavior of this problematic class of faults at the time scales of individual
earthquakes [e.g., Dolan et al., 2003; Sugiyama et al., 2003; Ishiyama et al., 2004; 2007;
Leon et al., 2007]. The difficulties in studying the paleoseismology of blind thrusts are
particularly well illustrated by the Compton thrust, a major blind-thrust fault beneath
metropolitan Los Angeles. Although this fault was declared to be inactive on the basis of
earlier studies [Mueller, 1997], my investigations demonstrate that the Compton fault is
indeed active and capable of generating large-magnitude (M
w
> 7) earthquakes. In this
chapter I discuss the implications of these results for seismic hazard assessment in Los
Angeles and describe a standard, multi-disciplinary methodology that can be used to
define seismic hazards posed by similar structures that underlie many densely populated
urban centers around the world.
6.1.1 Compton blind-thrust fault
Nowhere are blind thrust faults of more concern than beneath metropolitan Los
Angeles, home to more than 10 million people. Previous studies have identified several
major blind thrust faults beneath the city, with two – the Compton and Puente Hills faults
– being of particular concern because of their location directly beneath the urban center
[Shaw and Suppe, 1996; Shaw et al., 2002]. The Compton thrust fault was originally
identified by Shaw and Suppe [1996] using petroleum-industry seismic reflection profiles
and well data. Based on their analysis of the folding that occurs above the thrust fault,
they showed that the Compton fault extends northwest-southeast for ~40 km beneath the
western edge of the Los Angeles metropolitan region (Figure 6.1). The industry seismic
152
Figure 6.1 Map of the Compton Thrust and Adjacent Structures in the Los Angeles
Basin
My study site is located at the locus of active, backlimb folding along the northern
segment of the fault (white star). C – C’ marks the trace of the seismic reflection
profile shown in figure 3C. Also shown are the locations of Mueller [1997] study
sites located at the north-western end of the northern segment in Inglewood (white
triangle), and the south-eastern end of the southern segment in Los Alamitos (white
circles).
153
reflection data define a growth fault-bend fold associated with the gently northeast-
dipping thrust fault ramp, which, combined with well data, reveal compelling evidence
for Pliocene and Pleistocene activity. But evidence for recent (latest Pleistocene-
Holocene) activity of the fault was lacking, as the original seismic reflection data did not
image the uppermost 200 m of the stratigraphic section. Moreover, in marked contrast to
the Puente Hills blind thrust fault, which generated the 1987 moment-magnitude (M
w
) 6.0
Whittier-Narrows earthquake southeast of Los Angeles [Shaw and Shearer, 1999], the
Compton ramp has produced neither damaging historical earthquakes nor abundant,
instrumentally recorded micro-earthquakes that might assist in delineating the fault and
proving that it is seismically active.
6.2 Results
Successful analysis of the current state of activity and earthquake history of blind
faults such as the Compton thrust requires a multi-disciplinary methodology that relates
near-surface deformation to the deep seismogenic fault through observation of the entire
depth range of folding. I begin by analyzing available data on the location, geometry, and
extent of the folds and their relationship to the underlying thrust fault. Such data may
include seismicity on the thrust fault, deep seismic reflection images and well data,
geological mapping, and geomorphologic analysis.
6.2.1 High-Resolution Seismic Reflection Data
In the case of the Compton fault, I have access to numerous, high-quality
petroleum industry seismic reflection profiles that provide images of the upper 4-5 km of
the Compton-Los Alamitos fold. To bridge the gap between the petroleum industry
154
reflection data and the surface, I acquired higher-resolution seismic reflection data at two
different depth scales. Specifically, I used a Geometrics seismograph, available from
IRIS, using a 60-channel system with a weight-drop source to image the 30- to 600-
m-depth range, and a hammer source for 10- to 100-m-depth range. Geophones were
placed along the side of the street at 5-m intervals for the weight-drop source profile.
Four impacts from a truck-mounted Geometrics PWD weight-drop source (82 kg) using
elastomer technology were recorded at each source point (5 m apart). For the higher-
resolution seismic reflection profile with a hammer source (4.5 kg), geophones were
placed at 2-m intervals with four impacts recorded at 1-m intervals. Data processing was
routine for seismic reflection profiles and included velocity analysis, residual statics
corrections, pre- and post-stack bandpass filter and deconvolution, Stolt time migration,
and time-to-depth conversion. The combined industry and high-resolution seismic
reflection data provide an uninterrupted image of the full depth extent of the fold above
the Compton thrust ramp, from >4 km depth to within 20 m of the surface, allowing me
to accurately site boreholes across the locus of most recently active folding (Figure 6.3).
We acquired continuously cored boreholes that overlap with the shallow seismic
reflection data and provide a continuous image all the way to the earth’s surface (Figure
6.4).
6.2.2 Borehole Data
Our study site is located along Stanford Avenue in south-central Los Angeles,
~10 km south of the downtown high rise-district (Figures 6.1 and 6.2). Several features
combine to make this an excellent location to study the recent history of the Compton
155
Figure 6.2 Contour Map of the Stanford Avenue Study Site
Contour map of the location of the active, synclinal axial surface of the backlimb of
the Compton thrust and study site along Stanford Avenue, Los Angeles. Contour
intervals are 5 feet [USGS 7.5’ 1:24,000 Inglewood quadrangle topographic map
NGTCA33118H3, California]. A-A’, B-B’, C-C’ mark traces of the seismic
reflection profiles shown in Figure 6.3 (parts A, B, and C, respectively). A-A’ also
marks location of cross-section of borehole transect shown in Figure 6.4. Note that
C-C’ extends an additional 2.3 km southward off the figure. Also shown is the
location of Mueller [1997] study site located at the north-western end of the
northern segment of the Compton thrust on Van Ness Avenue in Inglewood (black
triangle). Note the prominent, north-facing scarp that passes through the Mueller
(1997) study site. This topographic scarp dies out east-southeastward into the active
Los Angeles River floodplain. Evidence for recent folding is completely buried at
our study site on Stanford Avenue, providing a continuous record of buried fold
scarps that formed during paleo-earthquakes on the Compton thrust ramp. N-IFZ
is the Newport-Inglewood fault zone.
156
Figure 6.2 Continued
157
thrust fault. Deep-penetration petroleum industry seismic reflection profiles provide an
exceptionally clear image of the folding above the Compton thrust ramp beneath the site.
In particular, these data show that the back-limb axial surface - the locus of folding
associated with fault slip across the base of the Compton ramp at depth - exhibits an
upward-narrowing growth triangle that extends to within ~200 m of the surface [Shaw
and Suppe, 1996]. This fault-bend fold, referred to as the Compton – Los Alamitos trend,
marks the base and northeastern extent of the Compton ramp. My study site is located
along this trend on a previously marshy, extremely low-gradient floodplain, facilitating
reconstruction of originally near-horizontal floodplain surfaces and allowing collection of
numerous organic-rich samples for radiocarbon dating. It is also far-removed from
structural interference associated with other active faults, such as the Newport-Inglewood
right-lateral strike-slip fault (Figure 6.2). Twentieth-century topographic maps of the site
show the ground surface as near horizontal, with an extremely gentle dip of 0.2
o
to the
south. However, presumably because of historical withdrawal of water and petroleum,
the ground surface now dips slightly (0.06
o
) to the north. The gentle, near-horizontal
regional slope at the study site has resulted in the deposition of laterally extensive
sedimentary layers that can be correlated over large distances.
As is important with all paleoseismic studies, I chose a site characterized by
continuous, relatively rapid sediment accumulation, leading to stratigraphic separation of
event horizons and preservation of event stratigraphy. I specifically chose a study site in
which the uplifted hanging wall exhibited no topographic expression in order to ensure
that our site was located in an area of continuous aggradation. The random occurrence
158
Figure 6.3 Northern Segment of the Compton Fault Seismic Reflection Data.
(A) Hammer-source seismic reflection profile (migrated; 8x vertical exaggeration).
The prominent reflector at 10- 17 m depth shows the kink band clearly. Thin,
vertical, white lines show borehole locations. Dashed white lines represent active,
synclinal and inactive, anticlinal axial surfaces. Black box indicates location of
borehole transect cross-section in figure 4. (B) Weight-drop source seismic
reflection profile (migrated). The prominent reflector at 120-200 m depth shows the
kink band well. The subhorizontal reflector at about 100 m depth is likely the water
table, as velocities above the reflector are consistent with unsaturated strata. The
reflector also cuts across the geologic structure consistently defined by the
underlying reflectors and overlying drill hole data, as would be expected for the
water table. Thin, vertical, white lines show borehole locations. Dashed white lines
represent active, synclinal and inactive, anticlinal axial surfaces. Black box
indicates location of hammer-source profile shown in part A. (C) Seismic reflection
image of the backlimb fold structure showing upward-narrowing zone of active
folding (growth triangle) delimited by sharply defined axial surfaces [Shaw and
Suppe, 1996]. Dashed white lines represent active, synclinal and inactive, anticlinal
axial surfaces. Back box indicates location of weight-drop source profile shown in
part B. (D) Fault-bend-fold model of backlimb fold structure of the Compton fault
with discrete axial surface yielding an upward-narrowing fold limb in growth strata
[Shaw and Suppe, 1996].
159
Figure 6.3 Continued
160
within any floodplain of topographic features would serve to decrease the consistency of
stratigraphic growth patterns. Moreover, historical evidence reveals that the floodplain
was so low gradient, that during rainy winters, the entire Los Angeles Basin (>20 km
wide) was a marsh or shallow lake with standing surface water [Gumprecht, 2001;
Linton, 2005].
I excavated a north-south transect of ten, 25- to 35-m-deep, continuously cored
boreholes across the zone of young folding imaged in the high-resolution seismic
reflection profiles. Excavation of boreholes allows me to document the details of the
most recent folding events, as well as to sample the sediments for age control. The cores
reveal sequences of fluvial sand and gravel, interbedded with cohesive overbank silt and
clay layers (Figure 6.4). Grain size was visually identified using a hand-held GSA grain-
size card, and I assigned soil colors on wet material using a Munsell color chart.
Deposition of these units in a low-gradient floodplain has resulted in remarkable
lateral continuity; more than 25 units are traceable continuously across the entire 760 m
length of the borehole transect. Calibrated radiocarbon dates from 30 bulk-soil, humic,
and charcoal samples indicate that sediment accumulation has been relatively continuous
at ~2 mm/yr over the past 14,000 years (Figure 6.5). The sediment accumulation reveals
no major hiatuses and the few soils observed in the section consist of weakly developed
A-C profiles that record brief periods of pedogenesis. I see exceptionally well-defined
growth in exactly the same location throughout the stratigraphic section. This argues
strongly that I am seeing a tectonic signal and that any natural variation of the floodplain
is not a significant source of error. I can also directly demonstrate through overlapping
161
Figure 6.4 Borehole Results from the Stanford Avenue Transect
(A) Cross-section of major stratigraphic units (16x vertical exaggeration) showing
the details of the most recent uplift event (Event 1). Thin vertical lines are
boreholes. Green horizontal line is the present ground surface. (B) Cross section of
major stratigraphic units (8x vertical exaggeration). Colors denote different
sedimentary units. Thin red lines mark the tops of major sand and gravel units.
Double-headed red vertical arrows along the right side of the figure show the
stratigraphic ranges of intervals of sedimentary thickening across the transect, with
the uplift in each event shown in red to the right of each arrow. Double-headed
green vertical arrows show intervals of no sedimentary growth.
162
Figure 6.4, Continued
163
radiocarbon dates that at least some of these units are synchronous along the length of the
transect. For example, three dates from the shallowest of several continuous marsh
deposits (Unit 11 – organic rich soil beneath Unit 10) yield overlapping radiocarbon dates
that confirm simultaneous development across the transect of this layer. I also observe
similar ages from multiple boreholes at 15.5 m and 29.5 m depths (relative to reference
borehole CF4). Minor mismatches in some layers are due to reworked older carbon (e.g.,
sample CF13).
6.2.3 Stratigraphic Evidence for Paleo-Folding Events
The borehole data reveal evidence for a discrete, upward-narrowing zone of
folding located at the updip projection of the synclinal axial surface seen on the industry
and high-resolution seismic reflection data (Figure 6.3). Specifically, the boreholes show
that all strata dip near-horizontally, parallel to the surface of the active floodplain, at the
northern and southern ends of the transect, but that the tops of all strata are deeper at the
northern end of the transect. This structural relief is produced by a panel of north-dipping
strata that lies directly at the up-dip projection of the growth triangle imaged
continuously on the seismic reflection data from >4 km depth.
Drilling the boreholes along the same transect as the high-resolution data
acquisition allowed me to directly compare the high-resolution seismic reflection and
borehole data (Figure 6.6). This comparison reveals generally good agreement between
the two data sets. Specifically, the high-resolution data accurately portray the location of
the active, synclinal axial surface, as well as the general form of the folding. Reflectors
as shallow as 6 m on the south side of the profile appear to correlate well with transitions
164
between thick sands and thick mud-rich units, which might be expected to produce the
significant acoustic impedance contrasts necessary to result in a discernible reflector.
The major reflector at ~20 m depth on the south side of the profile appears to correlate
with the top of the coarse-grained sand and gravel-dominated Pleistocene section
observed at the base of our boreholes, as might be expected for this major downward
change in sedimentary grain size. The continuity of folding illustrated by the combination
of seismic reflection and borehole data provides a direct link between the deformation
that we observe in the upper few tens of meters and the base of the Compton thrust ramp
at >7 km.
In situations such as I encountered at our study site, when deep seismic slip on the
thrust ramp is translated to the surface as fold growth leading to surface uplift, a fold
scarp develops at the updip tip of the active axial surface. If a laterally extensive fold
scarp develops at a high angle to a river, as at my study site, the river will immediately
begin to adjust its bed in order to return to the pre-earthquake, equilibrium stream
gradient. This may be achieved by either deposition of sediment on the downthrown
(upstream, in this case) side of the scarp, leading to growth of a thickened sedimentary
package until the pre-folded stream gradient is re-established, and/or erosion of the
upthrown (downstream, in this case) side of the scarp. This difference in sediment
accumulation rates is thus a direct manifestation of sedimentary growth across the fold.
In the Stanford Avenue transect, the northward stratigraphic thickening (or
“growth”) across this fold limb occurs in several discrete intervals that thicken
northward, reflecting the relative uplift of the hanging wall during folding and the
165
subsequent, post-uplift deposition that restores the original near-horizontal stream
gradient. These growth strata are interbedded with intervals that do not change thickness,
reflecting deposition during periods of structural quiescence at the near-horizontal
floodplain gradient. The stratigraphic relationships exposed in the boreholes demonstrate
that the youngest folding related to the most recent Compton thrust earthquakes is
localized within a narrow zone near the center of the transect between boreholes CF3 and
CF5, with more than half of the folding concentrated in a <30-m-wide zone between CF7
and CF6 (Figure 6.4).
In total, I observe six stratigraphically discrete, now-buried fold scarps. In each of
these, fold growth has uplifted the strata to the south of the zone of active folding,
resulting in a north-facing fold scarp that has been onlapped by the next-youngest
stratigraphic unit. I term these uplift Events 1 through 6 (from youngest to oldest [Table
6.1]).
The fold scarp that grew during the most recent Compton thrust earthquake is
particularly well defined (Figure 6.4A). The Unit 10 sand does not change thickness
across the transect, suggesting that it was deposited during a period of quiescence at the
equilibrium stream gradient. This unit was folded during Event 1, producing a 1-m-high,
north-facing fold scarp. This now-buried scarp was subsequently on-lapped by the Unit 8
sand. Later deposition occurred at the equilibrium stream gradient. Thus, the uplift event
that produced the fold scarp occurred after deposition of the Unit 10 sand, and before
deposition of the Unit 8 sand.
166
6.2.4 Structural Reconstructions
6.2.4.1 Methods
I reconstructed the structural evolution of these paleo-fold scarps using the
methodology of Novoa et al. [2000] (Figure 6.7). The heights of each buried fold scarp
determined from these reconstructions provide a robust measure of the minimum uplift
that occurred in each event [Benesh et al., 2007] and allow me to document the
incremental structural and stratigraphic evolution of the folded sediments. This process,
known as “inclined shear restoration”, is a general method of sequentially restoring
growth folds that does not imply a specific mechanism of fold growth. Rather, inclined
shear methods can resolve growth histories of folds that develop by kink-band migration
or limb-rotation mechanisms, or combinations of these end members [Shaw et al., 2002].
The critical points are that the restorations are sequential (providing path, not just initial
and final states), and that the balance of folding mechanisms is specified by the
geometries of the folded growth horizons, which generally distinguish between kink-band
migration and limb-rotation mechanisms. The various fault-related folding theories (e.g.,
fault-bend folding, trishear) all specify the kinematics of fold growth (i.e., if the folds
grow by kink-band migration, limb-rotation, or some specific combination of these).
Thus, inclined shear restorations, by determining folding mechanisms, are considered a
means of determining what fault-related folding theories might be most appropriate for
describing a structure. These reconstructions also allow incremental measurements of
sediment growth and the development of structural relief across the fold. In these
restorations, we used the 78
o
+1.5
o
S dip of the active, synclinal axial surface revealed by
167
seismic reflection data as the direction of shear [Shaw et al., 2002]. Uplift for each event
was then measured by averaging the depths of stratigraphic horizons in individual
boreholes for both the northern (CF4, CF8, CF2, CF9) and southern (CF1, CF10, CF3)
sections of the transect that are not folded (i.e., to the north and south of the north-
dipping dip panel). Uplift was then calculated as the difference between the average
depths of the northern and southern portions of the transect, with uncertainties estimated
as the difference between the average and maximum and/or minimum values in
individual boreholes. Additional uncertainties are attributed to loss of recovery within
individual drives (+15 cm in intervals of good recovery [>95%] and +30 cm in intervals
of poor recovery), and depth errors associated with drilling (+15 cm), which we were able
to quantify by measuring down hole with a tape measure every five feet. These potential
errors, however, are small compared to the +3
o
uncertainty of fault dip when converting
uplift to slip. In the following discussion all uncertainties (average uplift, loss of
recovery, driller’s depth errors, and fault dip uncertainty) are calculated and included
when reporting slip on the underlying thrust ramp (Table 6.1).
6.2.4.2 Shear Restoration of Folding Events
The shear restorations of the Stanford Avenue borehole data (Figure 6.7) demonstrate the
occurrence of six stratigraphically discrete uplift events during the past ~14,000 years, each of
which (with the possible exception of Event 6) were followed by sedimentary growth and
subsequent restoration of the regional, near-horizontal stream gradient.
I base my reconstructions primarily on the tops of major sand units, as these
horizons would have been deposited at (and provide a record of) the paleostream gradient
at the site. The scoured base of sand units is not an appropriate indicator of paleo-stream
168
Figure 6.5 Sediment Accumulation-Rate Curve for the Stanford Avenue Transect
Sediment accumulation-rate curve for the Stanford Avenue transect, showing
calibrated, calendric radiocarbon dates sorted by depth. Radiocarbon dates were
calibrated using the Oxford Radiocarbon Accelerator Unit (ORAU) calibration
program OXCAL v3.10 [Bronk Ramsey, 1995; 2001; using atmospheric data from
Reimer et al., 2004; Bronk Ramsey and van der Plicht, 2001]. True depth has been
corrected for folding by projecting all samples to a common reference point
(Borehole CF4), moving each sample along the nearest stratigraphic horizons to its
projected depth in borehole CF4. Probability distributions are black for bulk-soil
samples, red for charcoal samples, and purple for humic samples. Numbers next to
probability functions represent sample numbers in Table 6.1. Sediment
accumulation-rate curve is grey. Probability density functions of bulk-soil samples
that appear to be anomalously young are circled. Uncertainties in the ages of events
(Event 1 through Event 6) are marked by horizontal black lines, whereas depth
uncertainties are marked by vertical black lines. BH# indicates borehole location
for each sample. Sample type, depth and calendric age shown in Table 6.1.
169
Figure 6.5, Continued
170
gradient due to the erosional nature of these contacts, and therefore these basal contacts
cannot be used as reliable restoration horizons. These sands represent the highest-energy
deposits in the Holocene part of the section, and in almost all cases, the sands are overlain
by very fine-grained (clay/silt) deposits that record low-energy, slack-water deposition. I
therefore consider it highly unlikely that the tops of the sands have been modified by
erosion. Thus, although all uplift measurements discussed below must be considered
minima due to the possibility of erosion of the scarp, as noted above, we consider this
possibility unlikely. Conversely, any compaction of fine-grained sediments on the
thickened upstream, downthrown side of the scarp will make our uplift measurements
maxima. At these shallow depths, however, such differential compaction is likely to be
minor. In my reconstructions we use 20
th
Century topographic maps rather than the
current ground surface, which appears to have been tilted during the past few decades,
presumably due to ground water withdrawal.
For Events 1, 3, 4, 5, and 6, we use the tops of major sand units as restoration
horizons because the intervals overlying these stratigraphic units do not change thickness
along the transect, indicating deposition of at least the youngest part of each sand unit at
the equilibrium stream gradient. For Event 2, I use as the restoration horizon the top of a
thick, cohesive, organic-rich clay (Unit 15), which does not change thickness across the
transect. The intervals of sedimentary growth following the uplift events are separated
from one another by stratigraphic intervals that do not change thickness across the
transect (except between Events 5 and 6 [Figure 6.4; Figure 6.7-F1]). I interpret these
171
Table 6.1 Calibrated, Calendric Dates and Radiocarbon Ages of Samples from
Stanford Avenue Transect
All samples have been calibrated using Oxford Radiocarbon Accelerator Unit
(ORAU) calibration program OXCAL v3.10 [Bronk Ramsey, 1995; 2001; using
atmospheric data from Reimer et al., 2004; Bronk Ramsey and van der Plicht, 2001].
For sample type; ccl=charcoal fragment; bulk=bulk soil; bulk-h=bulk soil humic.
Calendric age is reported as 2 sigma (95% confidence limit) age range.
172
Table 6.1, Continued
PDF
#
Sample # Bore
hole
Sample
type
True
depth
(m)
Projected
depth to
bh 4 (m)
14
C Age BP
13
C Calendric age %
pro
ba
bilit
y
1 CF63 6 ccl 27.44 30.67 11650 + 35 -25 11690 – 11430
BC
95.4
2 CF65-H 6 bulk - h 27.31 30.54 12080 + 35 -25 12090 – 11860
BC
95.4
3 CF65-BS 6 bulk 27.31 30.54 11880 + 35 -28 11890 – 11700
BC
95.4
4 CF31 3 ccl 21.75 30.18 11995 + 35 -25 12020 – 11810
BC
95.4
5 CF62 6 ccl 26.22 29.45 11985 + 40 -25 12020 – 11790
BC
95.4
6 CF32-H 3 bulk - h 20.81 29.24 12240 + 70 -25 12550 – 11900
BC
95.4
7 CF32-BS 3 bulk 20.81 29.24 10085 + 50 -28 10050 – 9400 BC 95.4
8 CF55-BS 5 bulk 25.25 26.24 7800 + 60 -25 6820 – 6470 BC 95.4
9 CF46-BS 4 bulk 26.78 26.78 10290 + 35 -25 10450 – 10350
BC & 10300 –
10000 BC &
9950 – 9850 BC
2.2
90.9
2.4
10 CF61-BS 6 bulk 19.31 22.33 9205 + 35 -25 8550 – 8500 BC
& 8490 – 8300
BC
10.0
85.4
11 CF61-H 6 bulk - h 19.31 22.33 8870 + 270 -25 8550 – 7500 BC 95.4
12 CF59 5 ccl 20.48 20.48 12215 + 40 -25 12250 – 12000
BC
95.4
13 CF11 1 ccl 15.04 20.39 10105 + 40 -25 10050 – 9450 BC 95.4
14 CF44-BS 4 bulk 21.09 21.09 8480 + 80 -25 7650 – 7330 BC 95.4
15 CF43 4 ccl 15.45 15.45 7475 + 40 -25 6430 – 6240 BC 95.4
16 CF81-BS 8 bulk 15.32 15.19 8125 + 40 -25 7300 – 7220 BC
& 7190 – 7040
BC
6.2
89.2
17 CF33 3 ccl 11.38 15.31 7940 + 50 -25 7040 – 6680 BC 95.4
18 CF42-BS 4 bulk 12.07 12.07 6135 + 40 -25 5220 – 4960 BC 95.4
19 CF58 5 ccl 11.33 11.33 7590 + 35 -25 6495 – 6395 BC 95.4
20 CF53-BS 5 bulk 8.16 8.66 4740 + 50 -25 3640 – 3490 BC
& 3460 – 3370
BC
66.0
29.4
21 CF57 5 ccl 6.02 6.48 5925 + 45 -25 4940 – 4700 BC 95.4
22 CF56-BS 5 bulk 5.84 6.10 4890 + 35 -25 3770 – 3630 BC 95.4
23 CF64-BS 6 bulk 3.96 5.97 1140 + 35 -25 780 – 990 AD 95.4
24 CF52-BS 5 bulk 5.01 5.23 2965 + 45 -25 1380 – 1340 BC
& 1320 – 1040
BC
2.8
92.6
25 CF47 4 bulk 3.04 3.04 1845 + 35 -25 70 – 250 AD 95.4
26 CF71-H 7 bulk - h 1.98 2.92 2010 + 60 -25 180 BC – 130 AD 95.4
27 CF51-BS 5 bulk 2.87 2.92 1955 + 35 -25 40 BC – 130 AD 95.4
28 CF71-BS 7 bulk 1.98 2.92 1650 + 35 -28 260 – 290 AD &
320 – 540 AD
3.9
91.5
29 CF41-BS 4 bulk 1.52 1.52 2345 + 35 -25 540 – 360 BC 95.4
30 CF21-BS 2 bulk 3.0 0.91 890 + 90 -25 990 – 1280 AD 95.4
173
no-growth intervals as recording periods of structural quiescence characterized by
deposition at the regional, near-horizontal stream gradient.
In order to determine uplift in each folding event, I calculate the total minimum
scarp height, while taking into account all potential errors associated with reduced
recovery (Figure 6.7-A2-F2 and B3-E3). Uplift measured in Event 1, the
most recent event, is 1.1 m + 0.2. As described above, the cohesive unit above Unit 10
and the overlying Unit 8 sand thicken northward, recording sedimentary growth that
onlaps a now-buried, north-facing fold scarp that formed atop the Unit 10 sand. Thus,
Event 1 occurred sometime after the deposition of Unit 10 and before the completion of
deposition of Unit 8 (Figure 6.7-A1-A2). The maximum calendric age of a bulk-soil
sample from Unit 11 (an organic-rich soil at the base of Unit 10) of 1,750 calibrated years
BP provides a maximum age for Event 1, whereas the minimum age of a bulk-soil sample
from the cohesive unit directly above Unit 8 provides a minimum age of 730 cal yr BP.
Thus, Event 1 occurred between 730 and 1,750 years ago (Table 6.1).
The minimum measured uplift in Event 2 is 1.7 m + 0.5. The interval of
sedimentary growth within the laterally continuous units above Unit 15 and below Unit
10 indicates that these units were deposited on a north-facing paleo-scarp that formed
during Event 2, after the deposition of Unit 15 and before the completion of deposition of
Unit 11 (Figure 6.7-B1-B3). The maximum age of a bulk-soil sample from Unit 15
provides a maximum-possible age for this event at 3,400 cal yr BP, whereas the
minimum age of a bulk-soil sample from the cohesive unit directly above Unit 8 provides
a minimum age of 730 cal yr BP. Alternatively, the minimum age of a bulk-soil sample
174
Figure 6.6 High-Resolution Seismic Reflection Profile and Borehole Results
Cross-section of major stratigraphic units from the Stanford Avenue borehole
transect (Figure 6.4B) overlain on the high-resolution seismic reflection hammer
profile (Figure 6.3A) (8x vertical exaggeration). The strong reflector at 20- to 30- m-
depths probably represents the contact between Unit 60, a coarse-grained sand and
gravel, and the overlying cohesive silt and clay unit. Boreholes are represented by
thin, vertical, yellow lines.
175
Figure 6.6, Continued
176
from the top of Unit 11 provides a more-likely minimum age of 1,900 cal yr BP. Thus,
Event 2 occurred between 3,400-730 cal yr BP, and most likely between 3,400-1,900 cal
yr BP. In contrast, the lateral continuity and lack of growth within the Unit 10 sand
indicate that this unit was deposited at the near-horizontal paleo-stream gradient during a
period of structural quiescence following Event 2 (Figure 6.7-B2).
The minimum measured uplift for Event 3 is 1.2 m + 0.3. Uplift for this event is
recorded by the sedimentary growth that occurs below Unit 15 and above Unit 20 (Figure
6.7-C1-C3). The maximum age of a bulk-soil sample from Unit 21 (an organic-rich soil
at the base of Unit 20) of 7,200 cal yr BP provides a maximum age for Event 3, whereas
the minimum age of a bulk-soil sample from Unit 15 provides a minimum age of 5,600
cal yr BP. Thus, Event 3 occurred between 7,200-5,600 cal yr BP. The lack of growth
within Unit 15, a thick, laterally continuous, organic-rich soil, suggests that a period of
structural quiescence occurred between uplift Events 2 and 3 (Figure 6.7-C2).
The cohesive unit overlying the Unit 30 sand and below Unit 20 records another
period of deposition and deformation. Northward thickening of this stratigraphic interval
across the zone of active folding indicates a minimum uplift of 1.3 m + 0.3 for Event 4
(Figure 6.7-D1-D3). This interval of sedimentary growth records onlap of a north-facing
fold scarp that developed sometime after the deposition of Unit 30, and before the
completion of deposition of Unit 21 (an organic-rich soil at the base of Unit 20) and
subsequent deposition of Unit 20. The maximum age of a charcoal sample from Unit 30
provides a maximum-possible age for this event at 8,400 cal yr BP, whereas the
minimum age of a bulk-soil sample from Unit 20 provides a minimum age of 5,400 cal yr
177
BP. Thus, Event 4 occurred between 8,400-5,400 cal yr BP. The absence of sedimentary
growth in Unit 20 supports our inference that the top of this sand was deposited at a near-
horizontal paleostream gradient (Figure 6.7-D2). In contrast, the underlying interval
comprising the Unit 30 sand, cohesive overbank muds overlying Unit 40, and the Unit 40
sand exhibits no change in thickness along the length of the transect, indicating that these
sediments were deposited near-horizontally. Units 30 through 40 thus appear to record a
period of structural quiescence, with no discernible uplift (Figure 6.7-E2).
In contrast to the events described above, measuring uplift in Events 5 and 6 is
somewhat problematic. This ambiguity is caused by the fact that, unlike the situations
described above, there is no unit of continuous thickness separating the intervals of
sedimentary growth that occurred following Events 5 and 6, with the exception of the
Unit 42 sand at the top of the problematic stratigraphic interval. My basic observation is
that the total minimum uplift in Events 5 and 6 is 2.5 m. Although it is possible, given the
absence of evidence for prolonged structural quiescence within the Event 5-6 section, that
all of this uplift occurred during one exceptionally large earthquake, stratigraphic
relationships suggest that this uplift occurred during at least two separate events (Figure
6.7-E1-F1).
My reasoning that Event 5 was a separate earthquake comes mainly from the
coarse grain size of the Unit 50 sand. Specifically, although the cohesive silts and clays
overlying the Unit 50 sand might partially drape existing topography, the medium- to
coarse-grained sands and gravels of Unit 50 will buttress or onlap the scarp. I therefore
assume that the top of Unit 50 was deposited at the local stream gradient. Thus, for
178
Figure 6.7 Sequential Reconstructions for the Six Observed Uplift Events
For each event, major stratigraphic horizons have been incrementally restored to
the paleo-stream gradient (based on modern topographic maps) in the direction of
shear [Novoa et al., 2000]. Boreholes are shown as thin, black vertical lines. Thin
red lines mark the tops of major sand- and gravel-filled channels. Thin red, dashed
lines represent tops of major units prior to reconstructions. Graphs next to each
reconstruction step show plots of increasing sedimentary thickness from south to
north (blue line) versus increasing depth of top of major stratigraphic horizons
(pink line) for each event (A2, B3, C3, D3, E3, and F2). Additional graphs illustrate
periods of no growth by plotting sedimentary thickness versus depth for
stratigraphic units between growth intervals (B2, C2, D2, and E2).
179
Figure 6.7 Continued
180
Figure 6.7, Continued
181
Figure 6.7, Continued
182
Event 5, sedimentary growth occurred within the cohesive units overlying Unit 50, and
below Unit 42, indicating that this section was deposited against a now-buried fold scarp
that formed sometime after deposition of Unit 50 and before deposition of Unit 42
(Figure 6.7-E1-E3). If Event 5 is a separate earthquake, as we suspect, the minimum
measured uplift is 0.6 m + 0.2. The maximum age of a bulk-soil sample from Unit 50 of
12,500 cal yr BP provides a maximum age for Event 5, whereas the 10,300 cal yr BP
minimum age of a bulk-soil sample from unit 43 (an organic-rich soil at the base of Unit
42, which as a no-growth interval records deposition at the equilibrium stream gradient)
provides a minimum age for this earthquake. Thus, Event 5 occurred between 12,500-
10,300 cal yr BP.
During Event 6, a north-facing fold-scarp developed sometime after deposition of
the very coarse-grained Unit 60 sand and before the end of deposition of the Unit 50
sand. Sedimentary growth occurred from the top of Unit 60 to the top of Unit 50 (Figure
6.7-F1-F2).
Minimum measured uplift in this interval is 1.9 m + 0.2. The maximum age of a
charcoal sample from Unit 60 of 13,700 cal yr BP provides a maximum age for Event 6,
whereas the minimum age of a bulk-soil sample from an organic-rich soil at the base of
Unit 42 (Unit 43) provides a minimum age of 10,300 cal yr BP. Thus, Event 6 occurred
between 13,700-10,300 cal yr BP.
183
6.3 Discussion
6.3.1 Fault Slip Rate Estimates
To determine the reverse slip rate of the fault, I combine our measured minimum
uplifts with the dip of the causative Compton thrust ramp. The fault dip was defined by
the dip of the fold limb (28
o
) imaged in industry seismic reflection profiles [Suppe et al.,
1992; Shaw and Suppe, 1996], and corrected for apparent dip. Uncertainties in the
velocity structure that was used to depth process the seismic reflection data [Süss and
Shaw, 2003] leads me to estimate conservative fault-plane dip errors for the Compton
ramp at +3
o
. Slip estimates for individual events discussed below include fault-plane dip
uncertainties. It is possible that the dip of the Compton ramp may be greater than the dip
of the overlying fold limb, which would imply that the slip and slip rates required to
generate the observed uplift would be less than we calculate. However, the interaction of
the Compton ramp with adjacent structures strongly favors a dip of the ramp that is
similar to that of the overlying fold limb. Thus, I favor this interpretation in my analysis.
For further discussion on the dip of the Compton ramp, see Shaw and Suppe [1996]. I
divide the uplift measurement from the top of Unit 60 by the sine of the 28°±3° northeast
dip of the Compton ramp and derive a cumulative thrust displacement over the past
14,000 years of 16.9 +7.5/-6.9 m. This yields a minimum slip rate of 1.2 +0.5/-0.3
mm/yr, similar to the long-term slip rate of 1.4±0.4 mm/yr [Shaw and Suppe, 1996]. This
rate is a minimum because: (1) my uplift measurements fail to account for any potential
erosion of the hanging wall; (2) it assumes that all slip on the deep Compton ramp is
manifested as folding within the kink band. As noted above, however, I think that
184
Table 6.2 Uplift Amounts, Age Limits, and Estimated Paleo-magnitudes for the Northern Segment of the
Compton Blind-Thrust Fault
Event Age (ka) Growth
section
Uplift
(m)
Slip (m) M
w
(All-
slip-type
avg.
displace-
ment)
M
w
(All-
slip-type
max.
displace-
ment)
M
w
(Thrust-
fault-only
avg.
displace-
ment)
M
w
(Thrust-
fault-only
max.
displace-
ment)
1 0.7 – 1.75
ka
Top of Unit 8
to top of Unit
10
1.1 + 0.2 2.4 + 0.8/-
0.6
7.2 + 0.2 7.0 + 0.3 6.7 -0.2/+0.5 6.7 -0.2/+0.4
2 0.7 – 3.4
ka or 1.9 –
3.4 ka
Top of Unit 11
to top of Unit
15
1.7 + 0.5 3.7 +1.5/-
1.2
7.4 + 0.2 7.1 + 0.3 6.7 -0.2/+0.5 6.8 -0.2/+0.4
3 5.6 – 7.2
ka
Top of Unit 16
to top of Unit
20
1.2 + 0.3 2.5 +1.1/-
0.9
7.3 + 0.2 7.0 + 0.3 6.7 -0.2/+0.5 6.7 -0.2/+0.4
4 5.4 – 8.4
ka
Top of Unit 21
to top of Unit
30
1.3 + 0.3 2.7 +0.9/-
0.8
7.3 + 0.2 7.0 + 0.3 6.7 -0.2/+0.5 6.7 -0.2/+0.4
5 10.3 –
12.5 ka
Top of Unit 42
to top of Unit
50
0.6 + 0.2 1.3 +0.6/-
0.5
7.0 + 0.2 6.8 + 0.2 6.7 -0.2/+0.5 6.6 -0.2/+0.4
6 10.3 –
13.7 ka
Top of Unit 50
to top of Unit
60
1.9 + 0.2 4.2 +1.0/-
0.8
7.4 + 0.2 7.1 + 0.3 6.7 -0.2/+0.5 6.8 -0.2/+0.4
Uplift, age limits, and estimated moment-magnitude (M
w
) for Compton fault paleo-earthquakes 1 through 6
from the Stanford Avenue borehole results. Results based on the assumption that our measured displacements
represented either the average or the maximum displacement in each earthquake [Wells and Coppersmith,
1994].
185
erosion of the hanging wall uplift in each of the folding events is likely to be minimal
because of the low energy deposition recorded by the fine-grained deposits overlying
most of the event horizons. Moreover, my analysis of industry seismic reflection data
shows that at least 90% of the total structural relief due to folding of Quaternary horizons
above the Compton thrust fault occurs within the growth triangle. Thus, although I report
this rate as a minimum, we think that it is probably close to the actual slip rate of the
Compton thrust beneath the Stanford Avenue site. Using a large number of high-quality,
deep-penetration industry seismic reflection profiles, Lehle (2007) measured
displacements at various time intervals, ranging from 0.45 Ma to 2.4 Ma, along the length
of the Compton thrust. Her data demonstrate that displacement at our Stanford Avenue
study site occurs at a approximately 60% of the maximum long-term slip-rate along the
Compton thrust, which occurs along the central part of the fault approximately 15 km
south of our site. If the long-term displacement patterns documented by Lehle (2007)
extend to the present, these observations suggest that the maximum slip-rate along the
central part of the Compton thrust fault may be on the order of ~2.0 +0.8/-0.5 mm/yr.
6.3.2 Paleo-magnitude Estimates and Implications for Fault Behavior
The observation that uplift occurs in stratigraphically discrete intervals rules out
the possibility that this fold grew steadily, either through many smaller-magnitude
earthquakes or by quasi-continuous fault creep, on the Compton ramp. I cannot,
however, rule out the possibility that some component of folding occurs during
punctuated periods of aseismic fault slip, either as afterslip following large earthquakes
or during discrete periods of interseismic fold growth. It is also possible that each
186
stratigraphically discrete uplift event records a brief cluster of multiple, moderate-
magnitude earthquakes. If each uplift event does record multiple, smaller-magnitude
earthquakes, however then these must have occurred during relatively brief clusters (e.g.,
on the order of ≤800 years for Event 1). Moreover, uplift and fold growth are likely to
occur only during large-magnitude events that involve displacement of the hanging wall
through and across changes in the fault dip (e.g., from the flat down-dip from the
Compton ramp through the backlimb active, axial surface). Small- to moderate-
magnitude ruptures that are limited to the thrust ramp are unlikely to localize deformation
at the active, axial surfaces and will therefore not contribute substantially to the fold
growth that we measure at the surface (e.g. the 1987 M
w
6.0 Whittier Narrows
earthquake; Lin and Stein, [1989]). Thus, almost all growth of the Compton-Los
Alamitos fold probably occurs in large-magnitude events.
I estimate a conservative range of magnitudes for these paleo-earthquakes by
comparing our measured minimum displacements to global regressions of slip vs.
magnitude [Wells and Coppersmith, 1994], recognizing that the displacements measured
in this study represent slip at only one point along the Compton fault in earthquakes that
likely had laterally variable slip. If I make the simplifying assumption that my measured
displacements represent average displacement in each paleo-earthquake, the magnitude
estimates for the Compton thrust earthquakes range from M
w
7.0 + 0.2 to M
w
7.4 + 0.2
(Table 6.1). Alternatively, if I assume the extreme limiting case that all of my measured
displacements record the maximum displacements in each earthquake, the minimum
magnitude estimates range from M
w
6.8 + 0.2 to M
w
7.1 + 0.3. It is, of course, possible
187
that slip in any of the individual earthquakes we measured was the minimum slip in that
earthquake, but use of our displacements as minimum slip in the Wells and Coppersmith
[1994] regressions yields unreasonably large paleo-M
w
estimates, so we do not include
them here. If I use the thrust-fault-only regressions of Wells and Coppersmith [1994],
rather than their all-slip-type regressions, and each of my displacement measurements
represents an average displacement, the magnitude estimates are M
w
6.7 -0.2/+0.5. If my
measurements record maximum displacement, then the thrust-fault-only magnitude
estimates range from M
w
6.6 -0.2/+0.4 to M
w
6.8 -0.2/+0.4. The thrust regressions,
however, have many fewer data points than the regressions based on all-slip-type data,
and the uncertainty in the mean is therefore larger for thrust-slip-only regression [Wells
and Coppersmith 1994]. As a point of comparison, the 1994 M
w
6.7 Northridge
earthquake had a rupture area of ~300 km
2
(Wald et al., 1996), relative to the ~550 km
2
are of the Compton thrust ramp (Shaw and Suppe, 1996). Although, this leads us to
suspect that the Wells and Coppersmith (1994) thrust-only displacement regressions
underestimate the magnitude of the Compton fault paleo-earthquakes that I document, I
have nevertheless included them here for the sake of completeness. In summary, the
uncertainties associated with Wells & Coppersmith [1994] regressions convert into
uncertainties on the order of 0.1 to 0.3 M
w
in magnitude estimates for all-slip-type
regressions, with larger uncertainties of 0.1 to 0.5 M
w
in magnitude estimates for thrust-
only regressions due to the fewer number of available data for thrust-fault events.
Based on regressions of rupture area to magnitude (rather than slip), rupture of the
entire Compton ramp could generate an earthquake of ~M
w
7.1 [Shaw and Suppe, 1996],
188
in line with the smaller end of the magnitude estimates based on our measured
displacements. I consider it highly unlikely, however, that all of the minimum
displacements that we measured represent the largest displacements in all of these paleo-
earthquakes, and many of our M
w
estimates that assume average slip are larger than M
w
7.1. Thus, these events may have involved slip on faults other than just the Compton
ramp. Perhaps the most obvious explanation is that the decollement at the base of the
Compton ramp [Shaw and Suppe, 1996] also slipped in these earthquakes. Alternatively,
the Compton ramp may be underlain by a southwest-dipping fault, forming a structural
wedge, that also ruptured in these events. From a kinematic standpoint, these scenarios
seem plausible, as motion extending across this decollement-to-ramp transition or at the
wedge tip causes the large-scale folding that we observe in the seismic reflection and
borehole data [Shaw and Suppe, 1996].
6.3.3 Incremental Development of Dips within the Kink Band
The combined borehole and seismic reflection data described above indicate that
strata folded within the kink band acquired their dips incrementally. The progressive
downward increase in bed dip genuinely reflects an aspect of fold kinematics that is
distinct from classic growth fault-bend fold theories. This change in limb dips may
reflect some component of fold growth by limb rotation in addition to kink-band
migration. These hybrid kinematics are manifested in several types of fault-related folds,
including trishear fault-propagation folding [Erslev, 1991; Allmendinger, 1998] and shear
fault-bend folding [Suppe and Connors, 2004; Shaw et al., 2005]. Alternatively, the
component of limb rotation that we observe may reflect curvature of the fold hinge, as is
189
described by curved-hinge fault-bend folding theories [Suppe et al., 1997; Novoa et al.,
2000]. Finally, the progressive change in bed dips may reflect the mechanical response
of loosely consolidated, granular sediments in the shallow subsurface to folding at depth.
Recent work by Benesh et al., [2007] demonstrates that, even in cases of folding of pre-
tectonic layers exclusively by kink-band migration, shallow growth sediments acquire
their dips progressively. This is due to the finite width of the axial surface zone and its
rotation and migration with increasing deformation. In this particular example, the
structural restorations indicate that the axial surface trace has remained stationary from
event to event, consistent with theories of fault-bend-fold formation primarily by kink-
band migration [Suppe et al., 1992; Shaw and Suppe, 1994; Benesh et al., 2007].
6.3.4 Implications for Seismic Hazard in Southern California
Although earlier studies have effectively demonstrated the hazard associated with
the Puente Hills thrust [Shaw and Shearer, 1999; Pratt et al., 2002; Shaw et al., 2002;
Dolan et al., 2003; Field et al., 2005], the results of previous studies of the Compton
thrust [Mueller, 1997] have taken a very different path through both the regulatory
process and the public consciousness. Two previous studies involving paleoseismologic
trenches and closely spaced geotechnical penetration tests acquired across the fold trace
above the Compton thrust yielded indeterminate results [Mueller, 1997] leading to the
deletion of the Compton fault from the California Geological Survey (CGS) 2002 active
seismic sources. In the first of these two earlier studies, the researchers interpreted the
presence of flat-lying strata delineated in cone-penetrometer tests and shallow
paleoseismologic trenches excavated across the expected location of near-surface folding
190
in Los Alamitos as evidence for structural quiescence [Mueller, 1997]. Their study site,
however, was located near the southernmost extent of the southern segment of the
Compton fault and its overlying fold, where slip may be transferred to the southwest on
an en echelon structure (Figure 6.1). Industry seismic reflection data to the south of this
excavation site indicate that Quaternary strata in the uppermost 500 meters dip only a few
degrees to the northeast, in contrast to the ≥ 25° dips found elsewhere along the trend.
Given my observation that the dips of strata decrease significantly upward in the
Holocene section above the growth triangle, this suggests in hindsight that no discernable
trace of folding would be apparent at the Mueller [1997] excavation site, even if the
thrust ramp beneath the site was seismically active. A second study [Mueller, 1997],
using closely spaced geotechnical penetration tests across a suspected fold scarp at the
up-dip projection of the locus of recent folding observed on seismic reflection profiles
(Figures 6.1 and 6.2), indentified evidence of late Quaternary folding. The researchers
concluded, however, that shallow folding along the trend did not reproduce the growth
structures imaged at depth, and thus did not confirm the Compton thrust as an active
fault. The Compton thrust is currently included in the Southern California Earthquake
Center (SCEC) community fault model (CFM) [Plesch and Shaw, 2007] as a potentially
active fault due largely to my study.
The prospect for future large earthquakes on the Compton thrust similar to past
events documented in this study poses a significant seismic hazard to the Los Angeles
metropolitan region. As a point of comparison, the 1994 Northridge (M
w
6.7)
earthquake, which ruptured a similar type of fault but was much smaller than the events
191
documented here on the Compton ramp, caused widespread damage and remains one of
the costliest natural disasters in U.S. history [Scientists of USGS and SCEC, 1994].
Moreover, blind thrust earthquakes of the type forecast on the Compton thrust that occur
directly beneath metropolitan regions have been shown to produce very large ground
motions that pose particular concerns to high-rise buildings [Heaton et al., 1995; Field et
al., 2005; Sommerville and Pitarka, 2006; Abrahamson et al., 2008]. Thus, it is critical
that the activity and seismogenic potential of the Compton thrust be properly considered
in regional seismic hazards assessment. Finally, faults similar to the Compton blind
thrust lie beneath many cities around the world (e.g., Seattle, Washington [e.g., Liberty
and Pratt, 2008], Tokyo and Osaka, Japan [e.g., Sugiyama et al., 2003]). The multi-
disciplinary methodology described in this paper provides a means of defining the recent
seismic behavior, and therefore the hazard, posed by such structures to an increasingly
urbanized global population.
192
Chapter 6 References
Abrahamson, N., Atkinson, G., Boore, D., Bozorgnia, Y., Campbell, K., Chiou, B., Idriss,
I. M., Silva, W., and Youngs, R. (2008), Comparisons of the NGA Ground-
Motion Relations, Earthquake Spectra, 34, 1, 45-66.
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Field, E. H., H. A. Seligson, N. Gupta, V. Gupta, T. H. Jordan, and K. W. Campbell
(2005), Loss estimates for a Puente Hills blind-thrust earthquake in Los Angeles,
California, Earthquake Spectra, 21, 329-338.
Field., E. H. (2005), Collaborative SCEC/USGS efforts to improve seismic-hazard analysis;
RELM and OpenSHA: Proceedings of the 5th U.S.-Japan natural resources meeting and
Parkfield, California fieldtrip, Open-File Report - U. S. Geological Survey, Report: OF
2005-1131, pp. 18. Resource Location: http://pubs.usgs.gov/of/2005/1131/
193
Gumprecht, B. (2001), The Los Angeles River: It’s life, death, and possible rebirth,
Baltimore and London: John Hopkins University Press.
Heaton, T. H., J. F. Hall, D. J. Wald, and M. W. Halling (1995), Response of high-rise
and base-isolated buildings to a hypothetical Mw 7.0 blind thrust earthquake,
Science, 267, 206-211.
Ishiyama, T., K. Mueller, M. Togo, A. Okada, and K. Takemura (2004), Geomorphology,
kinematic history, and earthquake behavior of the active Kuwana wedge thrust
anticline, central Japan, J. Geophys. Res., 109, B12408, doi:
10.1029/2003JB002547.
Ishiyama, T., K. Mueller, H. Sato, and M. Togo (2007), Coseismic fault-related fold
model, growth structure, and the historic multisegment blind thrust earthquake on
the basement-involved Yoro thrust, central Japan, J. Geophys. Res., 112, B03S07,
doi: 10.1029/2006JB004377.
King, G. C. P., and C. Vita-Finzi (1981), Active folding in the Algerian earthquake of 10
October 1980, Nature, 292, 22-26.
Lai, K., Y. Chen, J. Hung, J. Suppe, L. Yue, and Y. Chen (2006), Surface deformation
related to kink-folding above an active fault: Evidence from geomorphic features
and co-seismic slips, Quat. Int., 147, 44-54.
Lehle, D. (2007), Geometry and slip history of the Compton thrust fault: Implications for
earthquake hazard assessment, thesis, Harvard Univ., Cambridge, Mass.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Liberty, L. M., and T. L. Pratt (2008), Structure of the eastern Seattle fault zone,
Washington state: New insights from seismic reflection data, Bull. Seismol. Soc.
Am., 98, 4, 1681-1695, doi: 10.1785/0120070145.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Linton, J. (2005), Down by the Los Angeles River, Wilderness Press, Berkeley,
California, 5-7.
194
Mueller, K. J., (1997), Recency of folding along the Compton-Los Alamitos trend:
Implications for seismic risk in the Los Angeles basin, EOS Trans. AGU, 78,
F702.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Plesch, A., and J. H. Shaw (2007), Community fault model (CFM) for Southern
California, Bull. Seismol. Soc. Am., 97, 6. 1793-1802.
Pratt, T. L., J. H. Shaw, J. F. Dolan, S. A. Christofferson, R. A. Williams, J. K. Odum,
and A. Plesch (2002), Shallow folding imaged above the Puente Hills blind-thrust
fault, Los Angeles, California, Geophys. Res. Lett., 29, 18-1 - 18-4.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Scientists of the U. S. Geological Survey and the Southern California Earthquake Center
(1994), The magnitude 6.7 Northridge, California, earthquake of 17 January 1994,
Science, 266, 389-397.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Sommerville, P. G., and A. Pitarka (2006), Differences in earthquake source and ground
motion characteristics between surface and buried earthquakes, Proceedings of
the Eighth National Conference on Earthquake Engineering, San Francisco,
California, U.S.A., 977, Earthquake Engineering Research Institute.
195
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
Stein, R. S., and R. S. Yeats (1989), Hidden earthquakes, Sci. Am., 260, 48-57.
Streig, A. R., C. M. Rubin, W. Chen, Y. Chen, L. Lee, S. C. Thompson, C. Madden, and
S. Lu (2007), Evidence for prehistoric coseismic folding along the Tsaotun
segment of the Chelungpu fault near Nan-Tou, Taiwan, J. Geophys. Res., 112,
B03S06, doi: 10.1029/2006JB004403.
Sugiyama, Y., K. Mizuno, F. Nanayama, T. Sugai, H. Yokota, T. Hosoya, K. Miura, K.
Takemura, and N. Kitada (2003), Study of blind thrust faults underlying Tokyo
and Osaka urban areas using a combination of high-resolution seismic reflection
profiling and continuous coring, Annals of Geophys., 46, 5.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
Süss, M. P., and J. H. Shaw (2003), P wave seismic velocity structure derived from sonic
logs and industry reflection data in the Los Angeles basin, California, J. Geophys.
Res., 108(B3), 2170, doi:10.1029/2001JB001628.
Wald, D. J., T. H. Heaton, and K. W. Hudnut (1996), The slip history of the 1994
Northridge, California, earthquake determined from strong-motion, teleseismic,
GPS, and leveling data Bull. Seismo.l Soc. Am., 86, 1B, S49-S70.
Wells, D. L., and K. J. Coppersmith (1994), New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface displacement,
Bull. Seismol. Soc. Am., 84, 974-1002.
196
CHAPTER 7:
Evidence for Holocene and Late Pleistocene Activity on the Compton Fault,
Lakewood, California: Implications for System-Wide Earthquakes and
Seismic Hazard
7.1 Introduction
The recent occurrence of several highly destructive earthquakes on thrust faults
(e.g., 1994 M
w
6.7 Northridge, California, 1999 M
w
7.6 Chi-Chi, Taiwan, 2005 M
w
7.5
Kashmir, Pakistan, and 2008 M
w
7.9 Wenchuan, China) highlights the need to better
understand the seismic hazards posed by these structures to an increasingly urbanized
global population. The keys to this understanding are the recognition of the rate at which
these faults store and release seismic energy, and the acquisition of detailed
paleoseismological data on the ages and displacements of ancient earthquakes that they
have generated. Although such data are routinely documented for many faults around the
world, one class of faults, the so-called “blind thrust faults”, in which the fault plane does
not extend all the way to the Earth’s surface, has proved particularly difficult to study.
Unlike most faults, which rupture to the surface in large earthquakes, near-surface
deformation above blind thrust faults is accommodated by folding, rather than faulting. In
many thrust fault earthquakes, co-seismic fold growth has been documented [King and
Vita-Finzi, 1981; Stein and King, 1984; Stein and Yeats, 1989; Lin and Stein, 1989; Lai et
al., 2006; Chen et al., 2007; Streig et al., 2007] and in very large thrust fault earthquakes,
discrete coseismic fold scarps have developed above bends and at the tips of propagating
faults [Stein and Yeats, 1989; Ishiyama et al., 2004; 2007; Lai et al., 2006; Chen et al.,
2007; Streig et al., 2007]. The youngest folded strata thus contain the information
necessary to understand the recent seismic behavior of these faults. However, extracting
197
paleoseismological data from blind thrusts has proved to be challenging, although the
development over the past several years of new inter-disciplinary methodologies for
documenting the sizes and ages of past events is beginning to yield a new understanding
of the behavior of this problematic class of faults at the time scales of individual
earthquakes [e.g., Dolan et al., 2003; Sugiyama et al., 2003; Ishiyama et al., 2004; 2007;
Leon et al., 2007]. The difficulties in studying the paleoseismology of blind thrusts are
particularly well illustrated by the Compton thrust, a major blind-thrust fault beneath
metropolitan Los Angeles. Although this fault was declared to be inactive on the basis of
earlier studies [Mueller, 1997], my previous investigations demonstrate that the Compton
fault is indeed active and capable of generating large-magnitude (M
w
> 7) earthquakes
[Leon et al., 2009 (Chapter 6)]. In this chapter I describe the results of an investigation of
the slip rate and paleo-earthquake history of the southern structural segment of the
Compton thrust at a site in Lakewood, California. I discuss the potential of multi-
segment rupture of both northern and southern segments of the Compton thrust and the
implications of these results for seismic hazard assessment in Los Angeles.
7.1.1 Compton blind-thrust fault
Nowhere are blind thrust faults of more concern than beneath metropolitan Los
Angeles, home to more than 10 million people. Previous studies have identified several
major blind thrust faults beneath the city, with two – the Compton and Puente Hills faults
– being of particular concern because of their location directly beneath the urban center
[Shaw and Suppe, 1996; Shaw et al., 2002]. The Compton thrust fault was originally
identified by Shaw and Suppe [1996] using petroleum-industry seismic reflection profiles
198
and well data. Based on their analysis of the folding that occurs above the thrust fault,
they showed that the Compton fault extends northwest-southeast for ~40 km beneath the
western edge of the Los Angeles metropolitan region (Figure 7.1). The industry seismic
reflection data define a growth fault-bend fold associated with the gently northeast-
dipping thrust fault ramp, which, combined with well data, reveal compelling evidence
for Pliocene and Pleistocene activity. But evidence for recent (latest Pleistocene-
Holocene) activity of the fault was lacking, as the original seismic reflection data did not
image the uppermost 200 m of the stratigraphic section. Moreover, in marked contrast to
the Puente Hills blind thrust fault, which generated the 1987 moment-magnitude (M
w
) 6.0
Whittier-Narrows earthquake southeast of Los Angeles [Shaw and Shearer, 1999], the
Compton ramp has produced neither damaging historical earthquakes nor abundant,
instrumentally recorded micro-earthquakes that might assist in delineating the fault and
proving that it was seismically active. The Pliocene/ Quaternary dip-slip rate is ~1.4+0.4
mm/yr for the entire Compton thrust ramp [Shaw and Suppe, 1996]. A comprehensive
study by Lehle [2007] analyzed forty industry seismic reflection profiles to provide
detailed mapping of the subsurface structure of the Compton-Los Alamitos fold trend,
and concluded that each fault segment experienced different slip rates during the past
~2.5 million years. The slip rate for the northern segment of the Compton fault remained
fairly consistent throughout Pliocene/ Quaternary time at 1.48+0.39 mm/yr. The
southern segment, however, experienced a marked increase in slip rate at ~1 Ma from
0.42 mm/yr to 2.69+0.67 mm/yr. Moreover, the temporally and spatially discrete slip
199
rates of each ramp segment suggest that either the faults merged early in the fault history
but were still able to behave in a segmented fashion, or that the fault has remained
segmented throughout most of its early history, but is capable of simultaneous, multi-
segment ruptures.
7.2 Results
Successful analysis of the current state of activity and earthquake history of blind
faults such as the Compton thrust requires a multi-disciplinary methodology that relates
near-surface deformation to the deep seismogenic fault through observation of the entire
depth range of folding. I begin by analyzing available data on the location, geometry, and
extent of the folds and their relationship to the underlying thrust fault. Such data may
include seismicity on the thrust fault, deep seismic reflection images and well data,
geological mapping, and geomorphologic analysis.
7.2.1 High-Resolution Seismic Reflection Data
In the case of the Compton fault, we have access to numerous, high-quality
petroleum industry seismic reflection data that provide images of the upper ~4-5 km of
the Compton-Los Alamitos fold. My study site is located along Briercrest Avenue in the
City of Lakewood (Figure 7.1). Several features combine to make this an excellent
location to study the paleo-earthquake history of the Compton thrust fault. Deep-
penetration petroleum industry seismic reflection profiles provide an exceptionally clear
image of the folding above the Compton thrust ramp beneath the site. In particular, these
data show that the back-limb axial surface - the locus of folding associated with fault slip
across the base of the Compton ramp at depth - exhibits an upward-narrowing growth
200
Figure 7.1 Map of the Compton thrust and adjacent structures in the Los Angeles
basin.
My study site is located at the locus of active, backlimb folding along the southern
segment of the fault (white star). A – A’ marks the trace of the seismic reflection
profile shown in Figure 7.2A and B – B’ indicates the trace of the high-resolution
seismic reflection profile in Figure 7.2B. Also shown are the locations of my
northern segment study site (white star; Chapter 6). The Mueller [1997] study sites
are located at the north-western end of the northern segment in Inglewood (white
triangle), and the south-eastern end of the southern segment in Los Alamitos (white
circles).
201
triangle that extends to within ~35 m of the surface [Shaw and Suppe, 1996]. This fault-
bend fold, referred to as the Compton – Los Alamitos trend, marks the base and
northeastern extent of the Compton ramp. My study site is located along this trend on a
previously marshy, extremely low-gradient floodplain, facilitating reconstruction of
originally near-horizontal floodplain surfaces and allowing collection of numerous
organic-rich samples for radiocarbon dating. The gentle, near-horizontal regional slope
at the study site has resulted in the deposition of laterally extensive sedimentary layers
that can be correlated over large distances.
To bridge the gap between the petroleum industry reflection data and the surface,
I acquired a 2.1 km long high-resolution seismic reflection profile utilizing a mini-vibe
source. Specifically, I used a geometrics seismograph with a 144-channel system and
mini-vibe T7000 source provided by the University of Nevada, Las Vegas. These
profiles were acquired with 60 hertz geophones at 5 m spacing. The high-resolution
seismic reflection data imaged reflectors from 30 to 1000 m depth range. Data
processing was routine for seismic reflection profiles and included velocity analysis,
residual statics corrections, pre- and post-stack bandpass filter and deconvolution, Stolt
time migration, and time-to-depth conversion. The combined industry and high-
resolution seismic reflection data provide an uninterrupted image of the full depth extent
of the fold above the Compton thrust ramp, from >4 km depth to within ~35 m of the
surface, allowing us to accurately site boreholes across the locus of most recently active
202
Figure 7.2 Southern Segment of the Compton Fault Seismic Reflection Data
(A) Seismic reflection image of the backlimb fold structure (in twtt [two way travel
time]) showing upward-narrowing zone of active folding (growth triangle) delimited
by sharply defined axial surface [Shaw and Suppe, 1996]. Dashed red line represents
the active, synclinal axial surface. Yellow box indicates location of mini-vibe source
profile shown in part B. Red box shows the location of the borehole transect shown
in Figure 7.3. (B) Mini-vibe source seismic reflection profile (migrated; no vertical
exaggeration). The vertical red lines indicate the location of cone penetration tests
(CPT’s) and the vertical yellow lines represent borehole locations. Dashed black
line represents the active, synclinal axial surface.
203
folding (Figure 7.2). I acquired continuously cored boreholes that overlap with the
shallow seismic reflection data and provide a continuous image all the way to the earth’s
surface (Figure 7.3).
7.2.2 Borehole Data
As is important with all paleoseismic studies, I chose a site characterized by
continuous, relatively rapid sediment accumulation, leading to stratigraphic separation of
event horizons and preservation of event stratigraphy. I specifically chose a study site in
which the uplifted hanging wall exhibited no topographic expression in order to ensure
that our site was located in an area of continuous aggradation.
I excavated a north-south transect of twelve, 25- to 42-m-deep, continuously
cored boreholes and eleven cone penetration tests (CPT’s) across the zone of young
folding imaged in the high-resolution seismic reflection profile. Excavation of boreholes
allows us to document the details of the most recent folding events, as well as to sample
the sediments for age control. The cores reveal sequences of fluvial sand and gravel,
interbedded with cohesive overbank silt and clay layers (Figure 7.3). Grain size was
visually identified using a hand-held GSA grain-size card, and we assigned soil colors
using a Munsell color chart on wet material. Deposition of these units in a low-gradient
floodplain has resulted in remarkable lateral continuity; more than a dozen units are
traceable continuously across the entire 1,340 m length of the borehole transect.
Specifically, these units exhibit identical stratigraphic sequences, with each unit defined
by its distinctive grain size, color, and texture.
204
Figure 7.3 Borehole Results from the Briercrest Avenue Transect
Cross section of major stratigraphic units (8x vertical exaggeration). Colors denote different sedimentary units.
Double-headed black vertical arrows along the right side of the figure show the stratigraphic ranges of intervals
of sedimentary thickening across the transect. Double-headed green vertical arrows show intervals of no
sedimentary growth
205
The uppermost ~20 meters of the stratigraphic section consists of fine-to medium-
grained sands, separated by units composed of cohesive silt and clay. At ~18 m depth
there is a marked change downward to a basal coarser-grained, sand and gravel-
dominated section. Major sand units are denoted by intervals of 10, from youngest to
oldest (e.g., 10, 20, 30, 40, 50, 60, and 70). Laterally continuous units of organic-rich,
cohesive silt-clay units have also been identified across the 1,340 m length of the
borehole transect. These units are numbered by adding 1 to the overlying major sand unit
(e.g. Unit 11 is the organic-rich soil beneath Unit 10 and Unit 21 is the organic-rich soil
beneath Unit 20).
7.2.2.1 Sediment Accumulation Rates
Calibrated radiocarbon dates from 42 bulk-soil and charcoal samples indicate that
the sediment accumulation rate averages 1.2 to 1.7 mm/yr over the past 32,000 years
(Table 7.1; Figure 7.4). This rate has varied quite considerably throughout Holocene and
latest Pleistocene time. Between ~32 ka and 13 ka, the sediment accumulation rate
averaged 0.6 mm/yr. There appears to be an unconformity due to a depositional hiatus
and/ or erosion that occurs within a fine-grained silt and clay unit above the basal coarse-
grained sand and gravel unit (Unit 70; Figure 7.3). The sediment accumulation rate
increased between 13 ka and 10 ka to 2.5 to 2.6 mm/yr. This rate subsequently
decreased throughout the Holocene to 2.1 mm/yr between 10 ka to 7.8 ka, then to the
mid-Holocene sediment accumulation rate of 1.7 mm/yr.
206
Table 7.1 Calibrated, Calendric Dates and Radiocarbon Ages of Samples from the
Briercrest Avenue Transect
All samples were calibrated using the Oxford Radiocarbon Accelerator Unit
(ORAU) calibration program OXCAL v4.11 [Bronk Ramsey, 1995; 2001; using
atmospheric data from Reimer et al., 2004 (IntCal04); Bronk Ramsey and van der
Plicht, 2001]. For sample type; ccl=charcoal fragment; bulk=bulk soil; bulk-h=bulk
soil humic. Calendric age is reported as 2 sigma (95% confidence limit) age range.
Samples 38 through 42 were calibrated using Voelker et al. and reported as
calibrated years before present.
207
Table 7.1, Continued
PDF Sample
number
Bore-
hole
Sample
type
Actual
depth
(m)
Projected
depth to
bh 12 (m)
14
C Age
BP
13
C Calendric age
(Cal BP)
%
pro
ba-
bilit
y
1 CLAF62 6 bulk 1.88 1.70 1405 + 35 -25 1371 – 1280 95.4
2 CLAF38 3 bulk 2.59 2.59 2315 + 45 -25 2463 – 2298
2258 – 2158
69.6
25.8
3 CLAF21 2 bulk 2.82 2.69 2435 + 35 -25 2701 – 2635
2617 – 2585
2572 – 2562
2545 – 2353
20.8
7.5
1.2
66.0
4 CLAF81 8 ccl 2.87 3.02 2245 + 30 -25 2341 – 2295
2270 – 2155
29.3
66.1
5 CLAF22 2 bulk 3.89 3.51 3635 + 35 -25 4082 – 4033
4006 – 3851
14.9
80.5
6 CLAF310 3 bulk 4.40 4.42 2455 + 35 -25 2705 – 2633
2618 – 2556
2551 – 2360
24.6
16.3
54.5
7 CLAF82 8 bulk 4.04 4.5 2045 + 35 -25 2115 – 1925 95.4
8 CLAF123 12 ccl 4.57 4.57 2460 + 35 -25 2706 – 2633
2619 – 2362
25.3
70.1
9 CLAF131 13 bulk 4.57 4.73 2665 + 35 -25 2845 – 2744 95.4
10 CLAF101 10 ccl 5.18 6.10 3005 + 35 -25 3330 – 3284
3272 – 3078
13.2
82.2
11 CLAF132 13 ccl 6.58 7.32 3590 + 35 -25 3985 – 3826
3790 – 3774
3744 – 3732
92.9
1.6
0.9
12 CLAF71 7 ccl 7.77 8.54 2215 + 35 -25 2331 – 2149 95.4
13 CLAF31 3 bulk 8.41 8.69 3595 + 40 -25 4070 – 4043
3990 – 3826
3791 – 3770
3745 – 3731
2.7
89.4
21
1.2
14 CLAF133 13 bulk 8.69 9.45 2155 + 35 -25 2308 – 2225
2209 – 2041
2017 – 2010
34.7
59.9
0.8
15 CLAF134 13 bulk 10.87 11.74 5145 + 35 -25 5990 – 5962
5952 – 5876
5826 – 5753
8.2
62.0
25.2
16 CLAF124 12 bulk 11.36 11.36 6360 + 40 -25 7418 – 7347
7342 – 7243
7215 – 7176
21.3
67.4
6.7
17 CLAF135 13 bulk 12.25 13.72 6820 + 35 -25 7706 – 7586 95.4
18 CLAF102 10 ccl 14.10 15.02 6300 + 35 -25 7301 – 7163 95.4
19 CLAF91 9 ccl 11.43 15.24 6915 + 40 -25 7838 – 7672 95.4
20 CLAF32 3 bulk 15.75 16.29 7475 + 50 -25 8380 – 8190 95.4
21 CLAF33 3 ccl 16.26 16.79 6060 + 35 -25 7003 – 6796 95.4
22 CLAF72 7 ccl 14.99 20.43 5730 + 35 -25 6636 – 6441 95.4
23 CLAF61 6 bulk 20.81 20.81 7875 + 35 -25 8951 – 8919
8862 – 8832
8780 – 8585
2.0
2.6
90.8
208
Table 7.1, Continued
PDF Sample
number
Bore-
hole
Sample
type
Actual
depth
(m)
Projected
depth to
bh 12 (m)
14
C Age
BP
13
C Calendric age
(Cal BP)
%
pro
ba-
bilit
y
24 CLAF52 5 ccl 20.655 20.655 7305 + 35 -25 8180 – 8025 95.4
25 CLAF63 6 ccl 21.7 21.7 9750 + 35 -25 11239 – 11134 95.4
26 CLAF73 7 ccl 17.73 24.39 7055 + 35 -25 10250 – 10185 95.4
27 CLAF34 3 bulk 24.08 24.39 10320 + 40 -25 12372 – 12265
12245 – 11988
17.4
78.0
28 CLAF65 6 bulk 24.67 24.67 8685 + 35 -25 9733 – 9720
9706 – 9545
1.8
93.6
29 CLAF64 6 ccl 24.695 24.695 9090 + 35 -25 10369 – 10357
10296 – 10191
1.1
94.3
30 CLAF23 2 ccl 29.8 29.8 9330 + 45 -25 10683 – 10407 95.4
31 CLAF125 12 ccl 31.1 31.1 10170+40 -25 12042 – 11705
11663 – 11652
94.6
0.8
32 CLAF136 13 bulk 27.185 31.4 10425+35 -25 12602 – 12456
12405 – 12124
24.0
71.4
33 CLAF121 12 bulk 32.01 32.01 10510+35 -25 12675 – 12385 95.4
34 CLAF35 3 ccl 31.5 31.89 17770+60 -25 21327 – 20654 95.4
35 CLAF74 7 bulk 22 32.01 11900+80 -25 13966 – 13575 95.4
36 CLAF37 3 bulk 32.75 32.75 9915+35 -25 11591 – 11578
11404 – 11230
0.6
94.8
37 CLAF36 3 ccl 33.435 33.435 10425+50 -25 12620 – 12446
12411 – 12099
27.6
67.8
38 CLAF122 12 bulk 41.16 41.16 29290+157
0
-25 33950 – 26450 V
39 CLAF103 10 ccl 38.03 41.62 25050+290 -25 25750 – 24450 V
40 CLAF105 10 bulk 38.03 41.62 28200+290 -25 30150 – 25050 V
41 CLAF111 11 bulk 35.82 42.23 27250+118
0
-25 28900 – 27600 V
42 CLAF83 8 ccl 31.7 ~54.57 42600+110
0
-25 45300 – 40600 V
209
Figure 7.4 Sediment Accumulation Rate Curve for the Briercrest Avenue Transect
Sediment accumulation-rate curve for the Briercrest Avenue transect, showing
calibrated, calendric radiocarbon dates sorted by depth (green shading). Alternative
sediment accumulation rate curve is indicated by the dashed green lines. Due to the
ambiguity in ages I have included both possible sediment accumulation rate curves
for completeness. Radiocarbon dates were calibrated using the Oxford
Radiocarbon Accelerator Unit (ORAU) calibration program OXCAL v4.11 [Bronk
Ramsey, 1995; 2001; using atmospheric data from Reimer et al., 2004 (IntCal04);
Bronk Ramsey and van der Plicht, 2001]. True depth has been corrected for folding
by projecting all samples to a common reference point (Borehole CLAF12), moving
each sample along the nearest stratigraphic horizons to its projected depth in
borehole CLAF12. Numbers next to probability functions represent sample
numbers in Table 7.1. Sediment accumulation-rate curve is green. Uncertainties in
the ages of events (Event 1 through Event 3) are marked by horizontal red lines,
whereas depth uncertainties are marked by vertical red lines. BH# indicates
borehole location for each sample. Sample type, depth and calendric age shown in
Table 7.1.
210
Figure 7.4, Continued
211
Figure 7.4 Continued
PDF Sample#
14
C Date BH# PDF Sample#
14
C Date BH#
1 CLAF62-BS 1405+35BP CLAF6 22 CLAF72 5730+35BP CLAF7
2 CLAF38-BS 2315+45BP CLAF3 23 CLAF61-BS 7875+35BP CLAF6
3 CLAF21-BS 2430+35BS CLAF2 24 CLAF52 7305+35BP CLAF5
4 CLAF81 2245+30BP CLAF8 25 CLAF63 9750+35BP CLAF6
5 CLAF22-BS 3635+35BP CLAF2 26 CLAF73 7055+35BP CLAF7
6 CLAF310-BS 2455+35BP CLAF3 27 CLAF34-BS 10320+40BP CLAF3
7 CLAF82-BS 2045+35BP CLAF8 28 CLAF65-BS 8685+35BP CLAF6
8 CLAF123 2460+35BP CLAF12 29 CLAF64 9090+35BP CLAF6
9 CLAF131-BS 2665+35BP CLAF13 30 CLAF23 9330+45BP CLAF2
10 CLAF101 3005+35BP CLAF10 31 CLAF125 10170+40BP CLAF12
11 CLAF132 3590+35BP CLAF13 32 CLAF136 10425+35BP CLAF13
12 CLAF71 2215+35BP CLAF7 33 CLAF121-BS 10510+35BP CLAF12
13 CLAF31-BS 3595+40BP CLAF3 34 CLAF35 17770+60BP CLAF3
14 CLAF133 2155+35BP CLAF13 35 CLAF74-BS 11900+80BP CLAF7
15 CLAF134-BS 5145+35BP CLAF13 36 CLAF37-BS 9915+35BP CLAF3
16 CLAF124-BS 6360+40BP CLAF12 37 CLAF36 10425+50BP CLAF3
17 CLAF135-BS 6820+35BP CLAF13 38 CLAF122-BS 29290+1570BP CLAF12
18 CLAF102 6300+35BP CLAF10 39 CLAF103 25050+290BP CLAF10
19 CLAF91 6915+40BP CLAF9 40 CLAF105-BS 28200+290BP CLAF10
20 CLAF32-BS 7475+50BP CLAF3 41 CLAF111-BS 27250+1180BP CLAF11
21 CLAF33 6060+35BP CLAF3 42 CLAF83 42600+1100BP CLAF8
212
7.2.3 Cone Penetration Testing Data
In addition to the twelve continuously cored borehole excavations, I acquired 11
cone penetration tests (CPT’s) to provide details of the stratigraphy within the zones of
deformation encountered at our study sites. Although cores are not collected directly,
CPT analysis combines three main parameters for soil behavior that are generally
consistent with the grain size of stratigraphic units logged from nearby borehole
excavations. The interpreted soil behavior (SBT) is reported for depths of up to 50 feet
after Robertson et al. [1986] and for depths greater than 50 feet SBTn is calculated after
Robertson [1990]. These geotechnical parameters are interpreted by the software as 12
“soil behavior zones” that may range from sensitive, fine-grained material to sand-
gravelly sand. These sedimentary characteristics are reported as vertical logs for each
CPT hole. The CPT logs provided additional confidence in correlating stratigraphic units
across the zone of deformation located by high-resolution seismic reflection and
continuously cored borehole data (Figure 7.3).
7.2.4 Stratigraphic Evidence for Paleo-Folding Events
The borehole data reveal evidence for a discrete, upward-narrowing zone of
folding located at the updip projection of the synclinal axial surface seen on the industry
and high-resolution seismic reflection data (Figure 7.2). Specifically, the boreholes show
that all strata dip near-horizontally, parallel to the surface of the active floodplain, at the
northern and southern ends of the transect, but that the tops of all strata are deeper at the
northern end of the transect. This structural relief is produced by a panel of north-dipping
strata that lies directly at the up-dip projection of the growth triangle imaged
213
continuously on the seismic reflection data from >4 km depth. The continuity of folding
illustrated by the combination of seismic reflection and borehole data provides a direct
link between the deformation that we observe in the upper few tens of meters and the
base of the Compton thrust ramp at >7 km.
In situations such as we encountered at my study site, when deep seismic slip on
the thrust ramp is translated to the surface as fold growth leading to surface uplift, a fold
scarp develops at the updip tip of the active axial surface. If a laterally extensive fold
scarp develops at a high angle to a river, as at my study site, the river will immediately
begin to adjust its bed in order to return to the pre-earthquake, equilibrium stream
gradient. This may be achieved by either deposition of sediment on the downthrown
(upstream, in this case) side of the scarp, leading to growth of a thickened sedimentary
package until the pre-folded stream gradient is re-established, and/or erosion of the
upthrown (downstream, in this case) side of the scarp. The difference in sediment
accumulation rates is thus a direct manifestation of sedimentary growth across the fold.
In the Briercrest Avenue transect, the northward stratigraphic thickening (or
“growth”) across this fold limb occurs in several discrete intervals that thicken
northward, reflecting the relative uplift of the hanging wall during folding and the
subsequent, post-uplift deposition that restores the original near-horizontal stream
gradient. We see exceptionally well-defined growth in exactly the same location
throughout the stratigraphic section. This argues strongly that we are seeing a tectonic
signal and that any natural variation of the floodplain is not a significant source of error.
These growth strata are interbedded with intervals that do not change thickness, reflecting
214
deposition during periods of structural quiescence at the near-horizontal floodplain
gradient.
My initial results suggest that at least three stratigraphically discrete, now-buried
fold scarps are present in the upper 16 m. The lower 26 m of the section exhibits
significant structural relief across the transect, which we interpret as evidence for
multiple uplift events, but additional analysis of the fine-grained stratigraphy will be
required to identify individual events in this part of the section. In the youngest three
uplift events, fold growth has uplifted the strata to the south of the zone of active folding,
resulting in a north-facing fold scarp that has been onlapped by the next-youngest
stratigraphic unit. I term these uplift Events 1 through 3 (from youngest to oldest), and
interpret them as evidence for large-magnitude (M
w
>7.0) earthquakes on the Compton
fault (Table 7.2).
7.3. Discussion
7.3.1 Fault Slip Rate Estimates
To determine the reverse slip rate of the fault, we need to combine my measured
minimum uplifts with the dip of the causative Compton thrust ramp. The fault dip was
defined by the dip of the fold limb (24
o
+2
o
) imaged in industry seismic reflection profiles
[Suppe et al., 1992; Shaw and Suppe, 1996], and corrected for apparent dip. Uncertainties
in the velocity structure that was used to depth process the seismic reflection data [Süss
and Shaw, 2003] leads me to estimate conservative fault-plane dip errors for the Compton
ramp at +2
o
. Slip estimates for individual events discussed below include these fault-
plane dip uncertainties. It is possible that the dip of the Compton ramp may be greater
215
than the dip of the overlying fold limb [Shaw and Suppe, 1996], which would imply that
the slip and slip rates required to generate the observed uplift would be less than we
calculate. However, the interaction of the Compton ramp with adjacent structures
strongly favors a dip of the ramp that is similar to that of the overlying fold limb. Thus,
we favor this interpretation in our analysis. For further discussion on the dip of the
Compton ramp, see Shaw and Suppe [1996].
I divide the uplift measurement from the op of Unit 70 sand and gravel by the sine
of the 24°±2° northeast dip of the Compton ramp and derive a cumulative thrust
displacement over the past 32,000 years of 19.5 m. This yields a minimum reverse slip
rate of 1.7+0.4 mm/yr. This rate is a minimum because: (1) our uplift measurements do
not account for any potential erosion of the uplifted block; (2) it assumes that all slip on
the deep Compton ramp is manifested as folding within the kink band. Moreover, our
analysis of industry seismic reflection data shows that at least 90% (and possibly almost
all) of the total structural relief due to folding of Quaternary horizons above the Compton
thrust fault occurs within the growth triangle. Thus, although we report this rate as a
minimum, we think that it is probably close to the actual slip rate of the Compton thrust
beneath the Briercrest Avenue site. Using a large number of high-quality, deep-
penetration industry seismic reflection profiles, Lehle [2007] measured displacements at
various time intervals, ranging from 0.45 Ma to 2.4 Ma along the length of the Compton
thrust. Her data demonstrate that displacement at my Briercrest Avenue study site occurs
at approximately 75% of the maximum long-term slip-rate along the Compton thrust,
which occurs along the central part of the fault approximately 5 km north of our site. If
216
the long-term displacement patterns documented by Lehle [2007] extend to the present,
these observations suggest that the maximum slip rate along the central part of the
Compton thrust fault may be on the order of ~2.0 +0.8/-0.5 mm/yr.
7.3.2 Paleo-magnitude Estimates and Implications for Fault Behavior
The observation that uplift occurs in stratigraphically discrete intervals rules out
the possibility that this fold grew steadily, either through many smaller-magnitude
earthquakes or by quasi-continuous fault creep, on the Compton ramp. I cannot,
however, rule out the possibility that some component of folding occurs during
punctuated periods of aseismic fault slip, either as afterslip following large earthquakes
or during discrete periods of interseismic fold growth. It is also possible that each
stratigraphically discrete uplift event records a brief cluster of multiple, moderate-
magnitude earthquakes. If each uplift event does record multiple, smaller-magnitude
earthquakes, however then these must have occurred during relatively brief clusters.
Moreover, uplift and fold growth are likely to occur only during large-magnitude events
that involve displacement of the hanging wall through and across changes in the fault dip
(e.g., from the flat down-dip from the Compton ramp through the backlimb active, axial
surface). Small- to moderate-magnitude ruptures that are limited to the thrust ramp are
unlikely to localize deformation at the active, axial surfaces and will therefore not
contribute substantially to the fold growth that we measure at the surface (e.g. the 1987
M
w
6.0 Whittier Narrows earthquake; Lin and Stein, [1989]). Thus, almost all growth of
the Compton-Los Alamitos fold probably occurs in large-magnitude events.
217
Table 7.2 Uplift Amounts, Age Limits, and Estimated Paleo-magnitudes for the Southern
Segment of the Compton Blind-Thrust Fault
Event Age (ka) Growth
section
Uplift
(m)
Slip (m) M
w
(All-
slip-type
avg.
displace-
ment)
M
w
(All-
slip-type
max.
displace-
ment)
M
w
(Thrust-
fault-only
avg.
displace-
ment)
M
w
(Thrust-
fault-only
max.
displace-
ment)
1 0 – 2.2 ka Surface
to top of
Unit 10
0.6 +
0.15
1.5+ 0.5/-
0.3
7.1 + 0.1 6.8 + 0.3 6.7 +0.5/-0.2 6.6 -0.2/+0.4
2 2.4 – 4.0
ka
Top of
Unit 11
to top of
Unit 20
0.8 +
0.5/-
0.1
2.0 +1.5/-
0.4
7.2 + 0.1 6.9 + 0.2 6.7 +0.5/-0.2 6.6 -0.2/+0.4
3 3.8 – 7.8
ka
Top of
Unit 20
to top of
Unit 30
1.6 +
0.5
3.9 +1.7/-
1.0
7.4 + 0.1 7.1 + 0.2 6.7 +0.5/-0.2 6.8 -0.2/+0.4
Uplift, age limits, and estimated moment-magnitude (M
w
) for Compton fault paleo-earthquakes 1
through 3 from the Briercrest Avenue borehole results. Results based on the assumption that our
measured displacements represented either the average or the maximum displacement in each
earthquake [Wells and Coppersmith, 1994]
218
I estimate a conservative range of magnitudes for these paleo-earthquakes by comparing
my measured minimum displacements to global regressions of slip vs. magnitude [Wells and
Coppersmith, 1994], recognizing that the displacements measured in this study
represent slip at only one point along the Compton fault in earthquakes that likely had
laterally variable slip. If I make the simplifying assumption that my measured
displacements represent average displacement in each paleo-earthquake, the magnitude
estimates for the Compton thrust earthquakes range from M
w
7.1 + 0.1 to M
w
7.4 + 0.1
(Table 7.2). Alternatively, if I assume the extreme limiting case that all of my measured
displacements record the maximum displacements in each earthquake, the minimum
magnitude estimates range from M
w
6.8 + 0.3 to M
w
7.1 + 0.2. It is, of course, possible
that slip in any of the individual earthquakes I measured was the minimum slip in that
earthquake, but use of our displacements as minimum slip in the Wells and Coppersmith
[1994] regressions yields unreasonably large M
w
estimates, so I do not include them here.
If I use the thrust-fault-only regressions of Wells and Coppersmith [1994], rather than
their all-slip-type regressions, and each of my displacement measurements represents an
average displacement, the magnitude estimates are M
w
6.7 -0.2/+0.5. If my
measurements record maximum displacement, then the thrust-fault-only magnitude
estimates range from M
w
6.6 -0.2/+0.4 to M
w
6.8 -0.2/+0.4. The thrust regressions,
however, have many fewer data points than the regressions based on all-slip-type data,
and the uncertainty in the mean is therefore larger for thrust-slip-only regression [Wells
and Coppersmith, 1994]. As a point of comparison, the 1994 M
w
6.7 Northridge
earthquake had a rupture area of ~300 km
2
[Wald et al., 1996], relative to the ~550 km
2
219
are of the Compton thrust ramp [Shaw and Suppe, 1996]. Although, this leads me to
suspect that the Wells and Coppersmith [1994] thrust-only displacement regressions
underestimate the magnitude of the Compton fault paleo-earthquakes that I document, I
have nevertheless included them here for the sake of completeness.
The data discussed in this chapter complement similar data paleo-earthquake data
acquired at the Stanford site (Chapter 6), along the backlimb fold above the northern
segment of the Compton fault. Thus I can compare the earthquake ages at our southern
segment Compton fault site to assess the possibility that the entire fault ruptures co-
seismically in fault-wide ruptures. The results from this study suggest that the
preliminary calculated displacements for Events 1 through 3 for the Briercrest Avenue
study site above the southern segment of the Compton thrust are similar to the calculated
displacements from the Stanford Avenue study site above the northern segment of the
Compton fault. Although we cannot uniquely confirm that the upper three co-seismic
folding events observed at the northern segment site of the Compton fault are the same
events observed in the southern segment study site, the timing constraints are permissive
of large-magnitude, multi-segment ruptures. Additional analysis should identify
individual uplift events throughout the lower part of the section at this study site.
Based on regressions of rupture area to magnitude (rather than slip), rupture of the
entire Compton ramp could generate an earthquake of ~M
w
7.1 [Shaw and Suppe, 1996],
in line with the smaller end of the magnitude estimates based on my measured
displacements. I consider it highly unlikely, however, that all of the minimum
displacements that I measured represent the largest displacements in all of these paleo-
220
earthquakes. Thus, these events may have involved slip on faults other than just the
Compton ramp. Perhaps the most obvious explanation is that the decollement at the base
of the Compton ramp [Shaw and Suppe, 1996] also slipped in these earthquakes.
Alternatively, the Compton ramp may be underlain by a southwest-dipping fault, forming
a structural wedge, that also ruptured in these events. From a kinematic standpoint, these
scenarios seem plausible, as motion extending across this decollement-to-ramp transition
or at the wedge tip causes the large-scale folding that we observe in the seismic reflection
and borehole data [Shaw and Suppe, 1996].
7.3.3 Incremental Development of Dips within the Kink Band
As at our northern segment study site (Chapter 6) the combined borehole and
seismic reflection data described above for the Lakewood site indicate that strata folded
within the kink band acquired their dips incrementally. The progressive downward
increase in bed dip genuinely reflects an aspect of fold kinematics that is distinct from
classic growth fault-bend fold theories. This change in limb dips may reflect some
component of fold growth by limb rotation in addition to kink-band migration. These
hybrid kinematics are manifested in several types of fault-related folds, including trishear
fault-propagation folding [Erslev, 1991; Allmendinger, 1998] and shear fault-bend
folding [Suppe and Connors, 2004; Shaw et al., 2005]. Alternatively, the component of
limb rotation that I observe may reflect curvature of the fold hinge, as is described by
curved-hinge fault-bend folding theories [Suppe et al., 1997; Novoa et al., 2000].
Finally, the progressive change in bed dips may reflect the mechanical response of
loosely consolidated, granular sediments in the shallow subsurface to folding at depth.
221
Recent work by Benesh et al., [2007] demonstrates that, even in cases of folding of pre-
tectonic layers exclusively by kink-band migration, shallow growth sediments acquire
their dips progressively. This is due to the finite width of the axial surface zone and its
rotation and migration with increasing deformation.
7.3.4 Implications for Seismic Hazard in Southern California
Although earlier studies have effectively demonstrated the hazard associated with
the Puente Hills thrust [Shaw and Shearer, 1999; Pratt et al., 2002; Shaw et al., 2002;
Dolan et al., 2003; Field et al., 2005], the results of previous studies of the Compton
thrust [Mueller, 1997] have taken a very different path through both the regulatory
process and the public consciousness. Two previous studies involving paleoseismologic
trenches and closely spaced geotechnical penetration tests acquired across the fold trace
above the Compton thrust yielded indeterminate results [Mueller, 1997] leading to the
deletion of the Compton fault from the California Geological Survey (CGS) 2002 active
seismic sources. In one of these two earlier studies, the researchers interpreted the
presence of flat-lying strata delineated in cone-penetrometer tests and shallow
paleoseismologic trenches excavated across the expected location of near-surface folding
in Los Alamitos as evidence for structural quiescence [Mueller, 1997]. Their study site,
however, was located near the southernmost extent of the southern segment of the
Compton fault and its overlying fold, where slip may be transferred to the southwest on
an en echelon structure (Figure 7.1). Industry seismic reflection data to the south of this
excavation site indicate that Quaternary strata in the uppermost 500 meters dip only a few
degrees to the northeast, in contrast to the ≥ 25° dips found elsewhere along the trend.
222
Given my observation that the dips of strata decrease significantly upward in the
Holocene section above the growth triangle, this suggests in hindsight that no discernable
trace of folding would be apparent at the Mueller [1997] excavation site, even if the
thrust ramp beneath the site was seismically active. A second study [Mueller, 1997],
using closely spaced geotechnical penetration tests across a topographically prominent
suspected fold scarp at the up-dip projection of the locus of recent folding observed on
seismic reflection profiles (Figure 7.1) indentified evidence of young folding at <3 km
depth. The researchers concluded, however, that shallow folding along the trend did not
reproduce the growth structures imaged at depth, and thus did not confirm the Compton
thrust as an active fault. The Compton thrust is currently included in the Southern
California Earthquake Center (SCEC) community fault model (CFM) [Plesch and Shaw,
2007] as a potentially active fault due largely to my studies.
The prospect for future large earthquakes on the Compton thrust similar to past
events documented in this study poses a significant seismic hazard to the Los Angeles
metropolitan region. As a point of comparison, the 1994 Northridge (M
w
6.7)
earthquake, which ruptured a similar type of fault but was much smaller than the events
documented here on the Compton ramp, caused widespread damage and remains one of
the costliest natural disasters in U.S. history [Scientists of USGS and SCEC, 1994].
Moreover, blind thrust earthquakes of the type forecast on the Compton thrust that occur
directly beneath metropolitan regions have been shown to produce very large ground
motions that pose particular concerns to high-rise buildings [Heaton et al., 1995; Field et
al., 2005; Sommerville and Pitarka, 2006; Abrahamson et al., 2008]. Thus, it is critical
223
that the activity and seismogenic potential of the Compton thrust be properly considered
in regional seismic hazards assessment. Finally, faults similar to the Compton blind
thrust lie beneath many cities around the world (e.g., Seattle, Washington [e.g., Liberty
and Pratt, 2008], Tokyo and Osaka, Japan [e.g., Sugiyama et al., 2003]), and the multi-
disciplinary methodology described in this paper provides a means of defining the recent
seismic behavior, and therefore the hazard, posed by such structures to an increasingly
urbanized global population.
224
Chapter 7 References
Abrahamson, N., Atkinson, G., Boore, D., Bozorgnia, Y., Campbell, K., Chiou, B., Idriss,
I. M., Silva, W., and Youngs, R. (2008), Comparisons of the NGA Ground-
Motion Relations, Earthquake Spectra, 34, 1, 45-66.
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Field, E. H., H. A. Seligson, N. Gupta, V. Gupta, T. H. Jordan, and K. W. Campbell
(2005), Loss estimates for a Puente Hills blind-thrust earthquake in Los Angeles,
California, Earthquake Spectra, 21, 329-338.
Field., E. H. (2005), Collaborative SCEC/USGS efforts to improve seismic-hazard analysis;
RELM and OpenSHA: Proceedings of the 5th U.S.-Japan natural resources meeting and
Parkfield, California fieldtrip, Open-File Report - U. S. Geological Survey, Report: OF
2005-1131, pp. 18. Resource Location: http://pubs.usgs.gov/of/2005/1131/
225
Heaton, T. H., J. F. Hall, D. J. Wald, and M. W. Halling (1995), Response of high-rise
and base-isolated buildings to a hypothetical Mw 7.0 blind thrust earthquake,
Science, 267, 206-211.
Ishiyama, T., K. Mueller, M. Togo, A. Okada, and K. Takemura (2004), Geomorphology,
kinematic history, and earthquake behavior of the active Kuwana wedge thrust
anticline, central Japan, J. Geophys. Res., 109, B12408, doi:
10.1029/2003JB002547.
Ishiyama, T., K. Mueller, H. Sato, and M. Togo (2007), Coseismic fault-related fold
model, growth structure, and the historic multisegment blind thrust earthquake on
the basement-involved Yoro thrust, central Japan, J. Geophys. Res., 112, B03S07,
doi: 10.1029/2006JB004377.
King, G. C. P., and C. Vita-Finzi (1981), Active folding in the Algerian earthquake of 10
October 1980, Nature, 292, 22-26.
Lai, K., Y. Chen, J. Hung, J. Suppe, L. Yue, and Y. Chen (2006), Surface deformation
related to kink-folding above an active fault: Evidence from geomorphic features
and co-seismic slips, Quat. Int., 147, 44-54.
Lehle, D. (2007), Geometry and slip history of the Compton thrust fault: Implications for
earthquake hazard assessment, thesis, Harvard Univ., Cambridge, Mass.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Leon, L. A., J. F. Dolan, J. H., Shaw, and T. L. Pratt (2009), Evidence for large Holocene
earthquakes on the Compton thrust fault, Los Angeles, California, J. Geophys.
Res., in press, doi: 10.1029/2008JB006129.
Liberty, L. M., and T. L. Pratt (2008), Structure of the eastern Seattle fault zone,
Washington state: New insights from seismic reflection data, Bull. Seismol. Soc.
Am., 98, 4, 1681-1695, doi: 10.1785/0120070145.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Mueller, K. J., (1997), Recency of folding along the Compton-Los Alamitos trend:
Implications for seismic risk in the Los Angeles basin, EOS Trans. AGU, 78,
F702.
226
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Plesch, A., and J. H. Shaw (2007), Community fault model (CFM) for Southern
California, Bull. Seismol. Soc. Am., 97, 6. 1793-1802.
Pratt, T. L., J. H. Shaw, J. F. Dolan, S. A. Christofferson, R. A. Williams, J. K. Odum,
and A. Plesch (2002), Shallow folding imaged above the Puente Hills blind-thrust
fault, Los Angeles, California, Geophys. Res. Lett., 29, 18-1 - 18-4.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Roberston, P.K. (1990), Soil Classification using the Cone Penetration Test, Canadian
Geotech. Journ., v. 27, p 151-158.
Robertson, P.K., Campanella, R. G., Gillespie, D., and Rice, A. (1986), Seismic CPT to
Measure In-Situ Shear Wave Velocity, J. Geotech. Eng. ASCE, v. 112, No. 8, p
791-803.
Scientists of the U. S. Geological Survey and the Southern California Earthquake Center
(1994), The magnitude 6.7 Northridge, California, earthquake of 17 January 1994,
Science, 266, 389-397.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
Sommerville, P. G., and A. Pitarka (2006), Differences in earthquake source and ground
motion characteristics between surface and buried earthquakes, Proceedings of
the Eighth National Conference on Earthquake Engineering, San Francisco,
California, U.S.A., 977, Earthquake Engineering Research Institute.
227
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
Stein, R. S., and R. S. Yeats (1989), Hidden earthquakes, Sci. Am., 260, 48-57.
Streig, A. R., C. M. Rubin, W. Chen, Y. Chen, L. Lee, S. C. Thompson, C. Madden, and
S. Lu (2007), Evidence for prehistoric coseismic folding along the Tsaotun
segment of the Chelungpu fault near Nan-Tou, Taiwan, J. Geophys. Res., 112,
B03S06, doi: 10.1029/2006JB004403.
Sugiyama, Y., K. Mizuno, F. Nanayama, T. Sugai, H. Yokota, T. Hosoya, K. Miura, K.
Takemura, and N. Kitada (2003), Study of blind thrust faults underlying Tokyo
and Osaka urban areas using a combination of high-resolution seismic reflection
profiling and continuous coring, Annals of Geophys., 46, 5.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
Süss, M. P., and J. H. Shaw (2003), P wave seismic velocity structure derived from sonic
logs and industry reflection data in the Los Angeles basin, California, J. Geophys.
Res., 108(B3), 2170, doi:10.1029/2001JB001628.
Wald, D. J., T. H. Heaton, and K. W. Hudnut (1996), The slip history of the 1994
Northridge, California, earthquake determined from strong-motion, teleseismic,
GPS, and leveling data Bull. Seismo.l Soc. Am., 86, 1B, S49-S70.
Wells, D. L., and K. J. Coppersmith (1994), New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface displacement,
Bull. Seismol. Soc. Am., 84, 974-1002.
228
CHAPTER 8:
Conclusions
8.1 Summary
I have exploited a multi-disciplinary methodology that combines high-resolution
seismic reflection data with borehole excavations and cone penetration testing to quantify
Holocene-latest Pleistocene slip rates for the Puente Hills thrust and Compton fault.
Moreover, this study documented that folding associated with slip on blind thrust ramps
is highly localized within narrow zones along both of these segmented blind-thrust
systems. The results of this research have led to a better understanding of both the
usefulness of long-term versus short-term slip rates for seismic hazard analysis, and
whether these segmented blind-thrust systems are capable of generating large, multi-
segment ruptures. Additionally, this study provided insights into the detailed kinematics
of earthquake-by-earthquake fold growth above the underlying blind thrust ramps. At all
our study sites, the borehole data show that the folded strata within the kink bands
acquired their dips incrementally, suggesting that fold kinematics involves components of
both kink-band migration and limb rotation. Alternatively, the progressive change in bed
dips may reflect the mechanical response of loosely consolidated, granular sediments in
the shallow subsurface to folding at depth. Recent work by Benesh et al., [2007]
demonstrates that, even in cases of folding of pre-tectonic layers exclusively by kink-
band migration, shallow growth sediments acquire their dips progressively. This is due
to the finite width of the axial surface zone and its rotation and migration with increasing
deformation.
229
These analyses of the geometry and structural evolution of young, shallow growth
strata have allowed us to decipher the previously unknown paleo-earthquake history of
the PHT and Compton fault, providing the basis for effective seismic hazard analysis.
The following discussion summarizes the results from this research.
8.2 Spatial and Temporal Variations in Slip Rates
By calculating slip rates from multiple study sites for the Puente Hills thrust, I
was able to show that Holocene-latest Pleistocene slip rates are similar to the long-term
(Plio-Quaternary) slip rates calculated for both the Los Angeles and Coyote Hills
segments of the Puente Hills thrust, but that the slip rate may have accelerated during late
Quaternary time on the Santa Fe Springs segment [Shaw et al., 2002].
In contrast, the laterally variable long-term slip rate of the Compton fault implies
that aspects of the system-level behavior of this segmented thrust system may
significantly influence the slip rate of the fault. For example, similarity of the 1.2 +0.5/-
0.3 slip rate I calculate with the 1.5+0.4 mm/yr long-term slip rate determined by Lehle
[2007] has remained relatively constant for the past ~2.5 million years. Moreover, the
1.2 +0.5/-0.3 mm/yr Holocene-latest Pleistocene slip rate calculated for the northern
segment suggests that the slip rate has continued to remain fairly constant throughout
Holocene time. The slip rate calculated for southern segment, however, changes at ~1
Ma from 0.4 mm/yr to 2.7+0.7 mm/yr, decreasing again for the Holocene-late Pleistocene
to 1.7 +/-0.4 mm/yr. This suggests that long-term slip rates for the Compton fault may
not be an effective proxy for shorter-term Holocene slip rates.
230
8.3 Implications for the Potential of System-Wide Earthquakes to Occur
The timing of events and calculation of uplift for paleoearthquakes recorded at
multiple study sites also facilitates comparison of uplift events between individual thrust
segments, allowing us to assess the degree to which these segments fail together, or
separately. Although, as with all paleoseismologic studies, we could not uniquely
confirm that a co-seismic folding event at the Los Angeles site of the PHT is the same
event observed at the Santa Fe Springs site, the timing and amount of displacement that
occurred in each event measured at the two sites suggests that the PHT is capable of
large-magnitude, multi-segment ruptures. Although we cannot rule out the possibility
that each stratigraphically discrete period of uplift represents a cluster of smaller events,
our preliminary data suggests that the large displacements observed at the study sites
above the Santa Fe Springs and Los Angeles segments of the PHT are most consistent
with large-magnitude, multi-segment ruptures. Our initial results for the Compton fault,
particularly the evidence for large (2-4 m) displacements in the paleo-earthquakes that
generated the fold scarps, as well as the timing of uplift events at both the northern and
southern segment study sites, are also most consistent with the idea of large-magnitude,
multi-segment ruptures.
8.4 Structural Characterization of Shallow Sub-Surface Folds
Analysis of the combined high-resolution seismic reflection and borehole data
indicates that deformation is localized within a discrete zone of folding at all our study
sites above both the Puente Hills thrust and the Compton. This upward-narrowing
growth triangle is recorded in petroleum-industry data that extend upward into the
231
shallow sub-surface (,250 m). The growth structures imaged in shallower, high-
resolution seismic reflection data I collected record deformation localized along axial
surfaces in kink bands, consistent with the deeper petroleum-industry data. This research
also indicates that folded strata within the kink band acquired their dips incrementally.
The progressive downward increase in bed dip genuinely reflects an aspect of fold
kinematics that is distinct from classic growth fault-bend fold theories. This change in
limb dips may reflect some component of fold growth by limb rotation in addition to
kink-band migration. These hybrid kinematics are manifested in several types of fault-
related folds, including trishear fault-propagation folding [Erslev, 1991; Allmendinger,
1998] and shear fault-bend folding [Suppe and Connors, 2004; Shaw et al., 2005].
Alternatively, the component of limb rotation that we observe may reflect curvature of
the fold hinge, as is described by curved-hinge fault-bend folding theories [Suppe et al.,
1997; Novoa et al., 2000]. Finally, the progressive change in bed dips may reflect the
mechanical response of loosely consolidated, granular sediments in the shallow
subsurface to folding at depth.
232
Chapter 8 References
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Lehle, D. (2007), Geometry and slip history of the Compton thrust fault: Implications for
earthquake hazard assessment, thesis, Harvard Univ., Cambridge, Mass.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P., (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p. 2946-
2960.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
233
REFERENCES
Abrahamson, N., Atkinson, G., Boore, D., Bozorgnia, Y., Campbell, K., Chiou, B., Idriss,
I. M., Silva, W., and Youngs, R. (2008), Comparisons of the NGA Ground-
Motion Relations, Earthquake Spectra, 34, 1, 45-66.
Allmendinger, R. W. (1998), Inverse and forward numerical modeling of trishear fault-
propagation folds, Tectonics, 17, 4, 640-656.
Allmendinger, R. W., and Shaw, J. H. (2000), Estimation of fault propagation distance
from fold shape: Implications for earthquake hazard assessment: Geology, v. 28,
n. 12, 1099-1102.
Argus, D. F., Heflin, M. B., Donnellan, A., Webb, F. H., Dong, D., Hurst, K. J.,
Jefferson, D. C., Lyzenga, G. A., Watkins, M. M., and Zumberge, J. F. (1999),
Shortening and thickening of metropolitan Los Angeles measured and inferred by
using geodesy: Geology, v. 27, n. 8, 703 – 706.
Argus, D. F., Heflin, M. B., Peltzer, G., and Crampe´, F. (2005), Interseismic strain
accumulation and anthropogenic motion in metropolitan Los Angeles, J. Geophys.
Res., 110, B04401, doi: 10.1029/2003JB002934.
Barbier, M. G. (1983), The mini-sosie method: International Human Resources
Development Corporation, Boston, MA, 186 pp.
Bawden, G. W., Thatcher, W., Stein, R. S., Hudnut, K. W., and Peltzer, G. (2001),
Tectonic contraction across Los Angeles after removal of groundwater pumping
effects: Nature, 412, 712-815.
Benesh, N. P., E. Frost, A. Plesch, and J. H., Shaw (2007), Mechanical models of
incremental fault-related folding: Insights into processes of coseismic folding
above blind thrust faults, J. Geophys. Res., 112, B03S04, doi:
10.1029/2006JB004466 .
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
234
Chen, Y. G., K.Y. Lai, Y. H. Lee, J. Suppe, W. S. Chen, Y. N. N. Lin, Y. Wang, J. H.
Hung, and Y. T. Kuo (2007), Coseismic fold scarps and their kinematic behavior
in the 1999 Chi-Chi earthquake Taiwan, J. Geophys. Res., 112, B03S02, doi:
10.1029/2006JB004388.
Crouch, J. A., and Suppe, J. (1993), Late Cenozoic tectonic evolution of the Los Angeles
basin and Inner California Borderland: A model for core-complex-like crustal
extension, Geol. Soc. Am. Bull., v. 105, 1415-1434.
Dolan, J. F., S. Christofferson, and J. H. Shaw (2003), Recognition of paleoearthquakes
on the Puente Hills blind thrust fault, Los Angeles, California, Science, 300, 115-
118.
Erslev, E. A. (1991), Trishear fault-propagation folding, Geology, 19, 617-620.
Field, E. H., H. A. Seligson, N. Gupta, V. Gupta, T. H. Jordan, and K. W. Campbell
(2005), Loss estimates for a Puente Hills blind-thrust earthquake in Los Angeles,
California, Earthquake Spectra, 21, 329-338.
Field., E. H. (2005), Collaborative SCEC/USGS efforts to improve seismic-hazard analysis;
RELM and OpenSHA: Proceedings of the 5th U.S.-Japan natural resources meeting and
Parkfield, California fieldtrip, Open-File Report - U. S. Geological Survey, Report: OF
2005-1131, pp. 18. Resource Location: http://pubs.usgs.gov/of/2005/1131/
Gumprecht, B. (2001), The Los Angeles River: It’s life, death, and possible rebirth,
Baltimore and London: John Hopkins University Press.
Hauksson, E. (1990), Earthquakes, faulting, and stress in the Los Angeles Basin, J. Geophys.
Res., 95, 15,365-15,394.
Hauksson, E., and Jones, L. (1989),The 1987 Whittier Narrows earthquake sequence in Los
Angeles, southern California: Seismological and tectonics analysis: J. Geophys. Res.,
94, 9569.
Heaton, T. H., J. F. Hall, D. J. Wald, and M. W. Halling (1995), Response of high-rise
and base-isolated buildings to a hypothetical Mw 7.0 blind thrust earthquake,
Science, 267, 206-211.
Hudnut, K. W., Shen, Z., Murray, M., McClusky, S., King, R., Herring, T., Hager, B., Feng,
Y., Fang, P., Donnellan, A., and Bock, Y.(1996), Co-seismic displacements of the 1994
Northridge, California, earthquake: Bull. Seismol. Soc. Am., v. 86, No. 1B, p. S19 – S36.
235
Ishiyama, T., K. Mueller, M. Togo, A. Okada, and K. Takemura (2004), Geomorphology,
kinematic history, and earthquake behavior of the active Kuwana wedge thrust
anticline, central Japan, J. Geophys. Res., 109, B12408, doi:
10.1029/2003JB002547.
Ishiyama, T., K. Mueller, H. Sato, and M. Togo (2007), Coseismic fault-related fold
model, growth structure, and the historic multisegment blind thrust earthquake on
the basement-involved Yoro thrust, central Japan, J. Geophys. Res., 112, B03S07,
doi: 10.1029/2006JB004377.
King, G. C. P., and C. Vita-Finzi (1981), Active folding in the Algerian earthquake of 10
October 1980, Nature, 292, 22-26.
Lai, K., Y. Chen, J. Hung, J. Suppe, L. Yue, and Y. Chen (2006), Surface deformation
related to kink-folding above an active fault: Evidence from geomorphic features
and co-seismic slips, Quat. Int., 147, 44-54.
Lehle, D. (2007), Geometry and slip history of the Compton thrust fault: Implications for
earthquake hazard assessment, thesis, Harvard Univ., Cambridge, Mass.
Leon, L. A., J. F. Dolan, S. A. Christofferson, J. H., Shaw, and T. L. Pratt (2007),
Earthquake-by-earthquake fold growth above the Puente Hills blind thrust fault,
Los Angeles, California: Implications for fold kinematics and seismic hazard, J.
Geophys. Res., 112, B03S03, doi: 10.1029/2006JB004461.
Leon, L. A., J. F. Dolan, J. H., Shaw, and T. L. Pratt (2009), Evidence for large Holocene
earthquakes on the Compton thrust fault, Los Angeles, California, J. Geophys.
Res., in press, doi: 10.1029/2008JB006129.
Liberty, L. M., and T. L. Pratt (2008), Structure of the eastern Seattle fault zone,
Washington state: New insights from seismic reflection data, Bull. Seismol. Soc.
Am., 98, 4, 1681-1695, doi: 10.1785/0120070145.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Res., 94, 9614-9632.
Linton, J. (2005), Down by the Los Angeles River, Wilderness Press, Berkeley,
California, 5-7.
Mueller, K. J., (1997), Recency of folding along the Compton-Los Alamitos trend:
Implications for seismic risk in the Los Angeles basin, EOS Trans. AGU, 78,
F702.
236
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Plesch, A., and J. H. Shaw (2007), Community fault model (CFM) for Southern
California, Bull. Seismol. Soc. Am., 97, 6. 1793-1802.
Pratt, T. L., J. H. Shaw, J. F. Dolan, S. A. Christofferson, R. A. Williams, J. K. Odum,
and A. Plesch (2002), Shallow folding imaged above the Puente Hills blind-thrust
fault, Los Angeles, California, Geophys. Res. Lett., 29, 18-1 - 18-4.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Roberston, P.K. (1990), Soil Classification using the Cone Penetration Test, Canadian
Geotech. Journ., v. 27, 151-158.
Robertson, P.K., Campanella, R. G., Gillespie, D., and Rice, A. (1986), Seismic CPT to
Measure In-Situ Shear Wave Velocity, J. Geotech. Eng. ASCE, v. 112, No. 8,
791-803.
Scientists of the U. S. Geological Survey and the Southern California Earthquake Center
(1994), The magnitude 6.7 Northridge, California, earthquake of 17 January 1994,
Science, 266, 389-397.
Shaw, J. H., Connors, C., and Suppe, J. (2005), Seismic interpretation of contractional
fault-related folds, An Am. Assoc. Petr. Geol. Seismic Atlas; studies in Geology
#53, AAPG, Tulsa, OK.
Shaw, J. H., A. Plesch, J. F. Dolan, T. Pratt, and P. Fiore (2002), Puente Hills blind-thrust
system, Los Angeles basin, California, Bull. Seismol. Soc. Am., 92, 2946-2960.
Shaw, J. H., and Shearer, P. M. (1999), An elusive blind-thrust fault beneath metropolitan
Los Angeles: Science, v. 283, p. 1516-1518.
Shaw, J. H., and Suppe, J. (1994), Active faulting and growth folding in the eastern Santa
Barbara Channel, California: Geol. Soc. Am. Bull., v. 106, 607-626.
Shaw, J. H., and Suppe, J. (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California: J. Geophys. Res., 101(B4), 8623 - 8642.
237
Sommerville, P. G., and A. Pitarka (2006), Differences in earthquake source and ground
motion characteristics between surface and buried earthquakes, Proceedings of
the Eighth National Conference on Earthquake Engineering, San Francisco,
California, U.S.A., 977, Earthquake Engineering Research Institute.
Stein, R. S., and G. King (1984), Seismic potential revealed by surface folding; 1983
Coalinga, California, earthquake, Science, 224, 869-872.
Stein, R. S. and Ekström, G. (1992), Seismicity and geometry of a 110-km-long blind
thrust fault. 2. Synthesis of the 1982 – 1985 California earthquake sequence: J.
Geophys. Res., 97(B4), 4865 - 4883.
Stein, R. S., and R. S. Yeats (1989), Hidden earthquakes, Sci. Am., 260, 48-57.
Stephenson, W. J., Odum, J. Shedlock, K. M., Pratt, T. L., and Williams, R. A., Mini-
sosie high-resolution seismic method aids hazards studies, Eos Trans AGU, 73,
473-476, 1992.
Streig, A. R., C. M. Rubin, W. Chen, Y. Chen, L. Lee, S. C. Thompson, C. Madden, and
S. Lu (2007), Evidence for prehistoric coseismic folding along the Tsaotun
segment of the Chelungpu fault near Nan-Tou, Taiwan, J. Geophys. Res., 112,
B03S06, doi: 10.1029/2006JB004403.
Sugiyama, Y., K. Mizuno, F. Nanayama, T. Sugai, H. Yokota, T. Hosoya, K. Miura, K.
Takemura, and N. Kitada (2003), Study of blind thrust faults underlying Tokyo
and Osaka urban areas using a combination of high-resolution seismic reflection
profiling and continuous coring, Annals of Geophys., 46, 5.
Suppe, J. (1983), Geometry and kinematics of fault-bend folding: Am. J. Sci., 283, p.
684-721.
Suppe, J., Chou, G. T. and Hook, S. C. (1992), Rates of folding and faulting determined
from growth strata, Thrust tectonics 105-121 Chapman & Hall, London, United
Kingdom.
Suppe, J. and Connors, C. D. (2004), Shear fault-bend folding, Thrust tectonics and
hydrocarbon systems, AAPG Memoir, 82, 303-323.
Suppe, J., and Medwedeff, D. A. (1990), Geometry and kinematics of fault-propagation
folding: Ecologae Geol. Helv., 83, no. 3, p. 409-454.
238
Suppe, J., Sàbat, F., Muñoz, J. A., Poblet, J., Roca, E., and Vergés, J. (1997), Bed-by-bed
fold growth by kink-band migration: Sant Llorenc de Morunys, eastern Pyrenees,
J. Structural Geol., 19, 443-461.
Süss, M. P., and J. H. Shaw (2003), P wave seismic velocity structure derived from sonic
logs and industry reflection data in the Los Angeles basin, California, J. Geophys.
Res., 108(B3), 2170, doi:10.1029/2001JB001628.
Wald, D. J., T. H. Heaton, and K. W. Hudnut (1996), The slip history of the 1994
Northridge, California, earthquake determined from strong-motion, teleseismic,
GPS, and leveling data Bull. Seismo.l Soc. Am., 86, 1B, S49-S70.
Walls, C., Rockwell, T., Mueller, K., Bock, Y., Williams, S., Pfanner, J., Dolan, J., and
Peng, F. (1998), Escape tectonics in the Los Angeles metropolitan region and
implications for seismic risk: Nature, 394, 356-360.
Wells, D. L., and K. J. Coppersmith (1994), New empirical relationships among
magnitude, rupture length, rupture width, rupture area, and surface displacement,
Bull. Seismol. Soc. Am., 84, 974-1002.
Wright, T. L. (1991), Structural geology and tectonic evolution of the Los Angeles basin,
California, in Active Margin Basins, edited by K. T. Biddle, AAPG Mem., 52, 35-
134.
Yeats, R. S., and Beall, J. M. (1991), Stratigraphic controls of oil fields in the Los
Angeles basin: A guide to migration history, in Active Margin Basins, edited by
K. T. Biddle, AAPG Bull. Mem., 52, 221-235.
Yerkes, R. F., McCulloh, T. H., Schoellhamer, J. E., and Vedder, J. G. (1965), Geology
of the Los Angeles basin, California – An introduction, U.S. Geol. Surv. Prof.
Pap., 420-A, 57.
239
APPENDICES
240
APPENDIX A:
Supplemental Material for Chapter 2
A1 Introduction
This data set contains error analysis for collection and processing of data included
in Chapter 2.
A2 Error Analysis
The most significant error for our hollow-stem excavations was loss of core
within any particular drive, leading to less than 100% recovery. Potential error arises
where an unrecovered interval coincides with the location of a stratigraphic contact,
resulting in an uncertainty in the exact depth of the contact. Errors were reduced by short
sampling drives of 60 cm for the Carfax Avenue transect, and 76 cm in the Gardenland
transect. These short sampling drives allowed us to isolate intervals of no recovery to the
depth interval of each particular drive. Moreover, recovery in both transects was very
high in most boreholes. Recovery in each borehole is shown on figure A1. Average
recovery in the Carfax transect was ~90%, reduced to 50% locally in drives of poor
recovery, particularly of the lower ~20 – 50 m of water-saturated sand and gravel
intervals. The depth error of each contact in intervals of good recovery (>90%) is
therefore + 15 cm, with +30 cm errors in areas of poor recovery. Depth was directly
measured by the drilling team by dropping a weighted tape measure down hole following
each drive (60 - 76 cm). An additional check is made every 1.52 m when a new section
of auger is added to the drill stem. In the Gardenland transect, recovery averaged ~86%,
with poor recovery in borehole 100A (65%) due to longer drives (1.52 m). The main
errors present for bucket auger recovery in the Carfax and SCE transects were in
241
estimation of penetration depth. This was also checked by dropping a weighted tape
measure down the hole, which could then be observed directly by a geologist to the
shallowest aquifer.
A3 High-Resolution Seismic Reflection Depth Conversion Errors
On the mini-sosie seismic reflection profiles, a 17o dip was measured on the
prominent reflector at 200 to 250 m depth. To change this deeper reflector to a 23o dip
(as measured in the deeper-penetration petroleum industry seismic reflection data) would
require a 33% increase in the velocities used to depth-convert the mini-sosie data. In
other words, the velocities would need to be in error by 33% if the dip on the mini-sosie
profile is actually the same as the dip of the reflector on the deeper industry data. This
error could be in either set of seismic reflection data, or a combination of both. Based on
our analysis of reflection travel times used in our depth conversions, however, we think
that our velocity estimates are correct to within about 10%. Therefore, as the mini-sosie
data definitely has less than a 23o dip, there must be significant up dip shallowing of dip
within the kink band.
242
Figure A1 Borehole Results from the Carfax Avenue Transect, Gardenland-
Greenhurst Transect
Borehole results from the Carfax Avenue transect, Gardenland-Greenhurst
transect, and SCE transect, showing major stratigraphic units together with
borehole lithology, core recovery, and sediment color (8x vertical exaggeration;
Figure 2.5). Numbers show key stratigraphic units discussed in the text. Thin red
lines mark the tops of major sand- and gravel-filled channels. Double-headed red
vertical arrows along the left side of the figure show the stratigraphic ranges over
which discrete uplift events occurred (see text for discussion). Solid green lines
between red arrows on left side of figure show intervals that do not change thickness
across the transect. Proposed correlations between Carfax boreholes and a water
well (1589S) located 175 m north of the transect are shown by dashed stratigraphic
contacts along the right side of the figure. Contacts at 12, 17, and 20 m depth are
correlative between borehole 22 and the water well.
243
Figure A1, Continued
244
Figure A2 An Example of the Inclined Shear Restoration Technique
An example of the inclined-shear restoration technique used in this study (based on
the methods described by Novoa et al. [2000]). (A) Sequential reconstruction of
fold growth during uplift Event Y in the Gardenland Avenue/Greenhurst Street
Transect (Figure 2.5A), showing present-day configuration of major stratigraphic
units (8x vertical exaggeration). Red line represents the modern stream gradient as
a proxy for the paleo-stream gradient. This provides a restoration horizon for the
top on Unit 20 to restore folding in the most recent uplift event (Y). Dark-blue,
inclined lines represent restoration vectors based on the angle of the anticlinal,
active axial surface determined from petroleum industry seismic reflection data by
Shaw et al., (2002). Green lines represent tops of major stratigraphic units prior to
restoration and light blue lines represent the tops of units following restoration of
event Y. (B) Configuration of major stratigraphic units following reconstruction of
event Y.
245
Appendix A References
Novoa, E., Suppe, J., and Shaw, J. H. (2000), Inclined-shear restoration of growth folds:
Am. Assoc. Petr. Geol. Bull., V. 84, No. 6, 787 – 804.
Shaw, J. H., Plesch, A., Dolan, J. F., Pratt, T., and Fiore, P. (2002), Puente Hills blind-
thrust system, Los Angeles basin, California: Bull. Seismol. Soc. Am., v. 92, p.
2946-2960.
246
APPENDIX B:
Supplemental Material for Chapter 6
B1 Introduction
This data set contains additional information on radiocarbon dating procedures
and error analysis for collection and processing of the data that are included in the paper.
Additional figures include a cross-section with superimposed borehole lithology, Figure
S1, and additional reconstruction of Event 4, Figure S2. An additional table, Table S1, of
calibrated calendric dates and radiocarbon ages of samples from Stanford Avenue
transect is also included. Items in this supplementary text appear in the same order as
they are discussed in the main text.
B2 Radiocarbon dating
All radiocarbon samples were prepared using a standard “acid-base-acid”
pretreatment of 1 normal (N) solution (containing 1 ‘gram equivalent weight’ of solute
per liter of solution) of hydrochloric acid (HCl), 1 N sodium hydroxide (NaOH), and 1 N
hydrochloric acid (HCl). The first acid soak of HCl removed authigenic carbonates,
fulvics, and acid-soluble organics. The NaOH removed base-soluble organics ("humic
acids") and was repeated until the supernatant was clear. The second series of acid soaks
removed any CO
3
that might have attached to the sample and any base-activated organics
that were not intrinsic to the parent material. The humic fraction was, therefore, removed
during pretreatment of bulk-soil samples. Thus, regardless of relative proportions of the
remaining fractions, the bulk-soil ages used in this paper can be used as robust maximum
ages. In addition to these standard bulk-soil age determinations, we dated several of the
humic fractions removed during the NaOH soak. The humic dates were designed to
247
provide a complementary estimate of the actual age of deposition, as this more-mobile
type of carbon can sometimes reflect the youngest organic carbon deposited in the unit.
As can be seen from examination of main text Figure 5, however, this common
simplifying interpretation of the significance of the humic dates does not appear to work
well in the Stanford Avenue cores, as several of the humic dates are actually older than
the standard bulk-soil dates from the same sample. We therefore place more emphasis on
the more-robust, easily interpreted bulk-soil and charcoal dates in generating the
sediment accumulation rate curves and our event ages.
B3 Potential borehole depth errors
The most significant potential error for our hollow-stem borehole transect was loss of
core within any particular drive, leading to less than 100% recovery. Potential error
arises where an unrecovered interval coincides with the location of a stratigraphic
contact, resulting in an uncertainty in the exact depth of the contact. Errors were reduced
by short sampling drives of 45 cm and 60 cm in the Stanford Avenue transect. These
short sampling drives allowed us to isolate the uncertainty associated with intervals of no
recovery to the depth range of a single drive. Moreover, recovery in this transect was
very high. Average recovery in the Stanford transect was ~93%, and approached 100%
in most cohesive intervals. Locally, however, particularly in the sand and gravel intervals
in the 25-35 m depth interval of the boreholes, recovery was only ~75%. Recovery in
each borehole is shown on Supplementary Figure 1. The depth error of each contact in
intervals of good recovery (>95%) is therefore + 15 cm, with +30 cm potential errors in
areas of poor recovery. During core recovery depth was measured directly by the drilling
248
team by dropping a weighted tape measure down the hole every 1.52 m. An additional
check was made by ensuring that whenever a new 1.52 m section of auger was added to
the drill stem that the bottom of the new auger coincided exactly with the ground surface.
249
Figure B1 Borehole results from the Stanford Avenue transect (Chapter 5)
Cross section of major stratigraphic units (8x vertical exaggeration) with borehole
logs. Green horizontal line represents ground surface. Colors denote different
sedimentary units. Thin red lines mark the tops of major sand- and gravel-filled
channels. Boreholes are shown with various colors that represent both grain size
and Munsell soil colors. Even numbers represent friable units with major sand
units numbered in units of ten (10, 20, 30, 40, 50, 60). Other key sand units are
represented by even numbers (8, 16, 42). Odd numbers represent cohesive units,
with white areas representing undifferentiated cohesive units. Organic-rich soils
are numbered based on the overlying major sand unit plus one, (e.g., 11, 21, 31, 41,
51). Red, double-headed vertical arrows along the right side of the figure show the
stratigraphic range of intervals of post-uplift sedimentary growth (thickening)
across the transect. Green, double-headed vertical arrows denote periods of no
sedimentary growth. Minimum uplift is noted for each event.
250
Figure B1, Continued
251
Figure B2 Reconstruction of Event 4 (Chapter 6).
Additional graphs showing growth intervals and scarp heights for the restoration of
Event 4. Unit 42 was not present in borehole 9, leading to ambiguity in the exact
depth of the event horizon. (A) Event 3 has been restored to the paleostream
gradient. Event 4 is represented by thickening of the cohesive interval between Units
20 and 30, with a north-facing scarp at the top of Unit 30. (B) Thickness of cohesive
unit between Units 20 and 30 (blue line) versus scarp height at top of Unit 30 (pink
line). (C) Thickness of sedimentary units from top of Unit 30 to top of Unit 42 (blue
line) versus scarp height at top of Unit 40 (pink line). (D) Thickness of sedimentary
units from top of Unit 30 to top of Unit 40 (blue line) versus scarp height at top of
Unit 40 (pink line). (E) Thickness of sedimentary units from top of Unit 40 to top of
Unit 42 (blue line) versus scarp height at top of Unit 40 (pink line). Parts C – E
demonstrate a lack of sedimentary growth (blue line) in all units other than cohesive
interval between Units 20 and 30, suggesting that the event horizon for Event 4 lies
below unit 20 and above Unit 30.
252
Figure B2, Continued
253
Appendix B References
Bronk Ramsey, C. (1995), Radiocarbon Calibration and Analysis of Stratigraphy: The
OxCal Program, Radiocarbon, 37, 2, 425-430.
Bronk Ramsey, C. (2001), Development of the Radiocarbon Program OxCal,
Radiocarbon, 43, 2A, 355-363.
Bronk Ramsey, C., J. van der Plicht, and B. Weninger (2001), 'Wiggle Matching'
radiocarbon dates, Radiocarbon, 43, 2A, 381-389 2001.
Lin, J., and R. S. Stein (1989), Coseismic folding, earthquake recurrence, and the 1987
source mechanism at Whittier Narrows, Los Angeles basin, California, J.
Geophys. Research, 94, 9614-9632.
Novoa, E., J. Suppe, and J. H. Shaw (2000), Inclined-shear restoration of growth folds,
AAPG Bull., 84, 787-804.
Reimer, P. J., and 27 others (2004), Radiocarbon calibration from 0-26 cal kyr BP –
IntCal04 terrestrial radiocarbon age calibration, 0-26 kyrBp, Radiocarbon, 46,
1029-1058.
Shaw, J. H., A. Plesch, J. F. Dolan, T. L. Pratt, and P. Fiore (2002), Puente Hills blind-
thrust system, Los Angeles basin, California, Bull. Seismol. Soc. Am., 92, 2946-
2960.
Shaw, J. H., and J. Suppe (1996), Earthquake hazards of active blind-thrust faults under
the central Los Angeles Basin, California, J. Geophys. Res., 101, 8623-8642.
Wells, D. L., and K. J. Coppersmith (1994), New Empirical Relationships among
magnitude, Rupture Length, Rupture Width, Rupture Area, and Surface
Displacement, Bull. Seismol. Soc. Am., 84, 974-1002.
Abstract (if available)
Abstract
In order to understand the paleo-earthquake history and structural evolution of blind-thrust faults and their associated folds, I use a multi-disciplinary methodology to link blind faulting at seismogenic depths directly to near-surface fault-related folding. My research focused on two major blind-thrust systems beneath metropolitan Los Angeles, California
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The impact of faulting and folding on earthquake cycles in collisional orogens
PDF
Paleoseismologic and slip rate studies of three major faults in southern California: understanding the complex behavior of plate boundary fault systems over millenial timescales
PDF
Characterizing the recent behavior of the Ventura blind thrust fault, Brookshire Avenue study site: implications for mutifault ruptures in the western Transverse Ranges of southern California
PDF
Towards an understanding of fault-system mechanics: from single earthquakes on isolated faults to millenial-scale collective plate-boundary fault-system behavior
PDF
Characterizing fault behavior and earthquake surface expression on timescales of single events to multiple millennia
PDF
Direct observation of fault zone structure and mechanics in three dimensions: a study of the SEMP fault system, Austria
PDF
Growth of the Cretaceous Tuolumne batholith and synchronous regional tectonics, Sierra Nevada, CA: a coupled system in a continental margin arc setting
Asset Metadata
Creator
Leon, Lorraine Annette
(author)
Core Title
Paleoseismology of blind-thrust faults beneath Los Angeles, California: implications for the potential of system-wide earthquakes to occur in an active fold-and-thrust belt
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Geological Sciences
Degree Conferral Date
2009-08
Publication Date
08/06/2009
Defense Date
05/04/2009
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Compton thrust fault,OAI-PMH Harvest,paleoseismology of blind-thrust faults,Puente Hills thrust fault
Place Name
California
(states),
Los Angeles
(city or populated place)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Dolan, James F. (
committee chair
), Davis, Gregory A. (
committee member
), Ershaghi, Iraj (
committee member
), Lund, Steven P. (
committee member
)
Creator Email
laztrak@aol.com,lorrainl@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m2497
Unique identifier
UC1177882
Identifier
etd-Leon-2969 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-184840 (legacy record id),usctheses-m2497 (legacy record id)
Legacy Identifier
etd-Leon-2969.pdf
Dmrecord
184840
Document Type
Dissertation
Rights
Leon, Lorraine Annette
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
Compton thrust fault
paleoseismology of blind-thrust faults
Puente Hills thrust fault