University of Southern California Dissertations and Theses

Revealing spatiotemporal patterns in fault-system behavior: insights from multi-millennial earthquake clustering and slip rate variability on the Garlock fault

(USC Thesis Other)

Revealing spatiotemporal patterns in fault-system behavior: insights from multi-millennial earthquake clustering and slip rate variability on the Garlock fault

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content REVEALING SPATIOTEMPORAL PATTERNS IN FAULT-SYSTEM BEHAVIOR: INSIGHTS
FROM MULTI-MILLENNIAL EARTHQUAKE CLUSTERING AND SLIP RATE VARIABILITY ON
THE GARLOCK FAULT
by
Dannielle Fougere
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)
May 2025
Copyright 2025 Dannielle Fougere

ii
ACKNOWLEDGEMENTS
This journey would not have been possible without the guidance, support, and encouragement of many
people who have played a crucial role in my academic and personal growth over the past several years.
First and foremost, I am grateful to my advisor, James Dolan, whose endless enthusiasm for scientific
research has been inspiring and motivating. Thank you for offering me the opportunity to pursue this PhD
at the University of Southern California, and for guiding me to where I am today.
I moved to the United States in August 2019 alongside two other international students, Judith Gauriau
and Chris Anthonissen, where we started our new lives in Los Angeles. I want to thank both of you for
your unwavering support in the office and the field. I would not have been able to complete this PhD
without either of you. I also want to thank lab mates who joined a little later on, PhD student Caje
Weigandt and MS student Luke Gordon, whose help during fieldwork in the Mojave Desert was
invaluable. I also extend my gratitude to my collaborators, Sally McGill, Ed Rhodes, and Andrew Ivester,
thank you for your kindness and unwavering support throughout this process.
To my qualifying committee members, Josh West, Sylvain Barbot, Craig Stanford, and John Vidale,
thank you for your guidance and thoughtful feedback. I am also deeply appreciative of Steve Nutt, my
external defense committee member from the Viterbi School of Engineering, for kindly agreeing to serve
on my defense committee last minute. I would like to thank the incredible staff at USC who have made
this journey smoother: Cindy Waite, Darlene Garza, Karen Young, Alexandra Aloia, Miguel Rincon,
John Yu, and Steve Lin. Special thanks to Darlene for organizing my PhD defense and ensuring I stayed
on track throughout this complex process!
To my family, despite a whole ocean between us, your unwavering support has been my anchor. From a
young age, you opened my eyes to the world, instilling in me the confidence to move to a different
country and build a life here. Your encouragement and support have been constant reminders of my roots.
To the Chandenburgs and Jan Goldhaber, thank you for providing me with the most incredible

iii
community and family here in Los Angeles. I will be forever grateful for the warmth and support you
have shown me. Anahi Carrera, thank you for the experiences we have had in academia and beyond. Our
conversations and shared moments have profoundly shaped who I am today, and I can’t imagine how
different this journey would have been without you. Lastly, to my friends, thank you for being a breath of
fresh air in my life. Your support, humor, and willingness to listen to my endless chatter about
earthquakes and rocks has made this journey not only bearable but enjoyable. You’ve reminded me of the
importance of life beyond research, and for that, I am deeply thankful.
This dissertation includes parts of the following manuscripts:
Fougere, D., Dolan, J., Rhodes, E., & McGill, S. (2024). Refined Holocene slip rate for the
western and central segments of the Garlock fault: record of alternating millennial-scale periods of fast
and flow fault slip. Seismica, 3(2). DOI: 10.26443/seismica.v3i2.1202
Fougere, D., Dolan, J., Ivester, A., McGill S., Rhodes, E., Anthonissen, C., and Gauriau, J. (in
prep.). The Garlock Fault Paleoseismic Record at Koehn Lake: Implications for Multi-Millennial
Earthquake Clustering and Fault Interactions.
Fougere, D., Dolan, J., Rhodes, E., and McGill S. (in prep.). Combined Incremental Slip Rate and
Paleo-Earthquake Records: An Earthquake-by-Earthquake Record of Slip for the Garlock Fault.

iv
TABLE OF CONTENTS
ACKNOWLEDGEMENTS...........................................................................................................................ii
LIST OF TABLES ......................................................................................................................................vii
LIST OF FIGURES ...................................................................................................................................viii
ABSTRACT ................................................................................................................................................xii
CHAPTER 1 Introduction .........................................................................................................................1
1.1 The Southern California Plate Boundary ............................................................................................2
1.2 The Garlock Fault ...............................................................................................................................3
1.3 Methods................................................................................................................................................5
1.3.1 Light Detection and Ranging Data Analysis................................................................................5
1.3.2 Paleoseismic Trenching ...............................................................................................................6
1.3.3 Geochronological Techniques .....................................................................................................9
1.3.3.1 Luminescence Dating ..........................................................................................................9
1.3.3.2 Radiocarbon dating ............................................................................................................10
1.4 Research Implications ......................................................................................................................12
CHAPTER 2 Refined Holocene Slip Rate for the Western and Central Segments of the
Garlock Fault: Record of Alternating Millennial-Scale Periods of Fast and Slow Fault Slip ....................15
2.1 Abstract .............................................................................................................................................15
2.2 Introduction ......................................................................................................................................15
2.3 The Garlock Fault .............................................................................................................................17
2.4 Previous Studies ................................................................................................................................17
2.5 New Slip Rate Data from the Central and Western Garlock Fault ..................................................20
2.5.1 Summit Range East ....................................................................................................................20
2.5.1.1 Offset Measurements .........................................................................................................21
2.5.1.2 Age Constraints ..................................................................................................................22
2.5.1.3 Slip Rate Calculation .........................................................................................................24
2.5.2 Summit Range West ..................................................................................................................25
2.5.2.1 Offset Measurements .........................................................................................................25
2.5.2.2 Age Constraints ..................................................................................................................26
2.5.2.3 Slip Rate Calculation .........................................................................................................27
2.5.3 Clark Wash ................................................................................................................................27
2.5.3.1 Offset Measurements .........................................................................................................28
2.5.3.2 Age Constraints ..................................................................................................................29
2.5.3.3 Slip Rate Calculation .........................................................................................................31
2.6 Discussion .........................................................................................................................................32
2.6.1 Multi-Millennial Slip Rate Variations .......................................................................................32

v
2.6.2 Comparison of Geodetic Slip-Deficit Rate with Incremental Geologic Slip Rates ...................35
2.6.3 Evaluation of Driving Mechanisms for Garlock Fault Slip .......................................................36
2.6.4 Implications for Probabilistic Seismic Hazard Assessment ......................................................37
2.6 Conclusions .......................................................................................................................................38
CHAPTER 3 The Garlock Fault Paleoseismic Record at Koehn Lake: Implications for
Multi-Millennial Earthquake Clustering and Fault Interactions..................................................................40
3.1 Abstract .............................................................................................................................................40
3.2 Introduction ......................................................................................................................................41
3.3 The Garlock Fault .............................................................................................................................42
3.4 The Koehn Lake Paleoseismic Site ...................................................................................................44
3.5 Structure and stratigraphy of the trench exposures .........................................................................46
3.5.1 Southern section stratigraphy ...................................................................................................47
3.5.2 Central section stratigraphy ......................................................................................................48
3.5.3 Northern section stratigraphy ...................................................................................................49
3.6 Event Evidence ..................................................................................................................................49
3.7 Age Control........................................................................................................................................56
3.7.1 Luminescence Ages ...................................................................................................................57
3.7.2 Radiocarbon dating ....................................................................................................................57
3.8 Koehn Lake paleo-earthquake chronology .......................................................................................58
3.9 Generation of preferred central Garlock fault paleo-earthquake age model ..................................58
3.10 Discussion .......................................................................................................................................59
3.10.1 Holocene paleo-earthquake recurrence pattern on the central Garlock fault ..........................61
3.10.2 Comparison of previous Garlock fault paleo-earthquake records ...........................................62
3.10.3 Possible physical controls on clustering behavior in southern California ...............................64
3.11 Conclusions .....................................................................................................................................66
CHAPTER 4 Paleo-Earthquake Evidence at Straw Peak Road from the Central Garlock Fault ...........68
4.1 Introduction ......................................................................................................................................68
4.2 Straw peak road paleoseismic site ....................................................................................................68
4.3 Paleoseismic trenching and trench stratigraphy ..............................................................................69
4.4 Interpretation of paleo-surface ruptures...........................................................................................71
4.5 Future Work ......................................................................................................................................73
CHAPTER 5 Combined Incremental Slip Rate and Paleo-Earthquake Records:
An Earthquake-by-Earthquake Record of Slip for the Garlock Fault ................................................74
5.1 Abstract .............................................................................................................................................74
5.2 Introduction ......................................................................................................................................75
5.3 New slip rate data from the central Garlock fault ............................................................................76
5.3.1 Koehn Lake berm ......................................................................................................................76
5.3.2 Searles Lake shoreline ...............................................................................................................78

vi
5.3.3 Pilot Knob Valley west ..............................................................................................................80
5.3.3.1 PKVw-A ............................................................................................................................81
5.3.3.2 PKVw-C ............................................................................................................................82
5.3.3.3 PKVw-E .............................................................................................................................83
5.3.4 Pilot Knob Valley east ....................................................................................................84
5.4 Updated incremental slip rate record for the Garlock fault .............................................................87
5.5 Discussion .........................................................................................................................................88
5.5.1 Correspondence of earthquake clusters with periods of faster-than-average slip .....................88
5.5.2 How much slip occurs per cluster? ............................................................................................89
5.5.3 Spatial observations ...................................................................................................................90
5.5.4 Fault interactions .......................................................................................................................91
5.5.5 Driving mechanisms ..................................................................................................................92
5.5.6 Probabilistic seismic hazard analysis .........................................................................................93
5.6 Conclusions .......................................................................................................................................94
CHAPTER 6 Conclusions .......................................................................................................................95
TABLES ....................................................................................................................................................100
FIGURES...................................................................................................................................................124
REFERENCES ..........................................................................................................................................152

vii
LIST OF TABLES
Table 2.1. Results of single-grain post-IR IRSL dating from Summit Range East (SRE), Summit
Range West (SRW), and Clark Wash (CW). ............................................................................................100
Table 2.2. Results of radiocarbon dating from the Clark Wash site. Calibrations based on IntCal20
calibration curve using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020).......................................102
Table 3.1. Summary of event evidence seen on the Koehn Lake trench exposures.................................103
Table 3.2. IRSL sample age estimates (variable OD) using a fading correction of 12%, modeled
water content that is more representative of actual conditions over the lifetime of burial of the
sediment taking into account modern, measured water content AND a calculated saturated water
content (from bulk density and porosity) for each sample, and overdispersion values of between
15 and 25% are used based on clear signals that indicate modes of well-bleached grains .......................108
Table 3.3. Radiocarbon dates and calibrated ages from the Koehn Lake 2021 paleoseismic trench .......112
Table 3.4. Age constraints on paleoseismic slip events at the Koehn Lake site calculated on
OxCal (V3) ................................................................................................................................................116
Table 5.1. Summary of slip rate data used in the Garlock fault incremental slip rate record...................118

viii
LIST OF FIGURES
Figure 2.1. Map of the Garlock fault and other active faults around the Mojave region of
southern California (gray), including the Mojave section of the San Andreas fault. Yellow star is
the location of the Clark Wash (CW) site of this study, blue star is the location of the Summit
Range West (SRW) and Summit Range East (SRE) sites of this study. White circles show the
locations of past slip rate studies; CCW is the Christmas Canyon West site of Dolan et al. (2016),
KL is the Koehn Lake site of Clark & Lajoie (1974), and PKV is the Pilot Knob Valley site of
Rittase et al. (2014). A gray square shows the location of the El Paso Peaks (EPP) paleoseismic
site of Dawson et al. (2003). Quaternary fault traces sourced from US Geological Survey &
California Geological Survey (2023).........................................................................................................122
Figure 2.2. (a) Annotated lidar-derived hillshade of the Summit Range West (SRW) study site
originally studied by Ganev et al. (2012) at N 35.479116° W 117.560319°, and the Summit Range
East (SRE) at N 35.485943° W 117.530138°. (b) Topographic map (2-m contour interval) derived
from the lidar dataset of the same areas as in (a). Garlock fault trace shown by the red line. ..................123
Figure 2.3. (a) Interpreted lidar-derived hillshade of the Summit Range East (SRE) site. Colors
show various mapped alluvial fan surfaces (Qfa, Qfb, Qfc1, Qfc2, and Qfd) and other mapped
rock units (Tss and Mzg). IRSL sample pit is shown by a blue circle and the fault trace is shown
by red lines. Inset shows a topographic profile marked by a white dotted line (A-A’) across Qfc1
and Qfc2 that suggests that the two fan segments may be the same fan. 38-m back-slipped
restoration of the SRE site shown with (b) interpreted lidar-derived hillshade and (c) lidar-derived
hillshade. The 38-m preferred offset measurement is based on the most plausible configuration of
the channel (Ch-1) incised into the Qfc2 fan.............................................................................................124
Figure 2.4. Age estimates for IRSL samples SRE14-01, SRE14-02, SRE14-03, and SRE14-04 that
were collected from the Qfc2 fan at the Summit Range East site. (a) Cross-section diagram of Qfc2
sample pit showing depths at which samples were collected and general stratigraphy. Sample
SRE14-04 is not included in the final slip rate calculation, (b) Single-grain potassium-feldspar
post-IR infrared stimulated luminescence distribution data and preferred age estimates for each s
ample. (c) Preferred estimate of the age of Qfc2 deposition by combining three shallowest sample
calendar dates with Gaussian errors using a chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009;
Reimer et al., 2020). ..................................................................................................................................125
Figure 2.5. (a) Interpreted lidar-derived hillshade of the Summit Range West (SRW) site. Colors
show various mapped alluvial fan surfaces (Qf1, Qf2, Qf3, and Qf4) and other mapped rock units
(QTc). IRSL sample pit is shown by a blue circle and the fault trace is shown by red lines. 70-m
back-slipped restoration of the SRW site shown with (b) interpreted lidar-derived hillshade and
(c) lidar-derived hillshade. The 70-m preferred offset measurement is based on the restoration of
the incised edge of the Qf2 fan surface on the western margin of the channel .........................................126
Figure 2.6. Age estimates for IRSL samples GF16-05, GF16-06, and GF16-07, which were
collected from the Qf2 fan at the Summit Range West site. (a) Cross-section view of Qf2 sample
pit showing depths at which samples were collected. (b) Single-grain K-feldspar IRSL distribution
data and preferred age estimates for each sample. (c) Preferred estimate of the age of Qf2
deposition by combining all three calendar dates with Gaussian errors using a chi-squared test
using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020)...................................................................127

ix
Figure 2.7. Regional topographic map of the Clark Wash site with a contour interval of 100 m.
The Clark Wash site is located at the northeastern extent of the western Garlock fault segment
adjacent to the 2-3 km wide step over at Koehn Lake at N 35.214608° W 118.075739°. The
black dashed box outlines the location of Figure 2.8a...............................................................................128
Figure 2.8. (a) Interpreted lidar-derived hillshade of the Clark Wash (CW) site. Colors
overlying lidar-derived hillshade show various mapped surfaces (disturbed surface, modern
channel, various aged alluvium, and colluvium). IRSL sample pit is shown by a blue circle and
the fault trace is shown by a red line. Eight of the ten trenches that were excavated by McGill
et al. (2009) fall within the area shown in this figure and are marked by gray boxes. The location
beneath which radiocarbon samples from Hoa were collected is indicated by the asterisk within
one of the trenches. The 66-m back-slipped restoration of the CW site is shown with
(b) interpreted lidar-derived hillshade and (c) lidar-derived hillshade. The 66-m preferred
offset measurement is based on the restoration of the northeastern channel wall.....................................129
Figure 2.9. Age estimates for IRSL samples GF16-01, GF16-02, GF16-03, and GF16-04
collected at the Clark Wash site. (a) Cross-section view of IRSL sample pit showing depths
that samples were collected. GF16-01, GF16-02, and GF16-03 were collected from Qfy,
whereas GF16-04 was collected from Qfo. (b) Single-grain K-feldspar post-IR IRSL distribution
data and preferred age estimates for each sample. (c) Preferred estimate of the age of Qfy
deposition by combining all four IRSL dates with Gaussian errors using a chi-squared test using
OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020). (d) Probability density functions for the
eight radiocarbon charcoal samples ages collected from the Clark Wash site calculated using
OxCal version 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020). Average age of all eight samples
was calculated using the OxCal combine function. (e) Age estimate of initial onset of fault offset
calculated by finding the PDF representing the interval between (c) and (d) ...........................................130
Figure 2.10. (a) Incremental slip history for the central and western Garlock fault including sites:
CCW - Christmas Canyon West slip rate site (Dolan et al., 2016), CW - Clark Wash (this paper,
building on McGill et al., 2009), MRE - most recent earthquake recorded by Dawson et al. (2003),
SR - Slate Range slip rate site (Rittase et al., 2014), SRE - Summit Range East slip rate site (this
paper), SRW - Summit Range West slip rate site (this paper, building on Ganev et al., 2012).
(b) Incremental slip rates for the central Garlock fault calculated using RISeR (Zinke et al., 2017,
2019) using only the most tightly constrained rates from a 16-km-long section of the central part
of the fault..................................................................................................................................................131
Figure 3.1. Map of the Garlock fault (red) and other active faults around the Mojave region of
southern California (gray), including the Mojave section of the San Andreas fault. The yellow
circle is the location of the Koehn Lake paleoseismic site (this paper). White circles show the
locations of past paleoseismic sites along the Garlock fault; CT - Campo Teresa (Gath and
Rockwell, 2018), TL - Twin Lakes (Madden Madugo et al., 2012), CW - Clark Wash (McGill
et al., 2009), EPP - El Paso Peaks (Dawson et al, 2003), CCW - Christmas Canyon West
(McGill et al., in prep), EP - Echo Playa (Kemp et al., 2016), SPR – Straw Peak Road (Chapter
4). The Garlock fault has been divided into three sections, western, central, and eastern, based on
changes in its strike or the presence of structural discontinuities (McGill, 1992) ....................................132
Figure 3.2. (a) Fremont Valley map with geology sourced from Macrostrat (Peters et al., 2018),
(b) Annotated satellite image of the Koehn Lake paleoseismic and slip rate site (Google Earth,
accessed December 2024), and (c) Annotated photograph of the Koehn Lake trench looking
eastwards....................................................................................................................................................133

x
Figure 3.3. Generalized schematic of the Koehn Lake trench stratigraphy and average ages for
the various units within the three sections of the trench............................................................................134
Figure 3.4. Annotated southern trench logs, (A) annotated west wall trench log, (B) annotated
east wall trench log, and (C) relative positions of event horizons (black lines), IRSL samples,
and radiocarbon samples from both walls. ................................................................................................135
Figure 3.5. Annotated northern trench logs, (A) annotated west wall trench log, (B) annotated
east wall trench log, and (C) relative positions of event horizons (black lines), IRSL samples,
and radiocarbon samples from both walls.................................................................................................136
Figure 3.6. Annotated northern trench logs, (A) annotated west wall trench log, (B) annotated
east wall trench log, and (C) relative positions of event horizons (black lines), IRSL samples,
and radiocarbon samples from both walls. This section is dominated by alluvium and highly
channelized with many laterally discontinuous units along the length of this section of the
trench exposure..........................................................................................................................................137
Figure 3.7. Paleo-earthquake age data. (A) Koehn Lake minimum paleo-earthquake record
(V3), (B) El Paso Peaks paleo-earthquake record, and (C) combined central Garlock fault
preferred paleo-earthquake record for the central Garlock fault (V4).......................................................138
Figure 4.1. Annotated Straw Peak Road trench log for the east wall exposure, shown from
m 7 to m 18 ................................................................................................................................................139
Figure 5.1. Map of the Garlock fault (red) and other active faults around the Mojave region
of southern California (gray), including the Mojave section of the San Andreas fault. Squares
along the Garlock fault are slip rate (white squares) and paleoseismic (purple squares) study
sites. Sites that have both slip rate and paleoseismic studies are shown with half white-half
purple squares. CT – Campo Teresa (Gath and Rockwell, 2018), TL – Twin Lakes (Madden
and Dolan, 2008); Madden Madugo et al., 2012), CW – Clark Wash (McGill et al., 2009;
Fougere et al., 2024), KL – Koehn Lake (Clark & Lajoie, 1974; Madden and Dawson, 2006,
this chapter), EPP – El Paso Peaks (Dawson et al., 2003), SRW – Summit Range West
(Ganev et al., 2012; Fougere et al., 2024), SRE – Summit Range East (Fougere et al., 2024),
CCW – Christmas Canyon West (Dolan et al., 2016; Pena, 2019; McGill et al, in prep),
SLS - Searles Lake Shoreline (McGill and Sieh, 1993, this chapter), SPR – Straw Peak Road
(Chapter 4), PKVw – Pilot Knob Valley West (Crane, 2014; Rittase et al., 2014, this chapter),
PKVe – Pilot Knob Valley East (Crane, 2014, this chapter). Quaternary fault traces sourced
from US Geological Survey & California Geological Survey (2023).......................................................140
Figure 5.2. Lidar-derived hillshaded images showing overview maps for the Koehn Lake berm
site, Searles Lake shoreline site, and Pilot Knob Valley ...........................................................................141
Figure 5.3. Koehn Lake berm slip rate site, luminescence (IRSL) sample locations and ages,
and preferred restoration based on the projected and estimated ridge crest ..............................................142
Figure 5.4. Searles Lake Shoreline slip rate site, luminescence (IRSL) sample locations and
ages, and preferred restoration...................................................................................................................143
Figure 5.5. Pilot Knob Valley west slip rate site overview ......................................................................144

xi
Figure 5.6. Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations
and ages, preferred restoration for PKVw-A, and trench log from Rittase et al. (2014)...........................145
Figure 5.7. Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations
and ages, and preferred restoration for PKVw-C ......................................................................................146
Figure 5.8. Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations
and ages, and preferred restoration for PKVw-E.......................................................................................147
Figure 5.9. Pilot Knob Valley east slip rate site and preferred restorations for PKVe-A,
PKVe-B, PKVe-C, and PKVe-D...............................................................................................................148
Figure 5.10. (A) Age-displacement measurements used in incremental slip rate record for
the western and central Garlock fault, (B) updated in incremental slip rate record
calculated using RISer (Zinke et al., 2017; 2019), (C) combined records showing
earthquake-by-earthquake slip, and (D) summary of periods of fault behavior........................................149

xii
ABSTRACT
Documenting the behavior of faults and fault systems is fundamental to understanding the driving
mechanisms that govern their activity. Mechanisms operating on faults and fault systems are complex,
interconnected, and poorly understood. Currently, there are too few comprehensive data sets of paired
incremental fault slip rates and paleo-earthquake ages from fault systems to allow the comparisons
necessary to fully determine how plate-boundary slip is partitioned amongst major faults in time and
space. This dissertation focuses on developing a comprehensive slip rate and paleo-earthquake record
spanning the Holocene for the Garlock fault, a major strike-slip fault within the southern California plate
boundary network that exhibits irregular earthquake recurrence and slip rate through time. Expanding and
resolving the incremental slip rate and paleo-earthquake record for the Garlock fault provides valuable
insights into fault behavior, enhancing our understanding of fault mechanics, potential interactions
between faults, and provides data necessary to facilitate the refinement of next-generation seismic hazard
assessments.
To generate a paired incremental fault slip rate and paleo-earthquake record for the Garlock fault, I use
lidar digital topographic data to measure offset landforms, infrared stimulated luminescence dating
(IRSL), 14C radiocarbon dating, and paleoseismic trenching methods. In this dissertation, I add 12 new
and revised slip rates for the Garlock fault and constrain the timing of 14 paleo-earthquakes on the central
Garlock fault. These results yield one of the longest and most detailed records of combined incremental
fault slip and paleo-earthquakes on any strike-slip fault in the world. These new slip rate data support the
idea that the Garlock fault experiences significant temporal variations (two-to-five fold) in slip rate
spanning multiple earthquake cycles, corresponding with periods of multi-millennial earthquake
clustering punctuated by periods of reduced earthquake recurrence and slower slip rate. New paleoearthquake data from the Koehn Lake trench confirm that earthquake clustering is a common pattern of
earthquake recurrence over the past ~13 ky, with two new temporal four-earthquake clusters documented
at 6-7 ka and 8-10 ka.

xiii
The results of this dissertation provide additional constraints that facilitate a more complete understanding
of fault system phenomena, provide insights into long-term fault behavior, and highlight the role of the
Garlock fault in the plate boundary system.

1
CHAPTER 1 Introduction
Plate boundaries are regions that accommodate the overall deformation between tectonic plates,
comprising a range of deformation styles. An integral component of plate boundaries are faults, zones of
weakness in Earth's lithosphere along which displacement has occurred, that act as the primary structures
accommodating plate motion. Many major faults are situated within complex, mechanically integrated
plate boundary networks that interact to accommodate total plate boundary motions in ways that we do
not yet fully understand. These faults have been shown to exhibit irregular behaviors such as temporal
clustering of earthquakes (e.g., Rockwell et al., 2000; Weldon et al., 2004; Ganev et al., 2010; McAuliffe
et al., 2013; Klinger et al., 2015), multi-millennial temporal slip rate variations (e.g., Wallace, 1987;
Weldon et al., 2004; Dolan et al., 2007; 2016; 2024; Friedrich et al., 2004; Gold & Cowgill, 2011; Ninis
et al., 2013; Onderdonk et al., 2015; Zinke et al., 2017; 2019; 2021; Hatem et al., 2020), and persistent
geologic-geodetic slip rate discrepancies (e.g., Dolan and Meade, 2017; Evans, 2018). The mechanisms
driving these fault behaviors in mechanically complex regions over millennial time scales are poorly
understood, and the degree to which each of these mechanisms operates remains an open question.
Gauriau and Dolan (2021) found that the more complex a fault system that accommodates strain at a plate
boundary, the more irregular the behavior of faults within such networks. Moreover, faults embedded
within complex plate boundaries that exhibit irregular behaviors appear to undergo coordinated switching
of slip between the sub-systems that accommodate plate motion (e.g., Dolan et al., 2007; Hatem and
Dolan, 2018; Dolan et al., 2024). Documenting the behaviors of individual faults is fundamental for
analyzing the collective behavior of a system as a whole, which is, in turn necessary to understand the
convoluted and interconnected driving mechanisms that govern fault activity.
Studying the behavior of faults and fault systems requires a multidisciplinary approach, using highresolution digital topographic data analysis, traditional field mapping, and paleoseismic trenching in
combination with contemporary geochronologic techniques. Currently, there are too few comprehensive
data sets combining paleo-earthquake ages with incremental fault slip rates from regional fault systems to

2
allow the systematic comparisons necessary to fully determine how plate-boundary slip is partitioned
amongst major faults, and how faults interact with one another at a wide range of temporal and spatial
scales. In this dissertation, I explore the basic controls and relative importance of earthquake occurrence
in time and space by developing the most comprehensive paired incremental slip rate and paleoearthquake record yet established for the Garlock fault, a major strike-slip fault within the complex
southern California plate boundary, providing a near earthquake- by-earthquake record of incremental slip
and extending the incremental slip record back to ca. 13 ky. Adding to the literature documenting
irregular behaviors on the Garlock fault (e.g., Dawson et al., 2003; Dolan et al., 2016) will elucidate how
the Garlock fault stores and releases energy through time and space. The Garlock fault incremental slip
record presented in this dissertation serves as a crucial reference for comparing similar data from other
major faults in the region, such as the San Andreas fault and faults of the eastern California shear zone.
1.1 The Southern California Plate Boundary
The Pacific-North American plate boundary is one of the world's most tectonically active and
geologically complex regions. In southern California, most deformation along this boundary has taken
place on two primary plate boundary shear zones, (1) the more localized San Andreas fault (SAF) system,
accommodating ~70% of the total ~50 mm/yr Pacific-North America relative plate motion (DeMets et al.,
1987), and (2) the Eastern California shear zone (ECSZ), which serves as a broad (<100-km-wide) zone
of N-NW-trending dextral shear, accommodating ~20-25% of the total relative plate motion (Dokka and
Travis, 1990a; Savage et al., 1990; Bennett et al., 1999; Dixon et al., 2000; Gan et al., 2000; Miller et al.,
2001; Faulds et al., 2005; Wesnousky, 2005; Faulds and Henry, 2008). The remaining 5-10% of relative
plate boundary motion is accommodated along the western boundary of the Basin and Range province
(Troxel et al., 1972; Davis & Burchfiel, 1973). The SAF is a right-lateral transform fault often split into
three segments, the northern, central, and southern, all of which exhibit different characteristics and
varying degrees of earthquake risk. Particularly germane to this study, the south-central SAF is a ~300-
km-long fault segment that represents a large restraining bend misaligned with regional plate motion by

3
~30°, making it less mechanically efficient in accommodating plate motions. The structurally simplest
part of this section of the San Andreas is the <200-km-long Mojave section of the San Andreas fault
(SAFm).
On the other hand, the ~500-km-long ESCZ, extending from the region of east-west extension bounding
the Sierra Nevada block to its southern extent at the eastern end of the compressive fault-bend of the San
Andreas fault, is oriented much more parallel to the relative plate-motion vector. The ECSZ comprises
two distinct regions, north and south of the Garlock fault, with markedly different topographic
expressions. The northern ECSZ (nECSZ), often called the southern Walker Lane (e.g., Wesnousky,
2005), comprises commonly discontinuous NNW-SSE-trending strike-slip and normal faults forming
multiple pull-apart basins (e.g., Owens Valley, Panamint Valley, Death Valley). The faults of the nECSZ
lie within the transition between the extensional Basin and Range Province and the zone of dextral strikeslip of the ECSZ. As a result of both slip mechanisms operating in this region, the slip contribution from
the nECSZ appears to be superimposed onto the WNW-ENE Basin and Range extension (Atwater and
Stock, 1998). The southern ECSZ (sECSZ), located in the western Mojave Desert, connects the southern
San Andreas fault (SAFs) and extension in the Gulf of California with dextral shear and extension from
the northern ECSZ and southwestern Basin and Range province (Plattner et al., 2010). Faulting in the
sECSZ comprises a series of commonly discontinuous N-NW-trending right-lateral faults accommodating
dextral shear in this region.
1.2 The Garlock Fault
The east-west Garlock fault is one of southern California's most prominent geological features, embedded
sub-perpendicular to the ECSZ and Pacific-North American plate boundary motion. The Garlock fault
spans more than half the width of California, forming an obvious physiographic boundary between the
east-west extension in the Basin and Range to the north and the non-extending Mojave block south of the
fault (Davis & Burchfiel, 1973). This left-lateral strike-slip fault is broadly arcuate over its 250-km length

4
from its intersection with the San Andreas fault at its western extent, to the southern end of Death Valley
at its eastern extent. The active easternmost extent of the Garlock fault is mapped using geophysical and
stratigraphic studies as terminating at the southern end of Death Valley (Jahns and Wright, 1960; Brady
III et al., 1989; Brady III, 1993), however, some authors map its eastern extent farther east (Davis and
Burchfiel, 1973; Plescia and Henyey, 1982).
The east-west Garlock fault is divided into three segments based on changes in strike and structural
complexity. The western Garlock fault segment extends 90 km from its intersection with the San Andreas
fault, eastwards through the Tehachapi Mountains to a two-to-three-km-wide transtensional step-over at
Koehn Lake. The central segment is defined by a 10° change in strike from the western segment,
extending eastwards for 100 km from the step-over at Koehn Lake through the Summit Range and Slate
Range, to a 15° change in strike near the Quail Mountains, which marks the central and eastern segment
boundary (McGill and Sieh, 1991). The eastern segment extends from the eastern end of Pilot Knob
Valley for 60 km to its termination at the southern end of Death Valley. Only the central and eastern
segments of the Garlock fault lie within the zone of dextral shear from the ECSZ, whereas the western
segment lies to the west of this demarcation. The total documented left-lateral from bedrock and structural
displacements along the Garlock fault is 48-64 km (Smith, 1962; Davis and Burchfiel, 1973). Past slip
rate studies along the western and central Garlock fault segments have recorded slip rates of 5-8 mm/yr
averaged over Holocene and latest Pleistocene time (Clark and Lajoie, 1974; McGill et al., 2009; Ganev
et al., 2012), requiring fault initiation to have occurred between 8 Ma and 12.8 Ma for the total 64 km of
slip, corresponding with past studies documenting the initiation of fault slip about 10-11 Ma (Burbank &
Whistler 1987; Loomis & Burbank 1988; Monastero et al., 1997; Frankel et al., 2008).
The tectonic role of the Garlock fault in the North American-Pacific plate boundary remains disputed.
Three models have been proposed to explain the driving mechanisms behind its formation and evolution:
(1) conjugate slip with the northwest-trending, right-lateral San Andreas fault (Hill and Dibblee, 1953),
(2) extension in the Basin and Range province north of the fault (Davis and Burchfiel, 1973), and (3)

5
clockwise rotation driven by dextral shear in the ECSZ (Garfunkel, 1974). Hatem and Dolan (2018)
proposed a model that encompasses all three tectonic drivers, with each mechanism influencing the three
fault segments to varying degrees, with slip rates increasing westward (Clark and Lajoie, 1974; McGill
and Sieh, 1993; McGill et al., 2009; Ganev et al., 2012; Crane, 2014). In this model, the western segment
of the Garlock fault is loaded primarily by lateral extrusion associated with N-S shortening and conjugate
shear in the region of the more westerly-striking section of the San Andreas fault bordering the Mojave
Desert. The central segment of the Garlock fault is loaded by a combination of conjugate shear and
extension in the Basin and Range to the north, as well as by a component of rotation-induced left-lateral
shear within the near-perpendicular zone of north-south ECSZ dextral shear, leading to complicated
patterns of strain accumulation. This model suggests that the western and central segments of the Garlock
fault will accommodate strain at an increased rate during periods when the San Andreas fault is slipping
faster (e.g., Dolan et al., 2007).
1.3 Methods
The geoscience toolkit for incremental slip studies has been greatly improved by advances in
luminescence dating and in the acquisition and processing of geospatial data, such as lidar digital
topographic data, allowing new investigations into fault system behavior. In this dissertation, I take
advantage of these improved techniques by utilizing a combination of geomorphic analyses,
paleoseismology, and geochronological dating techniques (luminescence and radiocarbon) to characterize
the behavior of the western and central Garlock fault in time and space.
1.3.1 Light Detection and Ranging Data Analysis
Over the past 20 years, Light Detection and Ranging (lidar) data have become a crucial tool in many
active tectonics studies, making the investigation of fault processes more accessible. Lidar is a remote
sensing method in which xyz points on the ground surface collected from an aircraft represent the Earth’s
surface in three dimensions, called a point cloud. Point clouds are used to generate detailed digital

6
elevation models (DEMs) of the terrain, as well as contour maps, topographic profiles, and hillshaded
images using GIS software, which allows for detailed analysis of fault-related landforms (e.g., alluvial
fans, terrace risers, river channels). The sub-meter (0.25-0.5 m) resolution of these data allows for precise
measurements of tectonically displaced landforms.
In particular, strike-slip faults can be readily identified using DEM-derived images, even when vegetation
cover masks the fault and tectonic landforms. Laterally extensive linear features are rare in the landscape,
making strike-slip faults identifiable. Several fault offset sites along the Garlock fault were initially
identified by Clark (1973) and later studied in detail using high-resolution (0.25–0.5 m) Airborne Laser
Swath Mapping (ALSM) lidar data. These lidar data, collected in April 2007 as part of the EarthScope
Southern and Eastern California airborne lidar project, are publicly available through OpenTopography
(www.opentopography.org). The lidar data clearly depict the Garlock fault trace, where it exhibits
abundant geomorphic evidence of recent tectonic activity, including numerous fault offsets ranging from
~3 m to over 100 m. Common tectonically offset features in the Mojave Desert include alluvial fans,
beheaded channels, and shutter ridges. Many channels flow across the fault, forming prominent S-shaped
bends after some number of earthquakes. Shutter ridges create irregular topography, whereas sag ponds
develop in areas with a transtensional shear component, making them identifiable in lidar-derived DEM
imagery. Detailed lidar analyses are complemented by extensive geomorphic field mapping to groundtruth observations of displaced landforms along the fault.
1.3.2 Paleoseismic Trenching
Paleoseismic trenching is used in active tectonics to reconstruct the earthquake history of an individual
fault. This technique provides a direct method to study the behavior of a fault that involves the excavation
of a trench (or multiple trenches) across an active fault to expose, map, and analyze subsurface
stratigraphic and structural evidence of past faulting events. Choosing the correct site to excavate a trench
is often difficult in paleo-seismological studies, as specific criteria must be met to achieve a successful

7
study. Firstly, excavation sites are ideally located along a narrow, recognizable fault zone where
deformation is localized. Secondly, the fault needs to be in a site with continuous sediment deposition, at
a rate ensuring that the fault is buried faster than the repeat time of earthquakes, allowing individual event
evidence to be recorded clearly. Sedimentary hiatuses can be present in the stratigraphic record where
sediment accumulation is too slow or erosional processes have removed part of the record, potentially
removing evidence of past earthquakes. Usually, the best sites for measuring earthquake recurrence are
sediment-filled fault-zone depressions. Additionally, adequate dating material to constrain paleoearthquake ages must be present within the trench to develop an earthquake history. The most widely
used geochronologic method in these studies is radiocarbon (14C) dating, however, restricting age dating
methods to only radiocarbon dating limits site selection to the rare sites that include dateable organic
material, such as the Wrightwood site along the central SAF (Weldon et al., 2004), and the Hokuri Creek
site on the Alpine fault (Berryman et al., 2012). However, new advances in luminescence dating
techniques (e.g., Rhodes, 2015) allow sites with sand-rich targets containing little to no organic material
to be dated, placing tighter constraints on the ages of paleo-earthquakes.
Paleoseismic trenching requires careful planning and preparation to ensure safety and effective data
collection. Trenches are typically excavated using tracked hydraulic excavators or backhoes, and the
walls are often benched for added stability and safety. Before logging can begin, the trench walls are
thoroughly cleaned to expose the stratigraphy and structures. Cleaning involves removing 2–5 cm of
material from the walls using various tools. Prior to logging, a reference grid is created using string
arranged horizontally at 50-cm or 1-meter intervals secured with large, labeled nails. Horizontal distances
are established at 1-meter intervals using vertical string lines. Since sedimentary deposits can change
appearance as they dry, making bedding and structures less visible, the walls can be artificially wetted
prior to logging to enhance the visibility of important features. Trenches are often logged manually, but
with technological advances, using an orthorectified photomosaic is becoming a more common practice.

8
Data collection in paleoseismic trenching involves a systematic approach to identifying, marking, and
logging stratigraphic and structural features. Lithologic units are defined as discrete sedimentary deposits
characterized by consistent attributes such as grain size, texture, sorting, bedding, fabric, or color. Subtle
stratigraphic boundaries that lack strong textural or color differences can be more visible in diffuse
lighting than under direct sunlight. Logging involves creating a geologic interpretation of a trench wall,
emphasizing key structural and stratigraphic relationships. The resulting trench log is schematic but
planimetrically accurate, providing a subjective approach to trench logging. This method is efficient and
simplifies interpretation by omitting extraneous details, but it can limit alternative interpretations due to
the selective inclusion of features. To address this limitation, orthomosaics created from high-resolution
imagery provide a detailed, objective record of the trench wall. These ortho-photo mosaics complement
traditional logging methods by preserving fine-scale details and offering a comprehensive view of the
trench wall that exists after the trench has been filled in. Event horizons identified from stratigraphic and
structural relationships are marked using nails and colored flagging tape, and dating samples are collected
above and below an event horizon in order to constrain its age. In my paleoseismic research, I follow a
philosophy of attempting to explain the paleo-earthquake evidence with as few paleo-earthquakes as
possible so as to avoid creating spurious “phantom” events.
In this dissertation, I apply trenching techniques to investigate paleoseismic activity along the central
segment of the Garlock fault at two key sites, (1) Koehn Lake [KL (Chapter 3)] and (2) Straw Peak Road
[SPR (Chapter 4)]. At the KL site, a paleoseismic trench was excavated perpendicular across the fault at
the edge of a dry lake, where flat-lying lacustrine clays and silts interfinger with distal alluvial fan
deposits derived from the north and northeast. This area is characterized by shallow channels with varying
orientations, predominantly flowing southwest, making it an ideal location to study sedimentary layering
and fault offsets. At the SPR site, a trench was excavated in an area dominated by recent sedimentation
from fine-grained alluvial fan deposits transported southward across the Garlock fault. This site revealed
evidence of left-lateral offset within a drainage channel and an associated shutter ridge, a hallmark feature

9
of this fault. The SPR trench was strategically placed where the fault is single-stranded, just west of a
double-stranded section, to optimize the clarity of fault-related deformation in the subsurface.
1.3.3 Geochronological Techniques
1.3.3.1 Luminescence Dating
A key tool I use in all of these studies is luminescence dating, which aims to estimate the last time a
mineral grain was exposed to sunlight. Sediment grains absorb environmental radiation from naturally
occurring radioactive elements in the surrounding sediment, accumulating trapped electrons within crystal
lattice defects over time. These trapped electrons can be released when exposed to heat or light, emitting a
measurable luminescence.
Advances in luminescence techniques have allowed the dating of previously undatable landforms in the
Mojave Desert. Specifically, this technique has been shown to reliably date Holocene-latest Pleistocene
fluvial and alluvial sands along the Garlock fault (e.g., Dolan et al., 2016), giving exclusive access to
previously undatable slip rate and paleo-seismological sites along this fault. I use the post-IR IRSL225
single-grain potassium feldspar luminescence dating method developed by Rhodes (2015) to estimate
sample ages collected from features of interest. Specifically, I use IRSL dating to (1) estimate the age of
offset landforms (e.g., alluvial fans, terraces) for slip rate calculations, and (2) constrain the ages of paleoearthquakes in paleoseismic trenches. For slip rates presented in Chapters 2 and 5, samples were collected
for IRSL analysis by hand-excavating pits or cleaning exposed natural faces on landforms of interest.
Sections of the exposure without younger disturbances (e.g., bioturbation, cracking) were carefully
selected, and the outer few centimeters of sediment on the exposure were removed. Steel tubes 6 cm in
diameter were hammered into (preferably) sand-rich deposits, and in-situ gamma spectrometer
measurements were recorded at each sample position to determine the gamma dose rate. A small
subsample of sediment was collected from each sample hole for inductively coupled plasma (ICP)
analysis of radioactive elements to determine the beta radiation dose rate. Chapters 3 and 4 use a similar

10
collection technique for paleoseismic trenching in which steel tubes were hammered horizontally into
sand-rich target strata below and above paleo-earthquake event horizons documented in the trench
exposure, and in-situ gamma spectrometer measurements were recorded at each sample position. Small
sediment samples were collected from within the sample hole and sealed in plastic bags for ICP and water
content analysis.
All sediment samples were processed at the University of California, Los Angeles, and the University of
West Georgia. Samples were removed from their tubes in a light-proof laboratory, and the water content
for each sample was measured using the outer ~2.5 cm of sediment from the tube by weighing before and
after drying. The remaining sample was wet-sieved to isolate the 180-220 μm grain-sized fraction and
then dried. A lithium metatungstate (LMT) solution and a centrifuge were used to density separate the
potassium-feldspar grains. All samples were measured at the University of Sheffield, UK, using a singlegrain regenerative-dose protocol using potassium feldspar to provide an equivalent dose (De) estimate.
Each sample underwent repeated cycles of paired IRSL measurements at 50°C and 225°C, and in the
second half of each single aliquot regenerative dose (SAR) cycle, the IRSL response to a standard test
dose is determined (Rhodes, 2011). A fading correction of 1.12 was applied to all samples, as this has
been the reliable value used in desert environments such as the Mojave Desert (e.g., Dolan et al., 2016).
The full protocol for this analysis can be found in Rhodes (2015).
1.3.3.2 Radiocarbon dating
Complementary to IRSL dating, radiocarbon dating was used when datable carbon was present in my
excavations. Radiocarbon is a geochronometer that uses the decay of radioactive carbon (14C) isotopes to
measure the age of carbon-containing material (Libby et al., 1949). Living organisms continuously absorb
carbon isotopes from their surrounding environment through photosynthesis or food consumption. When
an organism dies, this absorption stops, and the carbon-14 isotopes begin to decay to nitrogen-14. The age

11
since the organism died can be estimated by the number of carbon-14 atoms remaining, which have a
half-life of 5730 ± 40 years.
Radiocarbon dating can provide robust age estimates for the age of organism death. However, a problem
inherent to this technique is that the age estimated from a sample represents the maximum age of the layer
from which it was collected. This is because (1) the residence time in the environment between the
cessation of exchanging carbon and the time the charcoal is deposited within a landform, and (2) the
reworking of charcoal from an older landform was incorporated into a younger deposit. For example, the
downstream post-fire delivery of material containing detrital charcoal is dominated by mass movements,
typically during infrequent large storms, which could lead to a long residence time if the time between
storms is large. Since climatic cycling influences the presence of charcoal deposited in landforms, it is
rare to encounter such samples during fieldwork. A notable exception to the rarity of detrital charcoal in
desert settings is its occurrence in playa environments. For example, Dawson et al. (2003) collected
numerous detrital charcoal samples in the El Paso Peaks trench located within a playa that occupies a
small, internally drained basin. Similarly, Madden and Dawson (2006) found abundant detrital charcoal
samples throughout their trench at the northeastern corner of Koehn Lake, which is one reason this
location was chosen for the trench (Chapter 3). Detrital charcoal samples were collected from the Koehn
Lake trench and the adjacent offset lacustrine berm using tools that would not contaminate the samples.
Upon collection, samples were stored in individual glass vials.
Samples were processed and analyzed at the University of California, Irvine, Keck Carbon Cycle
accelerator mass spectrometer (AMS) lab. Firstly, charcoal samples had contaminants (e.g., roots)
physically removed using forceps under a stereo microscope, followed by a multi-step acid-base-acid (1N
HCl and 1N NaOH, 75°C) chemical rinse to get back to the core original component of carbon when the
organism was alive. After pretreatment, the charcoal samples were sealed in quartz tubes, along with
copper oxide (CuO) and silver (Ag) wire, and were combusted at 900ºC for 3 hours to obtain carbon
dioxide (CO2). The last step before the AMS measurement is graphitization. CO2 produced from

12
combustion is cryogenically purified and quantified. Using the Bosch reaction (Vogel et al, 1984; Lloyd,
et al., 1991), samples were reduced to graphite in two successive reductions, the first to carbon monoxide
and then to graphite. The graphite powder was condensed and loaded into aluminum targets and
transferred to an apparatus loaded into the particle accelerator's ion source and run on the AMS to
determine the proportion of radiocarbon atoms in the sample. Detailed protocols for each step can be
found at sites.ps.uci.edu/kccams/education/protocol.
Radiocarbon ages reported from AMS measurements are not true calendar ages. Instead, they are based
on the assumption that atmospheric radiocarbon has always been the same as in 1950. Therefore,
calibration curves are used to convert them to calendar age equivalents. For radiocarbon ages documented
in this dissertation, I use the IntCal20: Northern Hemisphere atmospheric curve (Reimer et al. 2020), and
dates are reported as calibrated years before present (cal. yr B.P.), with “present” defined as 1950 C.E.
1.4 Research Implications
This dissertation greatly expands the incremental slip rate and paleo-seismological data available for the
Garlock fault, generating one of the longest and most detailed records of combined incremental fault slip
and paleo-earthquakes on any strike-slip fault in the world. The research presented in this dissertation
comprises two main components (1) slip rate studies for the western and central Garlock fault (Chapters 2
and 5) and (2) paleoseismological studies for the central Garlock fault (Chapters 3 and 4). I then combine
these records to generate an earthquake-by-earthquake record of fault slip over the Holocene and latest
Pleistocene to characterize observed irregular behaviors (Chapter 5). The results presented here have been
published in an international, peer-reviewed journal [Fougere et al., 2024 (Chapter 2)], or are currently in
preparation for submission [Fougere et al., in prep (Chapter 3); Fougere et al., in prep (Chapter 5)].
Chapter 2 presents results from slip rates studies from sites on the western and central segments of the
Garlock fault using lidar- and field-based mapping coupled with single-grain infrared-stimulated
luminescence dating. These new slip rates provide insights into the behavior of the Garlock fault,

13
reinforcing and refining previous slip rates that show that the Garlock fault experiences significant
temporal variations in slip rate that span multiple earthquake cycles, with multi-millennial periods of very
fast (13-14 mm/yr) early and late Holocene slip separated by a mid-Holocene period of slow slip (3
mm/yr).
Chapter 3 presents the longest paleo-earthquake record on the Garlock fault, with a minimum of ten
surface-rupturing events recorded during the past ca. 12.5 ka at the Koehn Lake paleoseismic site. The
paleo-earthquake ages from this site, originally studied by Burke (1979) and Madden and Dawson (2006),
were constrained using luminescence and radiocarbon dates on samples strategically collected from strata
that directly over- or under-lie pale-earthquake event horizons. Combining this new paleo-earthquake
record with the nearby El Paso Peaks record of Dawson et al. (2003) generates a likely complete record of
the past 14 paleo-earthquakes on this stretch of the central Garlock fault. The results of Chapter 3 confirm
that multi-millennial clustering of earthquakes is a common pattern of earthquake recurrence on the
Garlock fault. Specifically, clusters of four earthquakes were observed to occur over ~2 ky periods (0.5-2
ka, 5-7 ka, and 8-10 ka), separated by periods of less frequent to zero earthquake recurrence.
Chapter 4 details paleo-earthquake evidence for the Straw Peak Road site on the central Garlock fault, in
Pilot Knob Valley, the easternmost paleoseismic site along the Garlock fault. I present evidence of 3-5
earthquakes recorded at this site, however, the IRSL age data from this trench are pending as of this
writing, so no age constraints have been placed on the identified event horizons.
Chapter 5 synthesizes the incremental slip rate and paleo-seismologic data from the Garlock fault to
develop an earthquake-by-earthquake record of fault slip spanning the Holocene and latest Pleistocene.
By integrating the new and refined slip rates (presented in Chapter 2 and Chapter 5) with the timing of 14
surface-rupturing earthquakes along the same fault section, this study demonstrates that multi-millennial
clustering of earthquakes is a common pattern of behavior for the Garlock fault, at least over the past ca.
14 ky. This record highlights significant temporal variability in fault behavior, with periods of faster-than-

14
average slip coinciding with ~2-ky-long clusters of paleo-earthquakes recorded at Koehn Lake and El
Paso Peaks (Chapter 3). This comprehensive record highlights the significant temporal variability in
incremental slip rates, with periods of accelerated slip accumulating ~25 m of slip during the interval
(about four earthquakes) that are punctuated by slower periods during which accommodate only 7 ± 3 m
of slip (corresponding to 0–1 earthquakes). The results presented in this chapter offer a more complete
understanding of fault and fault systems dynamics and contribute to ongoing efforts to better assess
seismic hazard in southern California.
This dissertation is structured so that each chapter stands alone, leading to some inevitable redundancy
between chapters.

15
CHAPTER 2 Refined Holocene Slip Rate for the Western and Central Segments of the Garlock Fault:
Record of Alternating Millennial-Scale Periods of Fast and Slow Fault Slip
This chapter is based on the following published article:
Fougere, D., Dolan, J., Rhodes, E., & McGill, S. (2024). Refined Holocene slip rate for the
western and central segments of the Garlock fault: record of alternating millennial-scale periods of fast
and flow fault slip. Seismica, 3(2). DOI: 10.26443/seismica.v3i2.1202
2.1 Abstract
We use lidar- and field-based mapping coupled with single-grain infrared-stimulated luminescence dating
to constrain three new slip rate estimates from the western and central segments of the Garlock fault in
southern California, revealing a more complete picture of incremental slip rate in time and space for this
major plate-boundary fault. These new rates reinforce and refine previous evidence showing that the
Garlock fault experiences significant temporal variations in slip rates that span multiple earthquake
cycles, with multi-millennial periods of very fast (13-14 mm/yr) early and late Holocene slip separated by
a mid-Holocene period of slow slip (3 mm/yr). Similar ca. 8 ka slip rates for the central Garlock fault of
8.8 ± 1.0 mm/yr and 8.2 (+1.0/-0.8) mm/yr for the western Garlock fault demonstrate that the fault has
slipped at a faster long-term average rate than suggested by previous studies. These fast rates are
consistent with kinematic models in which the western and central Garlock fault segments are driven
primarily by lateral extrusion associated with N-S contractional shortening, with additional slip driven by
WNW-ENE Basin and Range extension north of the fault and minor rotation of the Garlock within the NS zone of dextral ECSZ shear.
2.2 Introduction
Documenting patterns of elastic strain accumulation and release through time and space on major faults is
of fundamental importance for understanding the mechanisms governing the behavior of plate boundary

16
fault systems and assessing the associated seismic hazard. Increasingly detailed slip records suggest that
some faults display relatively constant slip rates through time (e.g., Berryman et al., 2012; Gold et al.,
2011; Kozacı et al., 2009; Noriega et al., 2006; Salisbury et al., 2018), whereas others experience
significant temporal variations in slip rates over multiple earthquake cycles (Dolan et al., 2007, 2016;
Friedrich et al., 2004; Gold & Cowgill, 2011; Hatem et al., 2020; Ninis et al., 2013; Onderdonk et al.,
2015; e.g., Wallace, 1987; Weldon et al., 2004; Zinke et al., 2017, 2019; Zinke et al., 2021). Incremental
slip rate records from major faults provide insights into the constancy (or non-constancy) of fault slip
through time, constraining temporal and displacement scales over which some faults exhibit non-constant
behavior. This information, in turn, provides constraints on what mechanisms may be controlling this
behavior. Yet, the underlying mechanisms modulating these behaviors are not well understood. One
reason for this is that this discussion remains highly data-limited – there are still too few detailed
incremental slip rate records developed from hybrid studies that combine both earthquake timings and
displacements to develop a true “dated path” of slip.
The key to understanding the mechanisms driving slip rate variability lies in first documenting patterns of
fault slip through time and space on major plate boundary faults. Generating a rich incremental slip rate
data set records the phenomenology of time-variable fault slip, and permits a comprehensive assessment
of patterns of strain release through time and space. Such data provide key constraints on insights into the
physical mechanisms modulating variations in slip behavior observed on major faults, such as the Garlock
fault discussed in this study. Moreover, geologic slip rates are used as fundamental inputs in probabilistic
seismic hazard assessment (PSHA) models, which often disregard temporal variability over long (i.e.,
multi-millennia) time periods (e.g., Field et al., 2015, 2017). New incremental slip rates such as those
presented here will thus help to advance our understanding of the use of geologic slip rates for the
development of the next generation of PSHA models (e.g., Van Dissen et al., 2020).
In this paper, we present three new and refined mid-to-early-Holocene slip rates from the central and
western Garlock fault. These slip rates add to the emerging incremental slip rate record of the Garlock

17
fault, notably helping to resolve the sparse record during mid- to early Holocene time. We discuss these
results in light of their implications for alternating millennial-scale periods of fast and slow fault slip on
the Garlock fault as well as for system-level behavior of the plate boundary fault system in southern
California.
2.3 The Garlock Fault
The Garlock fault is a major left-lateral strike-slip fault that extends eastward from its intersection with
the San Andreas fault for 250 km to southern Death Valley (Figure 2.1). It forms a prominent
physiographic boundary between the E-W extension of the Basin and Range Province to the north and the
non-extending Mojave block to the south (Davis & Burchfiel, 1973). Cumulative displacement along the
Garlock fault is recorded as 48-64 km since fault initiation (Davis & Burchfiel, 1973; Monastero et al.,
1997; Smith, 1962; Smith & Ketner, 1970), which occurred between 17 and 10 Ma (Andrew et al., 2015;
Burbank & Whistler, 1987; Loomis & Burbank, 1988; Monastero et al., 1997), and likely at ca. 11 Ma
(Andrew et al., 2015; Blythe & Longinotti, 2013). The Garlock fault has been divided into three segments
defined by changes in strike and structural complexities. Specifically, a two-to-three-km-wide step-over at
Koehn Lake between the western and central segments coincides with a 10° change in strike, and a 15°
change in strike near the Quail Mountains between the central and eastern segments (S. F. McGill & Sieh,
1991). The E-W Garlock fault is embedded sub-perpendicular to Pacific-North American plate boundary
motion within a zone of N-NW-trending dextral shear known as the Eastern California shear zone
(ECSZ). The central and eastern segments of the Garlock fault lie within the ECSZ, but the western fault
segment lies to the west of this zone of dextral shear.
2.4 Previous Studies
Although the Garlock fault displays abundant geomorphic evidence for past earthquakes, it has not
generated any large ground-rupturing earthquakes during the historic period (Dawson et al., 2003; Dolan
et al., 2016; Ganev et al., 2012; Madden Madugo et al., 2012; S. F. McGill et al., 2009; S. F. McGill &

18
Sieh, 1991; McGill & Sieh, 1993; Rittase et al., 2014). Several paleoseismic investigations, however,
have documented evidence for large-magnitude (Mw > 7.2) Holocene earthquakes along the western and
central segments of the Garlock fault (Dawson et al., 2003; Madden Madugo et al., 2012; S. F. McGill et
al., 2009; S. F. McGill & Sieh, 1991; S. H. F. McGill, 1992; McGill & Rockwell, 1998; Pena, 2019).
Additionally, the western and central segments exhibit relatively fast slip rates, with several studies
recording latest-Pleistocene to early-Holocene rates of 5 to 8 mm/yr (Clark & Lajoie, 1974; Ganev et al.,
2012; McGill et al., 2009; McGill & Sieh, 1993), which are comparable to longer-term rates averaged
over million-year time scales (Burbank & Whistler, 1987; Carter, 1994; Loomis & Burbank, 1988;
Monastero et al., 1997).
On the western segment of the Garlock fault at Clark Wash (Figure 2.1), S. F. McGill et al. (2009)
documented a 66 ± 6 m offset of an older alluvial fan into which Clark Wash has subsequently incised.
They determined this offset by back-slipping the riser (i.e., stream bank) between the older fan surface
and the top surface of the younger incised channel deposits until they achieved a sedimentologically
plausible configuration. Eight detrital charcoal ages collected from the deepest exposed deposits of the
younger channel fill in a fault-parallel trench all yielded similar early Holocene calibrated radiocarbon
ages of 7.7-8.1 ka (S. F. McGill et al., 2009). The older fan deposits into which Clark Wash incised
contained a detrital charcoal sample representing a maximum-possible age of fan abandonment at this site
of 13.3 ka, indicating Clark Wash incision occurred between ca. 8.0 ka and 13.3 ka. Consequently, S. F.
McGill et al. (2009) used an age of 9.3 +4.0/-1.2 ka as the preferred age of Clark Wash incision and
subsequent offset, with this age based on the correlation of climatically driven pulses of alluvial fan
deposition within the Mojave Desert during the 8.0-13.3 ka range allowed by the radiocarbon dates
(Harvey & Wells, 2003). Combining the 9.3 +4.0/-1.2 ka preferred age and 66 ± 6 m offset yielded a
broad potential slip rate of 7.6 +3.1/-2.3 mm/yr slip rate averaged over Holocene to latest Pleistocene time
(S. F. McGill et al., 2009).

19
Along the westernmost part of the central segment of the Garlock fault, 30 km west of Summit Range
West, Clark & Lajoie (1974) documented a slip rate from a site on the northeast corner of Koehn Lake
(Figure 2.1) by combining a ~75 m berm offset (Clark, 1973) with radiocarbon-dated berm-crest
lacustrine tufa deposits yielding a slip rate of 4.5-6.1 mm/yr (after Ganev et al., 2012) (application of
dendrochronological calibration to the uncalibrated radiocarbon dates reported in Clark & Lajoie, 1974).
This slip rate is considered a minimum since the tufa was deposited before offset of the berm began.
Thirty kilometers farther east, Ganev et al. (2012) documented a slip rate along the central segment at
what we refer to as the Summit Range West site of 5.3 +1.0/-2.5 mm/yr based on the 70-m offset of a
well-defined, incised edge of an alluvial fan and a 10Be cosmogenic radionuclide depth profile age of the
fan surface of 13.3 +5.9/-1.1 ka. Ganev et al. (2012) suggested two additional potential slip rate estimates
at the Summit Range West site based on a 58 m offset of the deeply incised stream thalweg. Possible ages
of thalweg incision were inferred to correspond to significant climate changes in the southwestern United
States, one change which occurred at the end of the Younger Dryas (~11.5 ka) resulted in a more humid
climate, and another at 8-10 ka that marked the onset of summer monsoonal rainfall patterns, providing
estimated slip rates of 5.1 ± 0.3 mm/yr and 6.6 ± 1.2 mm/yr, respectively. In this paper, we refine both the
Clark Wash (S. F. McGill et al., 2009) and Summit Range West (Ganev et al., 2012) slip rates using
newly collected luminescence age data, and we present a new slip rate for the Summit Range East site,
located 3 km east of the Summit Range West site.
Although these earlier studies reveal relatively similar long-term slip rates averaged over latest
Pleistocene and early Holocene time of ~5-8 mm/yr, recent studies suggest that the Garlock fault slip rate
has been highly irregular during the Holocene. For example, Dolan et al. (2016) documented an
accelerated, late Holocene slip rate of >14 +2.2/-1.8 mm/yr from a series of 1.9 ka alluvial fans that have
been offset by ~26 m at the Christmas Canyon West site (~15 km east of Summit Range West), at least 2-
3x faster than the long-term average rate. Similarly, Rittase et al. (2014) also documented an accelerated,
late Holocene slip rate of 10.8 to 12.5 mm/yr from a shutter ridge at the Pilot Knob Valley site (Figure

20
2.1). Age constraints were based on soil development and one optically stimulated luminescence age from
deposits that may correlate with those of the buttressing shutter ridge. Notably, the fast late-Holocene rate
at Christmas Canyon West appears to coincide with a cluster of four earthquakes recorded on the central
segment that occurred between ca. 0.5 and 2 ka at the El Paso Peaks site (Figure 2.1, Dawson et al.,
2003). The fast late-Holocene rate in Pilot Knob Valley of Rittase et al. (2014) also includes that period
but may include slip accumulated between ~ 3.5 ka and ~ 0.5 ka.
2.5 New Slip Rate Data from the Central and Western Garlock Fault
In order to refine the current incremental slip rate record (i.e., by calculating new slip rates) for the
western and central Garlock fault, we use field mapping and analysis of high-resolution (0.5 m)
GeoEarthScope lidar data collected along the Garlock fault (data available at www.opentopography.org)
to accurately measure tectonically offset geomorphic features, combined with age estimates using post-IR
infrared stimulated luminescence (post-IR IRSL225) protocol (Rhodes, 2015) and radiocarbon dating of
detrital charcoal. We document three mid-to-early-Holocene slip rates from (1) the Summit Range East
site, a new slip rate for the central Garlock fault; (2) the Summit Range West site, a refined slip rate for
the central Garlock fault, following earlier work by Ganev et al. (2012); and (3) the Clark Wash site, a
refined slip rate from the western Garlock fault following on from the study of S. F. McGill et al. (2009).
The addition of new post-IR IRSL225 data for the Summit Range West and Clark Wash sites
significantly narrows the slip rate ranges at both of the latter sites.
2.5.1 Summit Range East
The Summit Range East (SRE) site is located on the north-facing mountain front of the Summit Range on
the central segment of the Garlock fault, ~3 km east of the Ganev et al. (2012) Summit Range West site
(Figure 2.2). The Garlock fault is particularly well-defined in the geomorphology along this stretch of the
fault. High-resolution lidar data reveal a narrow, linear fault trace with numerous well-defined left-lateral
offsets of northward-flowing streams and associated alluvial fans.

21
At the SRE site, the alluvial fan of interest, Qfc2, and its associated distributary fan channels have been
offset left-laterally by the Garlock fault (Figure 2.3a). Analysis of topographic profiles derived from lidar
data measured perpendicular to flow direction across alluvial fans Qfc1 and Qfc2 reveal a classic convexup morphology, suggesting that they might be the same fan, and that difference in fine-scale surface
morphology may be due to enhanced erosion of Qfc1 on the western edge of this fan system. At this site,
the alluvial fan of interest, Qfc2, and its associated distributary fan channels have been offset left-laterally
by the Garlock fault (Figure 2.3a).
Whether Qfc1 and Qfc2 are actually different-aged fans does not affect our slip rate determination based
on the restorations of Channel 1 (Ch-1) and Channel 2 (Ch-2). Alluvium at this site was sourced from
Mesozoic granites exposed in the Summit Range, south of the fault, and has been transported northwards
across the fault towards Searles Valley. On its eastern edge, the depositional edge of the Qfc2 fan atop
Mesozoic bedrock is exposed south of the fault. Two northeastward-flowing channels (Ch-2 and Channel
3 [Ch-3]) have incised into the southeastern edge of the Qfc2 fan and were subsequently offset by the
fault. The modern channel extending through the study site (Ch-0) has incised into the northwestern edge
of the Qfc1 and Qfc2 fans.
2.5.1.1 Offset Measurements
The fault offset at the SRE site is based on the restoration of (1) a prominent distributary channel incised
into the Qfc2 surface (Ch-1 in Figure 2.3a), and (2) one of the northeastward-flowing offset drainages
incised into the southeastern edge of Qfc2 (Ch-2 in Figure 2.3a). Offset measurements were determined
by incrementally back-slipping one side of the fault relative to the other to determine the minimumpossible, preferred, and maximum-possible offset values. Maximum- and minimum-possible offset
uncertainty limits were determined by progressively back-slipping the fault in 50 cm increments until a
point was reached where the restored geomorphic features were no longer in a sedimentologically
plausible configuration (see Appendix A).

22
Using these criteria, we measured independent offset estimates for Ch-1 and the largest offset Ch-2 where
our reported uncertainties represent the maximum and minimum limits of sedimentologically allowable
geometries. Ch-1 yielded a preferred offset measurement of 38 ± 1 m (Figure 2.3b and c). This tight error
constraint was determined by a pronounced bend of Ch-1 at the fault crossing at restorations smaller than
37 m and larger than 39 m. Ch-2 yielded an offset measurement of 40 ± 3 m. The minimum estimate was
determined on the basis that restorations smaller than 37 m would have shut off the downstream segment
of the incised channel from its upstream equivalent, beheading the channel at the fault trace. The
maximum estimate was determined by the incised channel forming a prominent “s-bend” across the fault
at restorations larger than 43 m. The width of the channel, post-offset modification of topography slightly
north of the fault, and the oblique orientation of the channel with respect to the fault result in slightly
larger error bounds for Ch-2 than for the sharply defined Ch-1 incised into the Qfc2 surface. Therefore, in
the following slip rate calculations, we used the tightly constrained prominent distributary channel offset
of 38 ± 1 m as our preferred displacement value. The independent offset for Ch-2 may be slightly larger
than Ch-1, however, the measurements of 38 ± 1 m and 40 ± 3 m for these two channels overlap within
error. An additional channel (Ch-3b), located 15 m west of and adjacent to the 40 ± 3 m offset channel,
has also incised into the southeastern edge of the Qfc2 fan south of the fault. Reconstructing Ch3-b with
Ch-2a reveals that this feature is likely a result of a younger incision event following the accumulation of
~15 to 20 m of left-lateral fault slip after deposition of Qfc2 ceased. We, therefore, do not consider this
offset in the slip rate calculation.
2.5.1.2 Age Constraints
We use the post-IR IRSL225 single-grain potassium feldspar luminescence dating method of Rhodes
(2015) to determine the age of Qfc2 alluvial fan deposition, a detailed description of this protocol for all
three sites discussed in this paper can be found in Supplementary Material 3. This protocol has already
been successfully applied to dating sites on the central Garlock fault (Dolan et al., 2016; e.g., Rhodes,
2015) and at other sites in southern California (e.g., Del Vecchio et al., 2018; Kirby et al., 2018; Saha et

23
al., 2021). We hand-excavated a 1.5-m-deep by 1-m-square pit (N 35.486133°, W 117.529533°, at an
elevation of 854 m) into the Qfc2 fan surface and collected four IRSL samples in steel tubes from sandrich horizons in a vertical sequence at depths between 25-95 cm (Figure 2.4a). The sediments exposed in
the pit consist of sand and sandy-pebble gravel, with a relative downward fining of grain size. We
recorded in-situ gamma spectrometer measurements at each sample position and collected samples for
inductively coupled plasma (ICP) analysis of radioactive elements to determine the radiation dose rate.
All four IRSL samples were corrected for fading using a uniform 12% correction, which is the value used
for previous studies in the Mojave Desert (Dolan et al., 2016; Rhodes, 2015).
The single-grain age plots reveal two distinct age populations within the two uppermost samples, with the
older age populations yielding values of 5,560 ± 370 yb2020 (SRE14-01, 0.25 m depth) and 5,390 ± 410
years yb2020 (SRE14-02, 0.45 m depth), where IRSL ages are reported in years before 2020 when these
samples were analyzed with 1-σ uncertainties (Table 2.1; Figure 2.4b). These older age populations are
indistinguishable from the two lowermost samples (Figure 2.4b), which yielded ages of 5,640 ± 260
yb2020 (SRE14-03, 0.65 m depth) and 6,110 ± 370 yb2020 (SRE14-04, 0.95 m depth). We used the older
age populations of the upper two samples (SRE14-01 and SRE14-02) with the third lowest sample
(SRE14-03) to calculate an average age of deposition for the Qfc2 fan. We choose to not use the deepest
sample (SRE14-04) and instead use the three youngest samples for our age estimate as these are more
representative of the geomorphic feature used to measure displacement.
We calculate a preferred estimate of the age of Qfc2 deposition by combining the three shallowest sample
calendar dates with Gaussian errors using a chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009;
Reimer et al., 2020) yielding an age estimate of 5,570 ± 190 yb2020 (2-σ uncertainties) for the deposition
of the Qfc2 fan (Figure 2.4c). Using the method to combine these dates described in Griffin et al. (2022)
yields an indistinguishable age range of 5675 ± 356 yb2020. The younger age populations from samples
SRE14-01 and SRE14-02 yielded ages of 1,890 ± 150 yb2020 and 1,820 ± 180 yb2020 (both report 1-σ

24
uncertainties). See Supplementary Material 3.6 for a more detailed description of the interpretation of
single-grain age distributions.
2.5.1.3 Slip Rate Calculation
We used our preferred offset value of 38 ± 1 m based on the prominent distributary channel (Ch-1) and
the 5.57 ± 0.19 ka age as an estimate of Qfc2 fan deposition to determine a slip rate at the SRE site of 6.8
± 0.3 mm/yr, with 2-σ uncertainties calculated in quadrature. Specifically, we constructed a triangular
density function for the 38 ± 1 m offset of the Qfc2 and a Gaussian density function for the 5,680 ± 340
year age estimate of the Qfc2 fan deposition. We then summed both distributions using pointwise
addition to generate a joint probability density function (PDF) by combining offset and preferred age
values and their uncertainties. Pointwise addition (i.e., statistical union) is used to combine PDFs because
the samples were collected from the same unit, and we are interested in the probability that a date exists
within any one of those ages. We also calculated a slip rate at this site using the full potential error range
of 37-43 m and the 5.57 ± 0.19 ka age estimate of the Qfc2 fan, yielding a similar slip rate of 7.0 ± 0.7
mm/yr. The new ~6.8 mm/yr slip rate calculated using our tightly constrained offset for the SRE site is a
minimum rate because an unknown amount of time may have elapsed between the cessation of fan
deposition and the subsequent earthquake when offset of the Qfc2 fan began to accrue.
To determine the vertical component of slip of Qfc2, we measured the down-fan profile gradient north
and south of the fault using lidar data to compare the elevation difference, yielding <0.5 m of vertical
displacement over the same time period (ca. 5.6 ka) as the 38 ± 1 m offset. These measurements of leftlateral and vertical offset indicate that the Garlock fault at this site exhibits almost pure left-lateral slip
with a ratio of ~76:1 (south-side up component of slip) shown in Appendix A.

25
2.5.2 Summit Range West
The Summit Range West (SRW) site was first identified by Clark (1973) and studied in detail by Ganev
et al. (2012). Like the SRE site, the SRW site is located on the north-facing mountain front of the Summit
Range on the central segment of the Garlock fault, 3 km west of the SRE site (Figure 2.2). At the SRW
site, a large channel that drains much of the Summit Range flows northwards across the fault and has
deeply incised into two alluvial fans, Qf1 and Qf2, both of which emanated from this same source
drainage. The alluvial fan of interest (Qf2) was deposited atop the older and more laterally extensive Qf1
fan surface. The deeply incised channel has been sharply offset left-laterally by the Garlock fault.
2.5.2.1 Offset Measurements
Ganev et al. (2012) measured an offset at the SRW site of 70 ± 7 m based on the restoration of the incised
edge of the Qf2 fan surface on the western margin of the channel. This western edge of the channel is
present both upstream and downstream of the fault (Figure 2.5). The maximum offset estimate is based on
a restoration of 77 m, because larger restorations shut off northward flow through the deeply-incised
channel. The minimum offset estimate is based on restorations smaller than 63 m, resulting in a sharp,
sedimentologically unlikely deflection of the Qf2 fan edge at the fault on the western side of the channel.
Additionally, Ganev et al. (2012) measured an offset of 58 ± 4 m of the deeply incised thalweg of the
channel and interpreted this somewhat smaller offset to have occurred after deeper incision of the
channel, which they inferred might be related to major climate events at ~8-10 ka or 11.5 ka (younger
than the age of initial incision of Qf2). This smaller ~58 m offset was interpreted to reflect gradual
straightening of the channel during incision as the western channel wall south of the fault and the eastern
channel wall north of the fault is more exposed to erosion during incision and downcutting. We follow
Ganev et al. (2012) in using the 70 ± 7 m restoration of the incised edge of the Qf2 fan on the western
margin of the channel in our slip rate calculation. We think this best reflects the location of the initial
incision, which is the event that is constrained by the IRSL age data collected from Qf2.

26
2.5.2.2 Age Constraints
We collected three new post-IR IRSL samples from a 1.5-m-deep by 1-m-square pit (N 35.47834°, W
117.55943°, at an elevation of 1006 m) hand excavated into the Qf2 fan surface in steel tubes from sandrich horizons in a vertical sequence at depths between 40-85 cm (Figure 2.6). We chose our sample pit
location to be only a few meters from the pit from where Ganev et al. (2012) collected their 10Be depth
profile to ensure we were dating the same surface. We recorded in-situ gamma spectrometer
measurements at each sample position and collected samples for Inductively Coupled Plasma (ICP)
analysis.
The three IRSL samples, once corrected for fading using a uniform 12% fading correction (Dolan et al.,
2016; Rhodes, 2015), yielded ages of 7840 ± 760 yb2020 (GF16-05, 0.41 m depth), 9050 ± 740 yb2020
(GF16-06, 0.63 m depth), and 6710 ± 840 yb2020 (GF16-07, 0.84 m depth), where IRSL ages are
reported with 1-σ uncertainty in years before 2020 when these samples were analyzed (Table 2.1). We
calculated an average age of Qf2 deposition by combining all three calendar dates with Gaussian errors
using a chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020) yielding an age
estimate of 7960 ± 900 yb2020 (2-σ uncertainties), which provides a maximum age for the initial incision
of the offset channel (Figure 2.6c). These new IRSL data demonstrate that earlier age estimates for the
deposition of the Qf2 fan (Ganev et al., 2012) were too old. Specifically, Ganev et al. (2012) collected a
Beryllium-10 (10Be) radionuclide depth profile from a pit located <5 m from the IRSL pit where samples
GF16-05, GF16-06, and GF16-07 were collected, which yielded an age from the Qf2 fan surface of 13.3
+5.9/-1.0 ka. However, our new IRSL ages presented here demonstrate that the Qf2 fan was actually
deposited at ca. 8 ka. This anomalously old 10Be age is consistent with the observations of Owen et al.
(2011), who showed that there is significant 10Be terrestrial cosmogenic nuclide inheritance within
cobbles and boulders in desert settings characterized by highly intermittent, short transport streams,
suggesting that sediments are eroded and transported slowly in these regions, which can lead to
overestimating the age of the dated feature.

27
2.5.2.3 Slip Rate Calculation
We used our preferred offset value of 70 ± 7 m and the 8.0 ± 0.9 ka age of the Qf2 fan to determine a slip
rate at the SRW site of 8.8 ± 1.0 mm/yr, with errors calculated in quadrature and 2-σ uncertainty reported.
Specifically, we constructed a triangular density function for the 70 ± 7 m offset of the Qf2 riser and a
Gaussian density function for the 7960 ± 900-year age estimate of the Qf2 deposition. We then summed
both distributions to generate a joint PDF to combine offset and preferred age and their uncertainties. This
new ~8.8 mm/yr slip rate for the SRW site is a minimum rate because an unknown amount of time may
have elapsed between the cessation of fan deposition and the occurrence of the next earthquake (i.e.,
when offset of the Qf2 fan began to accrue).
2.5.3 Clark Wash
The Clark Wash (CW) site was first identified by Clark (1973) and studied in detail by S. F. McGill et al.
(2009). The CW site is located near the eastern end of the western segment of the Garlock fault, south of
the south-westernmost Sierra Nevada (Figure 2.7).
The CW site comprises an alluvial fan complex sourced from an ~1 km2 area of granitic bedrock (Smith,
1964) of the southern Sierra Nevada batholith via Clark Canyon. At the CW site, a latest Pleistocene
alluvial fan (Qfo) was deposited containing charcoal that yielded calibrated radiocarbon ages ranging
from 22.7 to 13.3 ka (Figure 2.8a). The Qfo fan is composed of alternating layers of grain-supported,
small-pebbly-to-coarse-sand stream deposits and matrix-supported debris flows (S. F. McGill et al.,
2009). Several trenches exposed Qfo capped by a buried soil which, in turn, was buried by a ~1-m thick
veneer of younger (latest Pleistocene to early Holocene) alluvial deposits (Qfy), which have not been
directly dated in previous studies. Following the deposition of Qfo, development of buried soil, and
deposition of Qfy, Clark Wash incised deeply into these fan deposits (Figure 2.8a) and partially filled
with fluvial-sourced, thickly bedded, gravelly sand (Hoa) and interfingering sandy colluvium derived
from the channel wall and from the Garlock fault scarp (Hoc; S. F. McGill et al., 2009). Hoa and Hoc

28
deposits were only exposed in trenches and are located beneath stream deposits (Hya) and Holocene
colluvium (Hyc) shown in Figure 2.6a. Calibrated radiocarbon dates from detrital charcoal and hearth
features within these deposits range from 8.0 – 6.8 ka. A younger episode of channel incision and fill
occurred during late-Holocene time, as shown by Hya deposits exposed in one of the S. F. McGill et al.
(2009) trenches south of the fault containing a ca. 2.5 ka charcoal sample (S. F. McGill et al., 2009),
which, when recalibrated in this study using Oxcal version 4.4 and the most up-to-date calibration curve
(IntCal20, Bronk Ramsey, 2009; Reimer et al., 2020) yielded an age of 2530 ± 220 cal. yrs B.P. (2-σ
uncertainty).
2.5.3.1 Offset Measurements
The refined early Holocene slip rate measurement presented here is based on the offset of Clark Wash
incised into the Qfy fan. Specifically, a 2-to-3-m-tall terrace riser separating Qfy from younger Hyc and
Hya (Figure 2.8a). This well-defined offset is based primarily on the restoration of the northeastern
channel wall. The offset value for this feature was measured as 66 ± 6 m by S. F. McGill et al. (2009)
using a combination of surveyed piercing points along the top of the terrace riser on the northeastern edge
of Clark Wash, the inner edge of the younger, Clark Wash fluvial strata (Hoa) exposed in trenches where
it had been buried by colluvium (Hoc and Hya), and a photogrammetric map with 0.5 m contours.
Reevaluation of this offset in this study using lidar-derived hillshade confirms that the northeastern
terrace riser is left-laterally offset by 66 ± 6 m (Figure 2.8b and c). The offset of the southwestern terrace
riser cannot be constrained as tightly because the piercing point where the southwestern wall of the
channel that once intersected the southeastern side of the fault has been eroded by Clark Wash. Using a
photogrammetric map with 0.5 m contours, S. F. McGill et al. (2009) estimated the offset to be 65 ± 7 m,
consistent with (but with broader uncertainty than) the offset measured for the northeastern wall. Figure
2.8a shows one likely projection of piercing points for the southwestern wall of the channel, which results
in an offset estimate of 60 m.

29
2.5.3.2 Age Constraints
Radiocarbon ages presented here were originally collected and calibrated by S. F. McGill et al. (2009)
and were recalibrated in this study using Oxcal version 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020).
We combined these ages with new IRSL data collected from a pit excavated into the Qfy fan, which was
not dated by S. F. McGill et al. (2009), because no charcoal samples were found from it. Because Qya is
the youngest deposit that has been incised by the offset channel, its age is critical to providing the tightest
possible maximum age constraint on the slip rate at this site. In this study, the minimum age estimates for
the onset of offset of the channel wall separating Qfy (and the underlying Qfo) from Hoa was determined
from recalibrated radiocarbon dates from the same charcoal samples from Hoa used in S. F. McGill et al.
(2009) slip rate estimate. However, whereas McGill et al.’s maximum age estimate for the incision of this
riser was based on the youngest age from Qfo (13.3 ka), in this study a much tighter maximum age
estimate was determined from our new IRSL data collected from the previously undated Qfy fan deposits.
Maximum age using post-IR infrared stimulated luminescence dating
We collected four IRSL samples in steel tubes from sand-rich horizons from a pit (N 35.20578°, W
118.08663° at an elevation of 864 m) excavated into the Qf2 surface in a vertical sequence at depths
between 35-104 cm (Figure 2.9a). We recorded in-situ gamma spectrometer measurements at each sample
position and collected samples for ICP analysis. Post-IR IRSL luminescence ages from these samples
reveal a layered fan structure. The three upper samples (GF16-01, GF16-02, and GF16-03) collected at
depths between 34-70 cm were from Qfy. After correcting for fading using a uniform 12% correction,
these samples yielded ages of 8490 ± 490 yb2020 (GF16-01; 0.34 m depth), 8420 ± 460 yb2020 (GF16-
02; 0.52 m depth), and 7100 ± 560 yb2020 (GF16-03; 0.70 m depth), where IRSL ages are reported with
1-σ uncertainties reported in years before 2020 (Table 2.1). We calculate the age of Qfy fan deposition by
combining the three upper samples with Gaussian probability distribution using a chi-squared test (Bronk

30
Ramsey, 2009; based on OxCal4.4; Reimer et al., 2020). This yields a preferred age of Qfy deposition of
8095 ± 575 years (2-σ uncertainty; Figure 2.9c).
We collected sample GF16-04 at 105 cm depth from the same pit as the three uppermost samples. This
sample, which yielded an age of 12500 ± 900 yb2020 after a uniform 12% fading correction (1-σ
uncertainty), was collected from Qfo deposits that underlie the Qfy deposits at this site. The Qfo fan
deposits were readily distinguished from Qfy deposits because they are very well consolidated with a
reddish matrix of translocated clay, whereas the overlying Qfy fan deposits are poorly sorted, massive,
and matrix supported. The depth of the contact between Qfy and Qfo was similar to that observed in S. F.
McGill et al. (2009) trenches. Since we are concerned with the 66 ± 6 m offset of the Qfy fan, we do not
use the ~12.5 ka age of GF16-04 in our slip rate calculations, but we note that it is consistent with the
youngest radiocarbon age from Qfo (13.3 ka) that was reported by S. F. McGill et al. (2009).
Minimum age using radiocarbon dating
S. F. McGill et al. (2009) collected eight charcoal samples from the deepest part of a trench excavated
into the Holocene channel fill of Clark Wash (Hoa) and colluvium (Hoc) sourced from the channel wall
of the incised Qfo fan and its overlying veneer of Qfy. We recalibrated radiocarbon dates for these eight
charcoal samples using OxCal version 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020). The oldest ages
from Hoa and Hoc provide a minimum bound on the age of initial incision of the Qfy fan by Clark Wash,
because Hoa and Hoc were deposited after the incision that formed the channel wall that is now offset by
66 ± 6 m. We derived minimum age constraints for this incision in two ways: (A) the average age derived
from all eight radiocarbon samples from the Hoa channel-fill deposits and the Hoc colluvial deposits that
interfinger with Hoa, and (B) the calibrated age of the stratigraphically lowest and oldest sample (H14-
6N-44).
We combined the eight radiocarbon ages using Gaussian probability distributions using a chi-squared test
(Bronk Ramsey, 2009; based on OxCal4.4; Reimer et al., 2020), yielding an average age of the channel-

31
fill deposits of 7,870 ± 88 cal. yrs BP (95% confidence interval; Figure 2.9d). A similar, but slightly
older, maximum age for channel incision of 8,065 ± 135 cal. yrs B.P was determined using the calibrated
age of the oldest charcoal sample, H14-6N-44, collected from the lowest stratigraphic level of the trench
(S. F. McGill et al., 2009).
Average age for fault offset
The age of the onset of fault offset for the channel wall that separates Qfo and its overlying cap of Qfy
from Hoa is between the age of deposition of Qfy (8095 ± 575 years ago), which is older than the initial
incision of the channel, and the age of the oldest deposits within the channel, Hoa/Hoc (7,870 ± 88 cal.
yrs BP), which is younger than age of initial channel incision. To calculate the time interval between the
minimum and maximum age, we used the RISeR program (Zinke et al., 2017, 2019) to calculate the
probability function representing the interval between the age PDFs derived from the upper terrace
(determined from IRSL) and the lower terrace (determined from radiocarbon). This yielded an age
constraint for the age of incision of 8010 +1115/-210 years (95% confidence interval, Figure 2.9e).
2.5.3.3 Slip Rate Calculation
To calculate a revised slip rate for the Clark Wash site, we combine the estimate for the age of incision of
the offset channel wall (8010 +1115/-210 years) with the preferred restoration of the offset channel wall
of 66 ± 6 m, yielding a slip rate of 8.2 +1.4/-0.8 mm/yr (2-σ uncertainty calculated in quadrature). This
rate is based on the well-constrained age of incision into the Qfy fan surface and subsequent deposition
into Clark Wash, and does not consider the possibility or the unknown duration of time between this
incision event and the first earthquake to occur after the incision. Therefore, this rate should be considered
a minimum slip rate.

32
2.6 Discussion
2.6.1 Multi-Millennial Slip Rate Variations
The three new slip rates presented here from the Summit Range East (SRE), Summit Range West (SRW),
and Clark Wash (CW) sites add to the growing incremental slip rate record for the western and central
Garlock fault (Dolan et al., 2016; Ganev et al., 2012; S. F. McGill et al., 2009; McGill & Sieh, 1993;
Rittase et al., 2014). Advanced luminescence dating techniques have allowed us to precisely date the
deposition of alluvial fans at the SRW site to revise and more tightly constrain the previous, slower slip
rate estimate of 5.3 +1.0/−2.5 mm/yr averaged over 13.3 ka documented by Ganev et al. (2012), to a
revised rate of 8.8 ± 1.0 mm/yr averaged over 8.0 ka. In addition, these new age data have allowed us to
date a previously undated unit at the Clark Wash site, resulting in a substantially tighter bound on the slip
rate at that site of 8.2 +1.0/-0.8 mm/yr averaged over the past 8.0 ka, compared to the previously
published rate of 7.6 +3.1/-2.3 mm/yr averaged over a time interval spanning 0 to 9.3 ka (S. F. McGill et
al., 2009). Finally, we also have documented a new slip rate of 6.8 ± 0.3 mm/yr over the past 5.6 ka at a
new site–SRE.
Combining the three new and refined slip rates with (a) previously published records of incremental slip
rate averaged over different time periods during Holocene to latest-Pleistocene time, and (b) the most
recent earthquake (MRE) age documented at the El Paso Peaks (EPP) site (Dawson et al., 2003) and the
Christmas Canyon site (Pena, 2019), allow us to construct a detailed time-displacement history for the
central Garlock fault since ca. 8 ka (Figure 2.10). This record facilitates testing for possible correlations
between periods of earthquake clustering and seismic lulls with potential accelerations and decelerations
of fault slip rate spanning multiple earthquake cycles, as has been suggested in previous studies (e.g.,
Dawson et al., 2003; Dolan et al., 2007, 2016; Ganev et al., 2012; Rittase et al., 2014).
The incremental slip history of the central Garlock fault is marked by significant (factor of 2 to 5 times)
variations in slip rate spanning cycles of thousands of years (~2 to 4 ky) during latest Pleistocene and

33
Holocene time. Specifically, the new slip rate averaged since 5.7 ka of 6.8 ± 0.3 mm/yr from the central
segment of the fault (SRE site) is consistent with previous studies of longer-term average slip rates since
latest-Pleistocene to early-Holocene time of 5 to 8 mm/yr (Clark & Lajoie, 1974; Ganev et al., 2012; S. F.
McGill et al., 2009; McGill & Sieh, 1993). However, our revised ca. 8 ka-averaged slip rates of 8.8 ± 1.0
mm/yr from the central segment (SRW site) and 8.2 +1.4/-0.8 mm/yr from the eastern extent of the
western segment (CW site), reveal two of the fastest long-term (>6 ka) slip rates of any site documented
along the Garlock fault.
Incremental slip rates for the central Garlock fault for various time intervals since the latest Pleistocene
vary from 3.1 mm/yr to 13.6 mm/yr (Figure 2.10b). We calculated incremental slip rates between agedisplacement pairs using RISeR’s analytical formulation (Zinke et al., 2017, 2019) based on Bayesian
statistics, which resamples the age and displacement measurements according to their PDFs to calculate a
slip rate over a specified time interval. We chose an analytical sampling approach (rather than Markov
Chain Monte Carlo sampling) as it is ideal for data sets in which dated markers are independent (i.e., do
not overlap within uncertainty), which is the case for the data presented here (see Supplementary Material
4.1). Differences in slip rate between the SRW and SRE sites are due to the different time periods that
they sample, not due to changes in slip rate along strike, because these two sites are separated by only a
few kilometers along a continuous, structurally simple section of the fault. The CCW slip rate also lies
along this same simple section of the fault, so we infer that all three incremental slips reflect the same
changes of rate along the central Garlock fault.
Periods of faster slip on the central Garlock fault (e.g., 0.5-1.9 ka and ca. 5-8 ka) appear to correspond
with bursts of earthquake recurrence recorded on the central segment, and slower periods appear to
correspond with periods of relative seismic quiescence. For instance, the paleoseismic record generated
by Dawson et al. (2003) at the EPP site, 20 km west of the SRE site on the central segment (Figure 2.1),
displays a four-earthquake cluster between 0.5 and 2.0 ka, corresponding to a faster-than-average slip rate
during this period of >14.0 +2.2/-1.8 mm/yr (Dolan et al., 2016). Previously, the absence of any

34
detectable paleo-surface ruptures between 1.9 ka and 5.1 ka in the EPP trench suggested that the Garlock
fault did not generate any surface-rupturing earthquakes for more than 3,000 years (Dawson et al., 2003),
which Dolan et al. (2016) interpreted to have resulted in a “zero” slip rate over this time interval.
However, given that 38 ± 1 m of left-lateral slip has occurred at the SRE site since ~5.6 ka, and 26 +3.5/-
2.5 m of slip has occurred at the CCW site since ~1.9 ka, it is clear that 12 +1/-2.5 m of slip occurred
during the ~3.7 ky interval between 1.9 ka and 5.6 ka. This results in an incremental slip rate of 3.1 ± 0.4
mm/yr during this period of relative quiescence. Additionally, Dawson et al. (2003) recorded one paleosurface rupture at ~5.1 ka within this slow-slip interval between 1.9-5.6 ka (Figure 2.10). However, as
noted above, we suspect there may have been one or more additional paleo-earthquakes during this slowslip interval that were not recognized at the EPP trench. Preceding the 1.9-5.6 ka slow-slip interval, we
use RISeR (Zinke et al., 2017, 2019) to calculate an incremental slip rate of 12.9 +3.2/-2.3 mm/yr
between 5.7 ka and 8.0 ka. Although we do not have incremental slip rate data for the western Garlock
fault, it is noteworthy that the average 8.0 ka slip rate at the CW site, on the western Garlock fault, is very
similar to the average slip rate for the past 8.0 ka on the central Garlock fault at the SRW site.
These new results add to the growing body of evidence suggesting that the Garlock fault experiences a
highly irregular temporal pattern of incremental slip (i.e., non-constant behavior) over periods that span
multiple earthquake cycles. Namely, the incremental slip history of the central Garlock fault can be
characterized as a period of relatively fast strain release during the late Holocene (0.5-1.9 ka), preceded
by a period of slow but not zero strain release during the mid-Holocene (1.9-5.0 ka), which in turn was
preceded by another period of fast strain release during the mid-to-late-Holocene (ca. 5-8 ka). The new
slip rate data presented here offer a more robust and nuanced view of the slip history of the Garlock fault
since the latest Pleistocene.

35
2.6.2 Comparison of Geodetic Slip-Deficit Rate with Incremental Geologic Slip Rates
Supporting the observations of variations in elastic strain release rate through time suggested by our new
incremental slip rate variations from sites along the western and central Garlock fault, the fast (8.2-8.8
mm/yr) average geologic slip rates since early Holocene time, averaged over 65-70 m of slip and
numerous earthquakes, do not match decadal-scale geodetic slip-deficit rates in the region. Specifically,
geodetic slip-deficit rates measured along the Garlock fault suggest negligible, to slow, left-lateral strain
accumulation (~0-3 mm/yr) over the past several decades (Evans et al., 2016; Gan et al., 2000; McClusky
et al., 2001; Meade & Hager, 2005; Miller et al., 2001; Peltzer et al., 2001; Savage et al., 1981, 1990,
2001). A recent block model suggests a slip-deficit rate for the central Garlock fault of ~2.6 ± 3.0 mm/yr
(Evans, 2017b). These observations are inconsistent with the geologic slip rates averaged since early
Holocene time that we document for the western and central Garlock fault (this study) and more closely
resemble the ~3 mm/yr slip rate averaged over the relative seismic lull between ca. 2 ka and 5 ka (Figure
2.10a). This geologic-geodetic rate discrepancy is consistent with the possibility that relatively slow rates
of interseismic fault loading for the Garlock fault could correlate with periods of relatively slow fault slip,
as has been suggested for other faults (e.g., Gauriau & Dolan, 2024), which may be correlated with the
current quiet period in seismic slip and earthquake occurrence since the MRE ca. 500 years ago (Dawson
et al., 2003). Moreover, geodetically constrained models of elastic strain accumulation along the Mojave
section of the San Andreas fault estimate a long-term slip rate of ~28 mm/yr (Moulin & Cowgill, 2023),
which suggest similar temporal slip variations as the Garlock fault, with decadal-scale geodetic slipdeficit rates of 15 ± 3 mm/yr (Evans, 2017b), only about half as fast as the long-term geologic slip rate of
35 mm/yr (Evans, 2017b) for the Mojave section of the San Andreas fault. E. H. Hearn et al. (2013) and
E. Hearn (2022) suggest that the geodetic slip-deficit and geologic slip rate discrepancy on the Mojave
section of the San Andreas fault can be partially explained by a “ghost transient” associated with the 1857
M~7.8 Fort Tejon earthquake, where the relatively slow elastic strain deformation on the Mojave section
of the San Andreas fault could be partially caused by long-term viscoelastic relaxation of the lithospheric

36
mantle and lower crust following past earthquakes (e.g., 1857 Fort Tejon). Interestingly, the GPS velocity
perturbations caused by large earthquakes on the Garlock Fault (i.e., MRE) do not influence block-model
geodetic slip-deficit rates, which suggests that the discrepancy between geodetic slip-deficit and geologic
slip rates for the Garlock Fault is not due to a ghost transient. If this apparent geodetic-geologic
discrepancy for the Garlock fault is accurate (and some have suggested that it is not; e.g., Chuang &
Johnson, 2011; Platt & Becker, 2013) (but see Evans, 2017a), it supports the notion that this region is
experiencing temporal variations in both strain accumulation and release spanning millennia and several
tens of meters of fault slip.
2.6.3 Evaluation of Driving Mechanisms for Garlock Fault Slip
The similar Garlock fault slip rates of 8.2 and 8.8 mm/yr averaged over the past ca. 8.0 ky that we
document at the CW and SRW sites indicate that, on average, the eastern part of the western segment and
the western part of the central segment have slipped at approximately the same average rate over the past
8 ky (Figure 2.10a). Although additional incremental slip rate measurements from the western segment of
the Garlock fault are needed to determine whether these two fault sections experienced similar slip
histories throughout Holocene time, the newly revised Clark Wash slip rate from the western segment,
together with detailed Holocene slip history for the central Garlock fault (Figure 2.10), allows us to
evaluate models of potential driving forces for slip on the central Garlock fault (e.g., Hatem & Dolan,
2018). For example, the SRW and SRE sites are located on a part of the Garlock fault embedded within
the westernmost part of the ECSZ region of N-S dextral shear, where the fault may be partially loaded by
clockwise rotation (Garfunkel, 1974; Guest et al., 2003; Hatem & Dolan, 2018; S. F. McGill et al., 2009).
In contrast, the CW site is located west of the western edge of this zone of dextral shear, and therefore this
segment of the fault is not being loaded via rotation within the ECSZ dextral strain field. Thus, the
similarity of the ca. 8 ka rates at SRW and CW indicates that very little (and potentially almost none) of
the slip on the central segment of the Garlock fault is a result of rotation-induced loading on these
timescales. Rather, these new rates support the model proposed by Hatem & Dolan (2018), where the

37
western and central segments of the Garlock fault are loaded primarily by lateral extrusion associated
with N-S shortening in the region of the more westerly-striking section of the San Andreas fault bordering
the Mojave Desert. This model suggests that the western and central segments of the Garlock fault will
accommodate strain at an increased rate during periods when the San Andreas fault is slipping faster (e.g.,
Dolan et al., 2007). Moreover, these new slip rates from both the western and central Garlock fault
provide a detailed record that can be compared with similar records from the San Andreas fault and ECSZ
which have yet to be documented.
2.6.4 Implications for Probabilistic Seismic Hazard Assessment
Most physics- and statistics-based seismic hazard models typically follow the assumption that faults are
loaded at a constant rate (i.e., strain accumulation and release rates remain constant between earthquake
cycles except for brief periods of post-seismic deformation). However, many studies, including this one,
challenge this assumption, and suggest that it may not be appropriate since at least some faults have been
shown to experience periods of faster-than-average slip separated by periods of slower-than-average slip
(Dolan et al., 2007, 2016; Hatem et al., 2020; Ninis et al., 2013; Onderdonk et al., 2015; e.g., Weldon et
al., 2004; Zinke et al., 2017, 2019; Zinke et al., 2021). As an example, it is evident from the updated
incremental slip rate record shown in Figure 2.10 that the central (and possibly western) Garlock fault
experiences temporally variable slip rates, during which fault slip has sped up and slowed down over
millennial time scales, with fast periods spanning multiple earthquakes and 25-30 m of displacement.
These results support the idea that non-constant behavior may be the expected mode of slip through time
on faults such as the Garlock fault that lie within structurally complex fault networks, as suggested by
Gauriau & Dolan (2021), who have shown that non-constant fault slip through time may be a response to
complex stress interactions within structurally complicated fault systems involving either temporal
variations in fault strength and/or kinematic interactions between faults within a plate boundary system. If
rates of elastic strain accumulation also vary through time, as suggested by the geodetic-to-geologic rate
discrepancy for the Garlock fault, then PSHA models based on the assumption of constant strain

38
accumulation and release may not provide a useful prediction of near-future fault behavior. Gauriau &
Dolan (2024) have shown that such geologic-to-geodetic discrepancies may be typical for faults such as
the Garlock fault that lie within complex plate boundary fault systems.
Such results are of fundamental importance for probabilistic seismic hazard assessment, which relies on
fault slip rates as a basic input parameter. But if fault slip rates on many faults vary through time, as in the
case of the Garlock fault documented in this paper, it remains unclear what the “correct” slip rate is for
use in PSHA. As detailed by Van Van Dissen et al. (2020), there are multiple ways of dealing with this
under-constrained aspect of hazard estimation, including assuming very large bounds on the potential
variability in earthquake recurrence that encompass the entire range of slip rate variability revealed by
detailed incremental fault slip-rate records like the one detailed herein. But treating the variability of
future earthquake recurrence in this manner yields little predictive value, given the large error ranges that
result from this approach. If, alternately, we use a single slip rate calculated over some arbitrary
displacement range, we risk either under- or overestimating the hazard based on whether that particular
displacement range yields a slip rate that is slower or faster than the current rate of fault slip (i.e., is that
fault currently experiencing an earthquake cluster or an earthquake lull). What is needed to more
accurately constrain this most basic input parameter for PSHA are many more incremental slip-rate data
sets like the one we document in this paper. Such records from many more faults of different kinematics
in different tectonic settings will be necessary to develop statistically meaningful ranges on the input
parameters used in next generation PSHA to ensure more accurate estimation of future earthquake
recurrence probabilities.
2.7 Conclusions
The new and revised slip rates presented here bolster and refine the previously sparse record of slip
during mid-to-early Holocene time on the Garlock fault, providing a more nuanced view of the behavior
of this fault than that suggested by previous results. Specifically, the new Summit Range East site data

39
suggest that the fault did slip, albeit slowly, within the ~3300-year apparent seismic lull documented in
earlier studies between 2.0 and 5.3 ka. Thus, earthquakes were occurring at a lower frequency or less
elastic strain energy was released per event during this “slow” period, rather than one subsystem
switching “off” entirely. As shown in this example, combining detailed incremental slip histories with
paleo-earthquake age and displacement data, and geodetic measurements, provides insights into the
behavior of major strike-slip faults, with implications for seismic hazard assessment and the geodynamics
controlling strain accumulation and release in the upper crust across the complex plate boundary fault
network in southern California. By documenting these factor of two to five-fold temporal variations in
slip rate for the Garlock fault, we can significantly improve our understanding of the manner in which
relative plate motions are accommodated through time in southern California.

40
CHAPTER 3 The Garlock Fault Paleoseismic Record at Koehn Lake: Implications for Multi-Millennial
Earthquake Clustering and Fault Interactions
This chapter is based on the following manuscript in preparation:
Fougere, D., Dolan, J.., Ivester, A., McGill S., Rhodes, E., Anthonissen, C., and Gauriau, J. (in
prep.). The Garlock Fault Paleoseismic Record at Koehn Lake: Implications for Multi-Millennial
Earthquake Clustering and Fault Interactions.
3.1 Abstract
Long paleoseismic records that span multiple earthquake cycles are essential for understanding patterns of
strain release along faults and fault systems. However, such records are rare, with only a limited number
of paleoseismic sites worldwide preserving complete sequences of ten or more prehistoric earthquakes on
major strike-slip faults. Previous studies along the western and central segments of the Garlock fault have
documented temporal clustering of earthquakes, suggesting variability in recurrence patterns over time.To
further investigate the mechanisms influencing the timing and spatial distribution of earthquakes in
southern California, we refine and extend the paleoseismic record from a trench excavated across the
central Garlock fault. Our study presents a newly developed paleo-earthquake chronology from a trench
site near the northeastern corner of Koehn Lake. By incorporating luminescence dating to constrain event
horizons observed in the trench exposures, we improve the resolution of the earthquake history along this
segment of the fault. Our findings extend the record of paleo-earthquakes into the early Holocene-latest
Pleistocene, providing a longer-term perspective on earthquake recurrence. The results confirm that
clusters of frequent earthquakes are punctuated by periods of relative quiescence, reinforcing the idea that
seismic activity on the central Garlock fault is not solely controlled by steady tectonic loading. Instead,
variations in earthquake recurrence appear to be influenced by static stress changes from interactions with
nearby faults, contributing to episodic rather than continuous seismic activity.

41
3.2 Introduction
Long paleoseismic records of earthquake recurrence spanning numerous earthquake cycles are key to
understanding patterns of strain release on faults and fault systems. However, such records are rare, with
only a few paleoseismic sites in the world yielding complete records of ten or more paleo-earthquakes on
major strike-slip faults (i.e., Wrightwood, San Andreas fault; Yammouneh fault, Lebanon; Hokuri Creek,
Alpine fault; Hog Lake, San Jacinto fault, Weldon et al., 2004; Daёron et al., 2007; Berryman et al., 2012;
Rockwell et al., 2015), limiting our ability to understand the spatial and temporal patterns of earthquake
occurrence and the underlying mechanisms. Existing paleoseismic records reveal some faults exhibit
relatively similar interevent times and nearly periodic behavior with very low coefficients of variation
(CoV) of <0.2 to ~0.3 (e.g., Okumura et al., 2003; Kondo et al., 2004, 2010; Berryman et al., 2012). In
contrast, other paleo-earthquake data sets demonstrate that on some faults the spatial and temporal
patterns of earthquake occurrence are much more complex (e.g., Dawson et al., 2003; Ganev et al., 2010;
McAuliffe et al., 2013; Weldon et al., 2004). For example, patterns of irregular slip rate through time
spanning multiple earthquakes have been observed on a number of major faults and fault systems, such as
at the Wrightwood site on the Mojave section of the San Andreas fault (Weldon et al., 2004), the Wairau
fault in New Zealand (Zinke et al., 2021), and the faults of the Eastern Californian shear zone (Rockwell,
2000; Ganev et al., 2010; McAuliffe et al., 2013), and southern end of the on-land segment of the Dead
Sea fault system (Klinger et al., 2015). These records suggest that temporally variable fault behaviors are
common features of seismically active regions, and support the idea that faults embedded within
tectonically complex settings are more prone to irregular fault slip and earthquake clustering (Gauriau and
Dolan, 2021; 2024).
To better understand mechanisms controlling the temporal and spatial patterns of earthquake recurrence
in southern California, in this study we examine the paleoseismic record from a trench we excavated
across the central Garlock fault. Past paleoseismic studies along the western and central segments of the
Garlock fault demonstrate variability in earthquake occurrence (e.g., Dawson et al., 2003; Gath and

42
Rockwell, 2018; Kemp et al., 2016; Madden and Dawson, 2006; Madugo et al., 2012; McGill et al., 2009,
in prep; Peña, 2019; Peña et al., in prep), with clusters of earthquakes encompassing ~25 m of cumulative
fault slip are concurrent with periods of fast slip (Dolan et al., 2016; Fougere et al., 2024). Herein, we
bolster the Holocene paleo-earthquake record for the central Garlock fault by documenting evidence for at
least ten surface-rupturing earthquakes during the past ca. 11,500 years at the Koehn Lake site. These new
data will help us to evaluate the potential mechanisms driving the pattern of earthquake recurrence, such
as multi-millennial supercycling between mechanically complementary fault subsystems (e.g., Dolan et
al., 2007, 2016), and examine potential fault interactions and consequences to evaluate the degree to
which these interactions influence the behavior of the Garlock fault.
3.3 The Garlock Fault
The Garlock fault is a major left-lateral strike-slip fault that extends eastward from its intersection with
the San Andreas fault for 250 km to the eastern termination of the fault at the southern end of Death
Valley (Figure 3.1), forming a prominent physiographic and structural boundary between the E-W
extension of the Basin and Range Province to the north and the non-extending Mojave block to the south
(Davis & Burchfiel, 1973). A total of 48-64 km of cumulative displacement has occurred along the
Garlock fault since its initiation (Davis & Burchfiel, 1973; Monastero et al., 1997; Smith, 1962; Smith &
Ketner, 1970), which occurred between 17 and 10 Ma (Andrew et al., 2015; Burbank & Whistler, 1987;
Loomis & Burbank, 1988; Monastero et al., 1997), and likely at ca. 11 Ma (Andrew et al., 2015; Blythe &
Longinotti, 2013). The Garlock fault has been divided into three segments defined by changes in strike
and structural complexities. Specifically, a two-to-three-km-wide step-over at Koehn Lake between the
western and central segments corresponds with a 10° change in strike, and a 15° change in strike near the
Quail Mountains between the central and eastern segments (McGill & Sieh, 1991).
Previous slip rate studies along the Garlock fault reveal similar long-term slip rates averaged over latest
Pleistocene and early Holocene time of ~5-8 mm/yr (Clark & Lajoie, 1974; Ganev et al., 2012; McGill et

43
al., 2009; McGill & Sieh, 1993). However, a recent slip rate study along the western and central Garlock
fault reveals significant (factor of 2 to 5 times) variations in slip rate spanning cycles of thousands of
years (~2 to 4 ky) and multiple earthquakes during latest Pleistocene and Holocene time (Fougere et al.,
2024). Specifically, incremental slip rates for the central Garlock fault for various time intervals since the
latest Pleistocene vary from ~3 mm/yr to ~14 mm/yr (Fougere et al., 2024). For example, Dolan et al.
(2016) documented an accelerated, late Holocene slip rate of >14 +2.2/-1.8 mm/yr from a series of 1.9 ka
alluvial fans that have been offset by ~26 m at the Christmas Canyon West site (~50 km east of Koehn
Lake), at least 2-3x faster than the long-term average rate. Additionally, Rittase et al. (2014) also
documented an accelerated, late Holocene slip rate of 10.8 to 12.5 mm/yr from an offset shutter ridge at
their Pilot Knob Valley site (Figure 3.1).
Although the Garlock fault has not generated any significant earthquakes during the historic period,
numerous studies have documented geomorphic and paleoseismic evidence for late Holocene earthquake
activity (e.g., Dawson et al., 2003; McGill et al., 2009; Madden Madugo et al., 2012; Kemp et al., 2016;
Gath and Rockwell, 2018; Pena, 2019; McGill et al., in prep). Three sites on the western segment of the
fault have documented evidence for past earthquakes. The western-most paleoseismic site, Campo Teresa
(CT), lies on the western segment of the Garlock fault, 25 km east of the intersection of the Garlock fault
with the San Andreas fault. This site, originally studied by LaViolette et al. (1980) and later by Gath and
Rockwell (2018) revealed two earthquakes in the last 1700 years and at least one (likely two) between
1700 years before present (yrs B.P.) and 4100 years yrs B.P. At the Twin Lakes (TL) site, which is
located ~20-km east of the CT site, Madden Madugo et al. (2012) documented evidence for two or three
earthquakes during the past 2550 ± 200 years. At the Clark Wash site, 40 km east of the TL site, McGill
et al. (2009) recorded evidence for at least two and possibly three prehistoric earthquakes, with the most
recent event occurring before the beginning of the twentieth century (Hanks et al., 1975) and after ca.
1460 CE. The third event, and possible second event, recorded at the CW site occurred before 2555 ± 205
yrs B.P.

44
On the central segment of the Garlock fault at the Koehn Lake (KL) site, Burke (1979) reported evidence
for 9-17 earthquakes constrained by a single radiocarbon age, recalibrated in this study using Oxcal
version 4.4 and the most up-to-date calibration curve (IntCal20, Bronk Ramsey, 2009; Reimer et al.,
2020), of 17530-18250 cal. years B.P. Madden and Dawson (2006) re-excavated the Burke (1979) trench
and documented seven to nine surface ruptures at the site since 10 ka, with three to five of these
earthquakes occurring since 4150 years ago. At the El Paso Peaks (EPP) site, located ~12 km east of the
KL site, Dawson et al. (2003) documented a cluster of four surface ruptures that occurred between 0.5 -
2.0 ka. Their data indicate that this cluster was preceded by an ~3-ky-long gap in evidence for surface
ruptures between 2.0 ka and 5.1 ka, which followed two older surface ruptures that occurred ca. 5.1 ka
and 6.8 ka. It is worth noting that at the EPP site, there was a period of reduced (but not zero) sediment
accumulation between 2.0-3.5 ka.
The easternmost paleoseismic record on the Garlock fault comes from the Echo Playa (EP) site, which is
located ~35 km east of the EPP site, where Kemp et al. (2016) recorded four events from ~3,500 years
ago. The most recent earthquake occurred between A.D. 1615 and 1820 and was preceded by the
penultimate event between A.D. 470 and 73, and an earlier event between A.D. 40 and 150. An earlier
earthquake was also recorded between 275 and 1950 B.C., which occurred during the period of apparent
seismic quiescence at the EPP site.
3.4 The Koehn Lake Paleoseismic Site
Koehn Lake is an ephemeral lake that is at present largely dry, occupying an area of oblique extension
caused by a left-releasing stepover in the Garlock fault that varies in width from 2 to 4 km, forming a
pull-apart basin between the western and central segments of the Garlock fault. The area of the pull-apart
basin, Fremont Valley, exhibits a flat-floored central basin bounded by the El Paso Mountains to the north
and to the south by the Rand Mountains (Dibblee and Minch, 2008). Primarily coarse-grained Quaternary
alluvial sediments derived from the Mesozoic granitic rocks of the El Paso Mountains and the Mesozoic

45
granitic rocks and Precambrian schists exposed in the Rand Mountains are deposited along the northern
and southern margins of the Fremont Valley (Figure 3.2a).
At the Koehn Lake site, the westernmost part of the central segment of the Garlock fault cuts through the
northeastern corner of the dry lake bed (Figure 3.2). To the north of Koehn Lake and the central Garlock
fault segment, the El Paso fault extends sub-parallel to the active Garlock fault for 26 km along the
southern front of the El Paso Mountains. This fault, however, does not show any geomorphic evidence of
Holocene displacement. In contrast, the central Garlock fault strand that extends along the northwestern
corner of the Koehn Lake basin displays abundant geomorphic evidence of Holocene displacements
(Clark, 1973; McGill and Rockwell, 1998; Madden and Dawson, 2006).
In this paper, we present a new paleo-earthquake record from a paleoseismic trench excavated near the
northeastern corner of the Koehn Lake just to the east of the Burke (1979) and Madden and Dawson
(2006) trenches (Figure 3.2). We located the trench site on a near-flat area at the northeastern corner of
the lake where the playa sediments interfinger with the distal portions of alluvial fans derived from the
north and northeast. These alluvial deposits are marked by shallow channel deposits at a variety of
orientations, but most exhibit flow towards the southwest. Although the site is nearly flat, as detailed
below, a central, fault-parallel graben that has opened between the two main strands of the Garlock fault
at the site does have subtle geomorphic expression, especially to the east of the trench site. Extending
eastward from the trench site, the central graben is bounded by oblique-normal left-lateral faults that have
very subtle geomorphic expression, with terrain to the south uplifted ~10-100 cm above the dry lake flat.
Previous paleoseismic trench investigations at the Koehn Lake site documented well-defined,
interfingering playa-margin and distal alluvial fan stratigraphy, with evidence for numerous well-defined
surface rupture event horizons (Burke and Clark, 1978; Burke, 1979; Madden and Dawson, 2006). In this
study, we returned to the Koehn Lake paleoseismic site to utilize a newer geochronological dating
technique, infrared-stimulated luminescence (IRSL; Rhodes, 2015), to better constrain the ages of

46
Garlock fault paleo-surface ruptures that extended through this site. To do this, we excavated a 37-mlong, 2.5-m-deep trench across the Garlock fault, ~20 m to the east of the paleoseismic trenches of Burke
(1979) and Madden and Dawson (2006) (Figure 3.2). To constrain event ages throughout the trench, we
collected sediment samples for IRSL analysis and detrital charcoal samples for radiocarbon analysis.
High-resolution photomosaics were generated for all faces of the excavated exposures by taking
photographs of the trench exposures from multiple angles to increase photographic overlap, using a
standard DSLR camera. The images were processed and stitched together using Agisoft Metashape
version 1.0.0.1 (Agisoft LLC, 2021) to produce point clouds and textured models that were transformed
into orthorectified photo mosaics. We used the photomosaics to map the exposed stratigraphy and faulting
to create logs for both walls of the trench where these features were directly mapped onto the photos in
the field.
3.5 Structure and stratigraphy of the trench exposures
The overall stratigraphy of the Koehn Lake paleoseismic trench is characterized primarily by an upwardcoarsening sequence of interbedded lacustrine-aeolian deposits that record the desiccation of pluvial
Koehn Lake, subsequent ephemeral stream flow along and across the Garlock fault, and deposition of
alluvium derived from the El Paso Mountains to the north. The top 0.5-1 m of the trench consisted of
highly friable silts, sands, and minor granule gravels that collapsed upon excavation, making
interpretation of stratigraphy and structural features near the top of the exposure difficult.
In the following, all locations in the trench are described by a distance in meters from the southern end of
the trench, which is designated as m 0. We refer to vertical distances in the trench using the system
implemented during trench logging, where positions above an arbitrarily defined “0 m” elevation depth
are denoted by a positive number (e.g., +0.5 m) and those below the 0 m elevation by a negative number
(e.g., -2.2 m). The location of any specific zone on the trench walls will therefore be denoted as the

47
horizontal measurements followed by the vertical measurement, separated by a comma (e.g., m 12.5, -
1.25 m).
Faulting was observed along most of the length of the trench. However, most strike-slip movement has
been concentrated on two major fault zones exposed in the trench at m 20-21 and m 25-26. Complete
mismatches of the stratigraphic sections juxtaposed by these faults in all by the shallowest section of the
trench exposure demonstrate that each of these two main strands of the Garlock fault has experienced
many meters of left-lateral strike-slip motion, effectively dividing the trench exposure into three separate
stratigraphic sections (Figure 3.3). Due to the resulting lateral complexity and extent of the stratigraphy
and structural features present within the trench, we describe these three sections, each of which exhibits
different, but internally consistent structural and stratigraphic features, separately. The three faultseparated packages exposed in the trench are, from south to north: (1) a southern section extending from
m 0 to m 20 m that is dominated by lacustrine clays, (2) a central section extending from m 20 to m 25
comprising a stratigraphic section of sediments deposited within a graben down-dropped between the two
main fault zones that exposes much younger strata, and (3) a northern section dominated by alluvium that
extends from m 25 to the northernmost end of the trench at m 38. Detailed trench logs can be found in the
Appendix B.
3.5.1 Southern section stratigraphy
The southern section of this trench, from m 0 to m 20, is characterized by very dark green-gray (almost
black on fresh exposure) lacustrine clays and silts that oxidize almost immediately (within an hour) to a
very pale gray color upon exposure that are overlain by a lake-drying sequence of thick lacustrine clays
interbedded with generally fine-grains (sands and silts) alluvium, which in turn is overlain unconformably
by fluvial channel cut-and-fill deposits (Figure 3.4). The basal, laterally extensive lacustrine deposits
comprise 2-25 cm thick, alternating, green-gray clay and silt units between -1.75 m and -2.5 m depth.
Interbedded silt beds comprise well-sorted grains, possibly deposited either by aeolian transport of mud-

48
cracked, desiccated sediments across the playa during drier periods, or deposition via distal, suspended
load flood deposits. The only organic material found within these lacustrine units was detrital charcoal. At
approximately -1.75 m, there is a sharp transition from green-gray lacustrine clays and silts upward to
clays alternating with red silt layers, indicative of a high-oxygen environment suitable for soil
development. The development of an organic-rich soil horizon nearby in the Mojave Desert at Little Dixie
Wash by ~10.7 ka (Rosenthal et al., 2017) matches the age of the IRSL sample (KL21-IRSL-52) collected
in the Koehn Lake trench at this horizon. An ~1-m-thick lake-drying sequence exposed between m 16 and
m 20.5, comprising interbedded clay, silt, and sand layers is preserved above this distinct boundary, These
units denote a change in the predominant depositional system from lacustrine to fluvial/alluvial, where the
lacustrine clay units exhibit increased spacing and reduced thickness. Between m 0 and m 16, a 0.5- to-1-
m-thick section of fluvial deposits postdates the drying of Koehn Lake. These deposits consist of crossbedded and laminated gravels and sands, and shallow, 1-4 m wide channel swales draped by cm-scale
clay and silt beds.
3.5.2 Central section stratigraphy
The central section of the trench between m 21 to m 25 is characterized by a 4.5-to-5 m wide, faultbounded graben that has been down-dropped. Specifically, the southern fault zone of the graben (m 20-
21) has experienced a maximum of 1.5 m of vertical separation from a depth of -0.5 m to the base of the
trench at -2 m. The northern fault zone of the graben (m 25-26) has experienced a similar 1.3 m of vertical
separation. The bottom part of the graben, below -1.75 m depth, preserves deformed lacustrine deposits.
Above the basal lacustrine unit, the graben is filled with alternating 25- to 50-cm-thick units of clay and
sand. The upper half of the graben above -1 m is filled with massive, faulted sandy and silty fluvial
deposits that post-date the pluvial lake, which were deposited unconformably above the alternating claysand units. Near the southern edge of the graben at m 21, the lacustrine and overlying clay-sand units
have been faulted to the extent that distinct sedimentary layers cannot be distinguished. Comparison of
similar strata exposed on both walls of the central section indicate that these strata dip ~15° eastwards.

49
This is consistent with the generally thicker stratigraphic section exposed at the same depths as that
exposed in the Burke (1979) and Madden & Dawson (2006) trenches 20 m to the west.
3.5.3 Northern Section Stratigraphy
The basal part of the northern section of the trench between m 25 and m 38 consists of lacustrine clays
and silts that are exposed below -2 m depth. These strata are tilted and dip ~25° to the south and many
exhibit soft-sediment deformation features (including convoluted strata, ball-and-pillow structures, and
pseudonodules). Most of the overlying northern section consists of fluvial and alluvial sand, silt, and
minor gravel deposits approximately 0.5-1 m thick that unconformably overlie the basal clay and silt
layers (Figure 3.6). Due to the proximity of the Koehn Lake site to the El Paso Mountains, these strata
likely represent distal alluvial fan sediments derived from the north and northeast. A massive, laterally
extensive, fine sand-to-silt unit deposited by sheet flows is exposed between +0.5 m and -2.25 m depth.
Within this massive fine sand-to-silt unit, two ~2-m-wide, southward-dipping, poorly sorted gravel to
coarse-grained sand channels are exposed between m 26 to m 31 approximately uniform in thickness with
a limited lateral extent.
3.6 Event Evidence
The main fault zone (MFZ) at the Koehn Lake paleoseismic site is expressed as two major fault strands
that bound a 5- to 6-m-wide central graben. As noted above, complete stratigraphic mismatches between
the sections bounded by these faults indicate that they have accommodated most of the left-lateral
displacement through the trench site. Additionally, minor faulting occurs throughout the northern and
southern sections of the trench and are less densely faulted than the MFZ. The lacustrine clay dried out
very quickly after exposure to air during excavation of the trench, making it difficult to identify faults
cutting through the lacustrine deposits (between -2.5 m to -1 m depth). At the southern end of the trench
(m 2 to m 5 on the west wall, and m 4 to m 7 on the east wall), a 3-m-wide zone of Riedel shears striking
at a 30° angle to the MFZ cuts the lower ~1 m of the southern section. The northern section of the trench

50
is also less densely faulted than the MFZ, however, it contains faults that have appeared to tilt multiple
beds toward the south by 10-20°.
Several lines of evidence were used to identify past surface ruptures in the trench, including upward fault
terminations, filled fissures, angular unconformities, growth stratigraphy, collapse features, and softsediment deformation structures. We document more than 30 event horizons throughout the two walls and
three sections of the trench. Each line of evidence is assigned a label based on the wall it was logged on
and an arbitrary assigned letter from a to u (e.g., WW-c). A detailed summary of all structural and
stratigraphic evidence of events documented at the Koehn Lake paleoseismic site can be seen in Table
3.1. Given that a single exposure is unlikely to record all evidence for an event, we attempted to combine
multiple lines of evidence from both walls and the different sections of the trench to establish a minimum
paleo-earthquake record at this site. Creating a minimum event record is important so as not to create
spurious events. Many event horizons we documented have age constraints that are too large to safely
assume they match a specific earthquake, or their age constraints overlap more than one event in the
minimum record. Therefore, we omit them from the preferred minimum paleo-earthquake record but we
want to note that it is possible that any of these events could potentially be another earthquake in the
record.
We determined the strength of evidence for each identified event by assessing the lines of mutually
consistent evidence by identifying: (1) multiple structural and stratigraphic features, (2) mutually
supportive evidence from both walls, and (3) recording of the same event on different structural blocks
within the trench (i.e., determining the lateral extent of the event horizon), and assigning a quality rating
of A (excellent), B (good), or C (fair) to each event. Events with excellent evidence of an earthquake
would include two or three of these factors and be assigned a rating of A. An event assigned a rating of C
has fair evidence and would only record one of these factors. We note that the extremely friable nature of
the upper 0.5-1 m of the trench made identification of event horizons within these sediments difficult,
resulting in a high likelihood of missing a few latest Holocene events at this site. In the following

51
sections, we describe the evidence for each well-defined surface rupture in the trench. Events are referred
to as KL1 (youngest event recorded at the site) to KL10 (oldest event).
Event 1 (KL1)
The most recent earthquake at this site (KL1) is defined by multiple fault strands on both walls that
terminate at the top of a gravel unit and the base or within the overlying friable sediments adjacent to the
southern edge of the graben (Figure 3.4). On the west wall in the southern section at m 19.9, +0.25 m, two
faults (F1 and F2) terminate either at the top of a gravel unit (S10) or extend upward through the basal
part of the friable sediments. Specifically, F2 at this location has experienced 15 cm of vertical separation
of unit S20. On the east wall in the central section, five faults (F3-7) terminate at the base or within the
friable sediments at m 20.3, +0.5 m depth, but any potential vertical separations are difficult to measure
due to this event horizon being located within one of the main fault zones (m 20-21). Based on a number
of upward terminations with associated moderate vertical separations, all of which terminated at the base
of the shallow, friable sedimentary section, together with the evidence being present on both trench walls,
we rate this event as quality A.
Event 2 (KL2)
Event 2 (KL2) is defined by two faults that terminate within a gravel unit in the central section of the
trench (Figure 3.5). These terminations are only seen on the western wall in the central section of the
trench at m 23.3, -0.25 m depth. Below these terminations, 1-2 cm vertical separations were observed on
faults F1 and F2. To be conservative, we define this event horizon by the first unbroken (and uncracked)
layer above the fault tips. Additional evidence for KL2 includes growth strata between m 23 and m 24.8,
and at -0.25 m depth, that were deposited above unit G70 as a 2-10-cm-thick, now-cemented sand layer
(G60) that pinches out to the north. Evidence for KL2 was not seen on the eastern wall because units G95
to G40 were likely deposited in a cut-and-fill event that was not laterally extensive. Based on two faults

52
that display distinct terminations within a gravel only documented on one wall and the overlying growth
strata, we rate this event as quality A/B.
Event 3 (KL3)
Event 3 (KL3) is defined by a 2-m-wide section exhibiting soft sediment deformation features on both
walls between m 22 and m 24 and in the lowest part of the graben, below -1.75 m depth. Specifically, cmthick, alternating layers of clay and silt exhibit plastic deformation along discrete shear zones (i.e., WW m
23, -2.25 m), as well as concave-upwards, bent layers of clay (i.e., dish structures) a few-cm wide,
separated by vertical “pillars” of silt (i.e., WW m 23.2, -2.15), and is overlain by a 0.25 m thick,
unsheared unit (G200). On the eastern wall, at m 22.5, -2.1 m, The top of unit G250 shows evidence of
plastic deformation, whereas the layers beneath appear to be faulted rather than plastically deformed.
More deformation has occurred near the edges of the graben compared to its center (i.e., m 23), likely as
the result of much more slip in multiple ruptures along the main fault strands that bound the graben.
Based on the soft-sediment deformation of the clay layers along discrete shear zones below a distinct
upward contact with overlying unsheared deposits, and the evidence being present on both trench walls,
we rate this event as quality A.
Event 4 (KL4)
Event 4 (KL4) is defined by multiple (at least 9) upward fault terminations that all extend upward to the
base of a channelized gravel deposit that exhibits two main channel thalwegs (at 26 m to 31 m). The
southernmost fault on the western wall at m 26.4, -2 m terminates at the deepest part of the channel.
Vertical separations documented along these upward terminating faults are <3 cm, some of which are
~0.5 cm. Bedding within both channels dips approximately 10-15° to the south, smaller than the dip of
underlying strata (N300 and N310). Based on a large number of upward terminations, all of which
terminated at the base of well-defined gravel layers, tilting of the gravel, alongside this event evidence
being present on both trench walls, we rate this event quality A.

53
Event 5 (KL5)
Event 5 is defined by a fluid-escape structure capped by a thin veneer of plastically-deformed clay.
Specifically, on the western wall a 15 cm-wide by 30 cm-tall dike exhibiting evidence of liquefaction of a
coarse to medium grained sand layer at m 26.7, -2 m. Within the dike, a medium-grain, 10-cm-diameter
spherical, sand body (i.e., pseudonodule) sharing similar characteristics to unit N250. The dike is bounded
by a fault on either side that warps and fragment the thin N245 clay unit. Additionally, two fault strands
terminate at the base of a sand unit (N210), and an additional strand terminates at the base of a 2-m long,
0.5 m deep channelized gravel deposit (N200). On the east wall, this event horizon is defined by ball-andpillow structures at m 25.5, -2.25 m. Based on the soft-sediment deformation features seen on both walls,
as well as evidence for multiple upward terminations, we rate this event quality A
Event 6 (KL6)
Evidence for event 6 (KL6) is defined by an angular unconformity between the top of a 20-cm-thick,
tilted clay unit (N300) at m 28.5 to 30 and -2 m depth exhibiting soft-sediment deformation and overlain
by a sand unit (N290), displaying growth strata. Convolute bedding is documented on both walls, at the
top of unit N300 at m 29.6 and -2.4 m on the west wall, and at m 32 and -2 m depth on the east wall. The
presence of undisturbed layers above and below these features indicates a seismic origin. The thickness of
N290 varies from approximately 50 cm south of unit N300 to 5-10 cm directly overlying it from a rapid
infilling of sedimentation post-earthquake. Multiple faults propagate upwards through this event horizon,
which has caused the clay units (N300 and N310) to dip 20° south from cumulative earthquakes.
Interestingly, KL4 overtop of KL6, has a smaller dip, indicating more events have tilted N300 and N310
than those that tilted the channelized gravel deposit (N150 and N160). Based on multiple lines of
evidence present on both walls, including soft-sediment deformed clay beds, tilted beds, and the overlying
growth strata, we rate this event quality A.
Event 7 (KL7)

54
Both lines of evidence for event 7 (KL7) come from the southern and the northern sections of the trench.
We believe that event horizon p (EH-p) from the northern section and event horizon t (EH-t) from the
southern section are the same event as they are both defined by soft-sediment deformation structures.
Event 7a (KL7a) (southern section)
Evidence for event 7b (KL7b) is defined by multiple soft-sediment deformation structures, as well as a
decreasing thickness of the unit underlying this event horizon. Specifically, this event horizon extending
from m 17 to m 21 on both walls (at -0.35 m depth on the west wall and between -0.1 and -0.25 m depth
on the east wall) is defined as the top of unit S80 as it contains significantly more convolute clay laminae
than units above (i.e., S55) and below (i.e., S100) this event horizon. Flame structures have formed in
clay laminae and extend into the sandy unit within unit S80, which can be seen on the west wall at m 19.5
and -0.4 m depth and on the east wall at m 20.6 and -0.3 m depth. Convolute laminae can be seen on the
west wall at m 18.9 and –0.45 m.
The southwards decrease in thickness of unit S80 indicates that this unit was deposited when the ground
surface at this location dipped somewhat northwards. At m 20.5 and -0.5 m depth, unit S80 is ~20 cm
thick, decreasing to a thickness of 2cm at m 16 and -0.3 m depth. Units above and below S80 also exhibit
a degree of thickening towards the north but to a much lesser extent than the unit underlying the event
horizon for KL7a. Unit S55 varies in thickness from 4 cm at 20.5 m to ~2 cm at m 17, and S100 varies in
thickness from 10 cm at m 20.5 to ~2 cm at m 17.
Event 7b (KL7b) (northern section)
Evidence for event 7b (KL7b) is defined by a ball-and-pillow structure seen on both trench walls.
Specifically, at m 29.5 and -0.22 m depth, there are several 10-to-15 cm-wide blocks containing the
internal structure of clay and silt from the surrounding material. The clay and silt layers comprising unit
(N300) overlying this event horizon are more continuous and less faulted than the unit below the event

55
horizon (N310). Multiple faults propagate through this event horizon indicator, some of which display
small (<2 cm) vertical separations of thin clay layers (e.g., at m 30 and -2.45 m depth), whereas other
faults display convoluted bedding. Based on the lateral continuity of this event documented on the
southern and northern sections, as well as both walls of the trench, and soft-sediment deformation
features, we rate this event quality A.
Event 8 (KL8)
Evidence for event 8 (KL8) is defined by multiple fissure fills within a laterally extensive clay layer
exhibiting a buttressing unconformity. Specifically, these fissures are located on the west wall at -0.5 m
depth and m 16.7, m 17.2, m 17.5, m 17.8, and m 18.3. The fissure fill located at m 17.8 on the west wall
is documented also at m 17.8 and -0.5 m depth. These fissures are filled with material from a thin (<2 cm)
veil of overlying silt, all underlain by at least one fault. These fissures range from 8 cm to 35 cm tall, and
2 cm to 15 cm wide.
The clay unit containing these fissure fill features (S130) varies in thickness, documented on both walls,
from 15 cm to 4 cm on the west wall, generally thinning towards the MFZ, with its greatest thickness
between m 17 to m 18.5. On the east wall, unit S130 varies in thickness from 8 cm to 4 cm between m
18.3 to m 20, where it tapers out at m 20 and -0.6 m depth. We observe that faulting during events can
cause topography on the relatively flat lake surface. On the east wall, the northernmost section of this
event horizon, from m 18.2 to m 17.5, has been eroded by a significantly younger gravel channel (S120),
which now overlies most of this event horizon. Additionally, one soft-sediment deformation structure was
identified on the east wall. Specifically, the base of clay unit S130 exhibits a convoluted clay layer at m
19.2 and -0.5 m depth. Based on the multiple fissure fills and buttressing unconformity documented on
both walls, as well as the soft-sediment deformation feature, we rate this event quality A.

56
Event 9 (KL9)
Evidence for event 9 (KL9) is defined by an angular unconformity between southward-dipping lacustrine
beds and an overlying clay unit. Specifically, unit S140 is unconformably overlain by unit S130 at m 16.5
to m 18 and -0.6 m depth on the west wall and between m 18 and m 20 and -0.5 m depth. The units (S140
and S150) underlying this event horizon comprise clay and silt laminations, dipping at ~10° towards the
south end of the trench, that interfinger with numerous <50 cm long gravel lenses. Based on the laterally
extensive angular unconformity documented on both walls of the trench, we rate this event quality A.
Event 10 (KL10)
Evidence for the oldest event (KL10) is defined by a large fissure fill feature within the lacustrine units in
the southern end of the trench. Specifically, a 1 m wide, 0.8 m deep fissure fill is documented on both
walls, between m 12 to 13 and -1.9 m depth on the west wall and between m 13.9 to 14.5 and -1.5 m
depth on the east wall. This event horizon is capped by a 5-10 cm thick reddish-brown clay layer and
underlain by alternating green-grey lacustrine clay and silt. This large fissure contains clay material from
the surrounding units, where minimal stratigraphy can be observed due to deformation. On the east wall,
the northern half of the fissure contains layers minimally displaced and warped compared to the south.
The fissure strikes at a 30° angle to the MFZ, and due to the extensional environment of the Fremont
Valley at Koehn Lake, we interpret this structure as a Riedel shear. Based on the large size of the fissure
fill being documented on both walls of the trench, we rate this event quality A.
3.7 Age Control
To constrain the ages of individual earthquakes, we dated post-IR infrared stimulated luminescence (postIR IRSL225) to date 50 sediment samples and radiocarbon analysis to date 47 detrital charcoal samples
collected from throughout the trench.

57
3.7.1 Luminescence Ages
The IRSL samples were collected from exposures without discernible cracks and younger disturbances
(e.g., bioturbation, faulting) by hammering 6-cm-diameter steel tubes horizontally into sand-rich target
strata located below and above paleo-earthquake event horizons documented in the trench exposures. All
luminescence samples were processed at the University of West Georgia, USA, and water content and
dose rates were determined. Although we collected samples for water content from the same holes as the
luminescence samples, in our age analyses we used a modeled water content that is more representative of
actual conditions over the lifetime of burial of the sediment, rather than simply the measured water
content. Specifically, these modeled values take into account the modern, measured water content as well
as the saturated water content calculated from bulk density and porosity of each sample. This approach
was used because much of the lacustrine sedimentary section was likely deposited under fully saturated
conditions that spanned a significant portion of the post-burial period. All samples were measured at the
University of Sheffield, UK, using a single-grain, regenerative-dose protocol using potassium feldspar
(Rhodes (2015), to provide an equivalent dose (De) estimate. We use overdispersion values between 15
and 25% based on clear signals that indicate modes of well-bleached grains (see Supplemental Material
for further discussion), and a fading correction factor of 1.12 for all samples. The resulting ages are
shown in Table 3.2.
3.7.2 Radiocarbon dating
We collected a total of 193 detrital charcoal samples encountered throughout the entire trench from as
many units as possible for radiocarbon dating. Samples were collected in glass vials with cleaned metal
tools. We chose 56 samples from key stratigraphic positions to date. All samples were prepared and
analyzed at the W.M. Keck accelerator mass spectrometer lab at the University of California, Irvine. We
visually analyzed each sample under a light microscope to remove any visible contaminants (e.g., roots,
surface dirt). The samples then underwent an acid-base-acid pretreatment protocol, and 47 samples

58
survived pre-treatment and combustion, yielding reliable ages. Radiocarbon ages were calibrated to the
IntCal20: Northern Hemisphere curve (Reimer et al. 2020) using OxCal 4.4 (Bronk Ramsey, 2009) results
of which can be seen in Table 3.3.
3.8 Koehn Lake paleo-earthquake chronology
To constrain the ages of Garlock fault paleo-earthquakes at the Koehn Lake site, we used OxCal v4.4
(Bronk Ramsey, 2009; Reimer et al., 2020) to model the paleo-earthquake record by incorporating all
available chronological constraints (i.e., both luminescence and radiocarbon) using a Bayesian approach
to trim probability density functions (PDFs) for sample age estimates and events. We observed that some
of the radiocarbon samples were older than other samples from the same or underlying deposits,
indicating that some of the detrital charcoal must have been reworked.Those ages that were clearly
reworked were not included in the age model. Similarly, for those sedimentary deposits within the trench
that contained multiple radiocarbon age estimates from the same horizon, we selected the youngest
detrital charcoal age(s) and omitted sample age estimates that exhibited clear pre-depositional history.
The structural and stratigraphic complexity of the trench, driven by the juxtaposition of distinct
sedimentary sections during left-lateral strike-slip along the two main strands of the Garlock fault, has
resulted in different sedimentary sequences observed in fault-perpendicular trench exposures. For this
reason, we build three initial sub-models for different sections of the trench (V1s – southern, V1c –
central, V1n – northern), each of which incorporates all relevant IRSL age estimates and event horizons in
stratigraphic order. Following initial statistical trimming of these ages using OxCal, we then ran a second
set of age models in which we incorporated all non-reworked, calibrated radiocarbon ages into a second
version of the age models (V2s, V2c, and V2n) to further constrain the ages of event horizons.
To create our preferred age model, we combine the V2 models for each section of the trench into a single
comprehensive minimum paleo-earthquake age model for the Koehn Lake site (V3). We also created an
additional model for event horizons identified throughout the trench from stratigraphic horizons that we

59
cannot directly correlate with well-documented event horizons elsewhere in the trench, incorporating
IRSL and radiocarbon data for these “floating” events (Vfloating).
3.9 Generation of preferred central Garlock fault paleo-earthquake age model
To generate our preferred model of paleo-earthquake recurrence at the Koehn lake site (V4), we must
consider the possibility that the paleoseismic record at the Koehn Lake site could be incomplete. There
are several factors that could lead to an incomplete record of surface ruptures at the Koehn Lake site. For
example, sediment accumulation rates and erosional processes are important to consider, as periods of
non-deposition can obscure the record, and erosion can remove evidence of past earthquakes. The highly
friable nature of the sediments exposed near the ground surface in the Koehn Lake trench have likely
resulted in an incomplete paleo-earthquake record over the past 2 ky, resulting in the possibility of
missing event horizons during this period. Comparisons to the EPP paleo-earthquake record reveal that
our preferred minimum KL paleo-earthquake record (V3; Table 3.4) is likely missing three younger
events. Dawson et al. (2003) recorded four events between A.D. 25 and 1640. Interestingly, our KL1
event, recorded at the base or within the friable sediments atop the entire trench, matches the timing of the
most recent event at EPP (between A.D. 1450 and 1640; Dawson et al., 2003). The erosion and deposition
processes that have occurred on the ground surface during this time has likely resulted in an incomplete
paleo-earthquake record, resulting in the high possibility of missing this cluster of events during this
period. However, in addition to our minimum paleo-earthquake model at Koehn Lake, we record several
anachronous event horizons with large age constraints. At least one of these event horizons (WW-r)
match the timing of the third and four events back in the EPP paleo-earthquake record, whereas the
penultimate event at EPP was not recorded at KL anywhere in the trench.
Using the assumption that the latest Holocene paleo-earthquakes recorded nearby at the EPP site ~10 km
to the east of the Koehn Lake over the past ~2 ky (Dawson et al., 2003) likely propagated to the KL site,
but were not recorded at the Koehn lake trench due to the very slow sediment accumulation rate and

60
highly friable nature of the sediments in the uppermost 25-50 cm, we have also generated a preferred
central Garlock fault paleo-earthquake model (V4) that includes all events recorded at both the KL and
EPP trenches (Table 3.5), allowing us to evaluate a more complete picture of events over the past ~11.5
ka.
Sediment accumulation rates differed between the three sections of the Koehn Lake trench. In the central
graben, sediment accumulation was relatively continuous between ca. 2.4 to 7 ka at 0.43 mm/yr (two
meters in 4600 years). Sediment accumulation in the southern section varied, with rates five times faster
during the period of lacustrine deposition between 8 to 14 ka of 0.38 mm/yr (2.25 m in 6000 years), than
the subsequent period of very slow sediment accumulation at 0.07 mm/yr between 8 and 0.5 ka (0.5
meters in 7500 years). Over the 8-14 ka period, shallow sub-aqueous sediment accumulation was
relatively continuous. Nearby records of paleo-climate proxies suggest that the end of the Pleistocene (ca.
11.7 ka) was marked by cool and moist conditions (Miller et al., 2010; Reheis et al., 2014; Kirby et al.,
2015), which likely ended between 7.8 and 7.4 ka (Kirby et al., 2015), seemingly around the transition
from relatively continuous sediment accumulation in the lacustrine section observed in the Koehn Lake
trench until ~8 ka.
Further supporting climate-controlled sediment accumulation rates is the occurrence of temporal clusters
of radiocarbon ages estimated from detrital charcoal. Specifically, the radiocarbon samples we dated
revealed ages in groups between 2.5 to 3.5 ka, 5 to 7 ka, and 8 to 10 ka. In contrast, there are minimal
radiocarbon samples from all sections of the trench that reveal ages that do not fall into the previous
groups, suggesting these periods are marked by reduced sediment transport. Additionally, we note a
sharply defined desiccation event marked by the uppermost lacustrine clay section, which could indicate a
short period of slower sedimentation accumulation. Sediment accumulation in the northern section of the
trench was made difficult to unravel due to tilting of beds, multiple cut-and-fill channels that eroded parts
of the section, and laterally discontinuous strata.

61
3.10 Discussion
3.10.1 Holocene paleo-earthquake recurrence pattern on the central Garlock fault
The ten surface ruptures documented at the Koehn Lake trench, combined with the El Paso Peaks record
of Dawson et al., (2003), reveals the occurrence of at least fourteen earthquakes along this stretch of the
central segment of the Garlock fault during Holocene-latest Pleistocene time. Combining these two paleoearthquake records allows us to examine the pattern of earthquakes using quantitative measures.
The Koehn Lake trench data reveal a complicated pattern of earthquake recurrence on the central Garlock
fault through time, highlighted by multiple periods of temporal clusters of four earthquakes. Overall, the
entire Holocene central Garlock fault paleo-earthquake record (Figure 3.7c) has an average earthquake
recurrence of 803 ± 552 (1σ) years. But this average includes what appear to be periods of less-frequent
earthquake recurrence. To more fully quantify the variability of earthquake recurrence patterns on the
central Garlock fault, we calculate the coefficient of variation (CoV) to test the aperiodicity (i.e., the
degree of periodicity in earthquake occurrence) of this record. For example, we identified four
earthquakes (KL6, KL7, KL8, KL9) that occurred between 8 ka to 10 ka during the early Holocene
(Cluster 3 in Figure 3.7). This temporal cluster shares similarities with the late Holocene temporal cluster
identified by Dawson et al. (2003) at nearby El Paso Peaks, during which four surface ruptures occurred
during a 1500-year-long period between 2.0 and 0.5 ka. Both of these clusters comprise four events over a
~2 ky period, and are both preceded by a single event ~1000 years prior to each cluster. In addition to
these four-event clusters, we documented a cluster of three earthquakes (i.e., KL3, KL4, KL5) was
documented at Koehn Lake during the mid-Holocene time between 6 to 7 ka (Cluster 2). Separating
Cluster 2 and Cluster 3 is an apparent seismic lull between 7 to 8 ka, similar to but significantly shorter
than the ~3 ky period separating the 0.5 to 2 ka cluster identified by Dawson et al. (2003) (Cluster 1) from
the 6-7 ka Cluster 2. It is apparent that this long (i.e., 3000-year) period separating Cluster 1 and 2 was

62
punctuated by at least one and most likely two earthquakes, making this a period of reduced earthquake
occurrence, rather than a complete cessation of surface ruptures along the central Garlock fault.
To incorporate the 95% uncertainty for each of our paleo-earthquake age estimates into the CoV
calculation, we calculated interevent times (N=13) as the absolute differences between consecutive
median ages, propagating uncertainties using the square root of the sum of squared errors for each pair of
events, and derived the mean, standard deviation, and coefficient of variation along with their propagated
uncertainties (see Supplemental Material), to yield a CoV of 0.69 ± 0.52. The average CoV calculated
from the median ages of paleo-earthquakes of ~0.7 is characterized as weakly periodic/strongly aperiodic
(0.5 < CoV < 1; Goes and Ward, 1994). However, the analytical uncertainties regarding paleo-earthquake
ages are large relative to the recurrence intervals, ranging from a minimum CoV value of 0.17 and a
maximum value of 1.21. Specifically, the three clusters identified have CoV values of 0.57 (Cluster 1;
460 ± 260 year average recurrence interval), 0.74 (Cluster 2; 640 ± 470 year average recurrence interval),
and 0.45 (Cluster 3; 395 ± 175 year average recurrence interval). These values imply that the central
Garlock fault earthquake recurrence behavior may range from “quasi-periodic” (CoV between 0 and 1) to
“clustered” (CoV > 1; Goes and Ward, 1994). Throughout the entire record, the occurrence of surface
ruptures on the central Garlock fault appears to follow a generally quasi-periodic pattern. However, the
record also reveals clusters of 3–5 earthquakes that deviate from this trend and may represent significant
periods of increased activity, juxtaposed with periods of lower earthquake recurrence.
3.10.2 Comparison of previous Garlock fault paleo-earthquake records
Comparison of the new event record from Koehn Lake on the westernmost extent of the central Garlock
fault with other paleo-earthquake records from the western and central Garlock fault segments allows us
to examine surface-rupture extents. Previous paleo-earthquake records on other sites along the Garlock
fault do not record any events older than 8 ka, so comparisons with the record presented here is only done
for the past 8 ky. The most recent event (ca. 460 years ago), was recorded at all paleoseismic sites along

63
the Garlock fault (Figure 3.8), suggesting this earthquake ruptured from its western extent at the
intersection with the San Andreas fault at least as far east as the Echo Playa site in a large (Mw<7.8)
earthquake (McGill and Sieh, 1991). A similar rupture occurred in the penultimate event, recorded at CT,
TL, possibly CW, EPP, possibly CCW, and EP. This event was not documented in the KL trench, likely
because of the difficulty in we encountered on identifying paleo-surface ruptures in the poorly exposed
uppermost 50 cm of the trench, but is inferred to have ruptured through this site because it is recorded at
sites to the west and east. Cluster 2 earthquakes identified in the central Garlock paleo-earthquake record
between 6 ka and 7 ka are possibly documented at sites east of Koehn Lake, including at El Paso Peaks
(EPP-F; Dawson et al., 2003) and Christmas Canyon West (CCW-8 and -9; Peña, 2019; Peña et al., in
prep.). From the combined paleo-earthquake records along the Garlock fault, it appears that this fault
commonly ruptures at least the western and central segments of the fault. We do not, however, have
enough resolution in the available paleoseismic records to identify possible partial ruptures of different
segments of the fault.
It is possible that these previous paleo-earthquake records are not complete, making it difficult to
correlate individual fault ruptures. For example, the EPP paleo-earthquake record (Dawson et al., 2003)
was likely not complete between 2-5 ka, during which no paleo-surface ruptures were observed. This ~3
ky period matches the timing in a decrease in sedimentation. Specifically, between 2 ky and 5 ky ago,
sedimentation in the EPP playa was being deposited at 5 mm/yr (1.5 m in 3250 years), compared with the
long-term sedimentation rate for this 7 ka record was 13 mm/yr (9 m in 7175 years). It is possible that this
period of slower sediment accumulation failed to capture any evidence of ground-rupturing earthquakes
during this period as at least one earthquake (KL2) was recorded at KL west of the EPP site during the 2-
5 ka apparent seismic lull period, which is similar to the timing of an event (E4 or E5) observed at the
CCW paleoseismic site to the east of the EPP site (Peña, 2019; Peña et al., in prep.).

64
3.10.3 Possible physical controls on clustering behavior in southern California
The new paleo-seismological data presented here add to the growing body of evidence that the Garlock
fault experiences millennia-long strain supercycles, during which the Garlock fault (GF) trades off
displacement with other faults in the southern California plate boundary (Dolan et al., 2007; 2016; Hatem
and Dolan, 2018). Specifically, these authors propose the nECSZ-GF-SAFm-SAFs fault network subsystem creates a closed kinematic loop acting as a mechanically efficient alternative to N-S dextral shear
in the ECSZ-Walker Lane-SAFs system within the southern Californian plate boundary alternating
between the two subsystems over periods millennia long. These individual faults within a sub-system
(i.e., SAFm, GF) experience periods of earthquake clustering followed by periods of much less frequent
earthquake activity during Holocene time, as shown by paleo-earthquake records presented in this paper
and by others (e.g., Rockwell et al., 2000; Dawson et al., 2003).
These fault behaviors could be controlled by changes in the relative strength of the major faults that
comprise the southern California plate boundary and/or changes to relative plate motion rates (Dolan et
al., 2016, Cawood and Dolan, 2024), and possibly also by associated with temporal changes in elastic
strain accumulation rates on individual faults. Gauriau and Dolan (2021) demonstrated the importance of
the basic control exerted by the relative structural complexity of a fault system in which a fault is
embedded by demonstrating that faults that lie within structurally complex plate boundary fault systems
exhibit more irregular slip behavior through time than isolated faults that lie within simple plate boundary
fault networks. Similarly, a comparison of geologic vs geodetic slip-deficit rates by Gauriau and Dolan
(2024) suggests that major strike-slip faults within complex plate boundary systems experience periods of
faster and slower elastic strain accumulation matching with changes in the rates of deformation of their
ductile roots.
Notably, earthquake clusters documented in this paper and by Dawson et al., (2003) match variations in
incremental slip rate documented along the central Garlock fault (e.g., Dolan et al., 2016; Fougere et al.,

65
2024). Specifically, the late Holocene (0.5-2.0 ka) cluster documented by Dawson et al. (2003) occurred
during a period of accelerated slip at ≥13 mm/yr, relative to the long-term slip rate of the Garlock fault of
~8 mm/yr documented by Fougere et al. (2024). Similarly, the mid-Holocene Cluster 2 (6-7 ka)
documented in the Koehn Lake trench occurred during a period of faster-than-average slip on the central
Garlock fault between 6 to 9 ka (Fougere et al., 2024). The older Koehn Lake Cluster 3 that occurred
between 8 and 10 ka falls partially within this period of faster slip, however, the incremental slip rate
record does not extend back beyond 9 ka. Therefore, additional incremental slip rate data from the early
Holocene-latest Pleistocene are needed to evaluate the relationship between Cluster 3 and the rate of slip
on the Garlock fault. These results suggest that the Garlock fault experiences periods of fast slip that span
multiple earthquakes, followed by periods of much slower fault slip and less frequent earthquake
occurrence. Interestingly, the slow periods do not reflect a complete absence of surface-rupturing
earthquakes, as originally suggested by Dawson et al., (2003) and reinforced by Dolan et al. (2016), but
rather record periods or reduced earthquake occurrence.
Notably, paleo-seismological records from the Garlock fault and Mojave section of the San Andreas fault
(SAFm) all document an event during the same period ca. 500 years ago, supporting a potential
mechanical linkage between these fault systems. Specifically, an event recorded at Pallett Creek and
Wrightwood on the SAFm occurred between 410-570 years ago (Scharer et al., 2011; Bemis et al., 2021),
displaying similar timing to the most recent earthquake on the central Garlock fault recorded at the KL
and EPP trenches (and possibly recorded at the Campo Teresa and Twin Lakes sites on the western
Garlock fault; Madden and Dolan, 2008; Madden Madugo et al., 2012; Gath and Rockwell, 2018) at 270-
650 years ago (Dawson et al., 2003; this paper). McAuliffe et al. (2013) investigated the likelihood of
triggering events on an adjacent fault by conducting Coulomb Failure Function (CFF) modeling using
various combinations of Garlock fault and Panamint Valley fault (PVF) ruptures, revealing significant
stress interactions between the two faults, and found that a PVF earthquake with magnitudes >7.27 will
increase Coulomb stress on the central Garlock segment, and at the same time decrease Coulomb stress

66
on the eastern Garlock segment (McAuliffe et al., 2013). More recently researchers have modeled stress
changes imparted by the Mw=6.4 and Mw=7.1 2019 Ridgecrest earthquakes on the Garlock fault (e.g.,
Ramos et al., 2020; Toda and Stein, 2020). Barnhart et al. (2019) used interferometric synthetic aperture
radar and satellite optical imagery to document surface displacements and subsurface fault slip
characteristics, demonstrating that static stress changes from the 2019 Ridgecrest sequence promoted slip
on the Garlock fault on the 20-25 km stretch of the central segment where surface creep was documented,
however, this triggered slip on Garlock was very shallow and likely involved fluids. Similar CFF
modeling of a magnitude 7.6 earthquake rupturing the SAFm showed this would likely increase CFF on
the western Garlock fault (McAuliffe et al., 2013). Similarly, Rockwell et al. (2000) present paleoearthquake ages from trenches along faults of the ECSZ in a compilation with previous investigations also
revealing ruptures through multiple faults (South Camp Rock, Old Woman Springs, Helendale, Lenwood)
occurring since ~1 ka.
3.11 Conclusions
This study enhances our understanding of earthquake recurrence on the central Garlock fault by extending
the record of paleo-earthquakes back to early Holocene–latest Pleistocene time. Our findings confirm that
periods of high earthquake recurrence (i.e., clusters) are separated by periods of less frequent earthquake
activity (i.e., lulls). The observed variability in earthquake recurrence strengthens the idea that the central
Garlock fault is not only driven by steady tectonic loading but is also influenced by static stress changes
resulting from interactions with nearby faults, which may contribute to sporadic earthquake recurrence
during lulls. The use of improved luminescence dating techniques allow us to refine the timing of paleoearthquakes, while radiocarbon constraints enhance our ability to construct a more precise chronology.
However, large analytical uncertainties present challenges in correlating paleo-earthquake events across
different records. Ultimately, the processes that control earthquake recurrence are highly complex and
interconnected, often producing patterns that can appear quasi-random. By documenting at least ten
surface-rupturing earthquakes at the Koehn Lake site over the past ~12,500 years, this study provides

67
crucial insights into the long-term behavior of the Garlock fault and its role within the broader fault
system of southern California. These findings underscore the need for continued research into the
mechanisms governing fault behavior to improve our ability to assess seismic hazard in tectonically active
regions.

68
CHAPTER 4 Paleo-Earthquake Evidence at Straw Peak Road from the Central Garlock Fault
4.1 Introduction
To add to the growing body of paleo-seismological evidence aiding understanding of the mechanisms
controlling temporal and spatial patterns of earthquake recurrence in southern California, this chapter
presents paleo-earthquake evidence from a paleoseismic trench excavated across the central Garlock fault
at the Straw Peak Road site (Figure 3.1). This chapter only presents paleo-earthquake evidence identified
on the Straw Peak Road trench exposures as the luminescence sample ages that aim to constrain the
paleo-earthquake record at the Straw Peak Road site in Pilot Knob Valley are pending at the time of
writing.
4.2 Straw peak road paleoseismic site
Pilot Knob Valley is an east-west-elongated depression located at the southern end of the Slate Range,
composed of deformed clastic and evaporitic rocks of Pliocene to Pleistocene age (Andrew et al., 2015;
Rittase et al., 2020). The Garlock fault cuts through the northern third of Pilot Knob Valley and is the
primary, through-going, left-lateral, strike-slip fault in the area. Another active strand of the Garlock in
the area is the Marine Gate fault, however, this fault is currently active as a north-side-up, high-angle fault
(Rittase et al., 2020), rather than accommodating left-lateral, strike-slip motion. At the boundary between
the Garlock cutting through a 2-km-thick package of Late Cenozoic strata (Rittase et al., 2020) in the
northern third of PKV, and the lower elevations of this valley, multiple drainages transport sediment
southwards where they are deposited as alluvial fans as the drainage opens up into the valley. These
alluvial fans emanate from multiple drainages along this section of the fault and coalesce with adjacent
fans.
In this chapter, I present paleo-earthquake evidence for large, surface-rupturing earthquakes from the
Straw Peak Road (SPR) paleoseismic trench located in Pilot Knob Valley. This stretch of the Garlock

69
fault is characterized by numerous deflected streams with “s-shaped” bends, coalescing alluvial fans
aggrading south of the fault from higher topography on the north side of the fault, and small linear
valleys. At the SPR trench site, the fault trace is relatively simple and linear, becoming double-stranded
~100 m eastwards of the trench site. The ground surface at the trench location has ~25 cm of elevated
surface topography, indicating this area has previously ruptured and has been filled in recently with
alluvium from the slope of the hill north of the fault at this site. This site was chosen because it has been
the locus of recent sedimentation, which could preserve the record of several past earthquakes on the
fault.
4.3 Paleoseismic trenching and trench stratigraphy
We excavated a 20-meter-long, 3-meter-deep trench oriented sub-perpendicular to the fault trace at the
SPR paleoseismic site (N 35°33.126', W 117°14.457'). This excavation was conducted to log stratigraphic
and structural features associated with Holocene to late-Pleistocene surface ruptures. To constrain the
ages of faulting events within the trench, 27 sediment samples were collected for IRSL analysis. Highresolution photomosaics of all exposed trench faces were created by capturing multiple overlapping
photographs using a standard DSLR camera. These images were processed and stitched together using
Agisoft Metashape version 1.0.0.1 (Agisoft LLC, 2021) to generate point clouds and textured models,
which were subsequently transformed into orthorectified photo mosaics. The photomosaics served as a
basis for mapping exposed stratigraphy and faulting. Logs for both trench walls were created by directly
mapping these features onto the photographs during fieldwork. In the following, all locations in the trench
are described by a distance in meters from the northern end of the trench, which is designated as m 0. We
refer to vertical distances in the trench using the system implemented during trench logging, where
positions above an arbitrarily defined “0 m” elevation depth are denoted by a positive number (e.g., +0.3
m) and those below the 0 m elevation by a negative number (e.g., -1.4 m).

70
The Straw Peak Road paleoseismic trench exposed Holocene to latest Pleistocene alluvial fan deposits,
characterized by multiple stratified and massive units that reflect varying times of fan deposition (Figure
4.1). Unit A, interpreted as very late Holocene in age due to its stratigraphic position near the ground
surface, is composed of loose silty sand, and weakly stratified. Unit B is a loosely packed, gray in color,
coarse and fine sand layer. This unit is massive in the upper half and exhibits stratification in the lower
half. Beneath this, Unit C comprises dense brown sand with granules, displaying local bedding features.
Unit D transitions into denser, gravelly sand with angular carbonate-coated clasts. It is weakly stratified,
with bedding dipping more steeply than the bedding in Unit C. Subunits of Unit D include D',
characterized by dense brown sand with thin vertical carbonate streaks, and D'', which features stratified
gravelly sand with carbonate splotches and carbonate-coated pebbles, both within the main fault zone.
The underlying Unit E comprises dense, well-stratified sand and gravel. It is divided into subunits, E' and
E'', that include horizontal carbonate-rich layers, though carbonate streaks are more continuous in E' than
in E''. Unit F, a continuation of the stratified sand and gravel sequence, is dense with minimal carbonate
content and slightly lighter in color compared to Unit E. Unit G is a dense debris flow with cobbles and
pebbles, characterized by carbonate-coated clasts. Unit H is composed of loose, coarse dark brown sand
and pebbles and is not a laterally continuous unit. Unit I is composed of moderately dense, grainsupported brown granules. Unit I contains pebbles in a silt matrix with a higher pebble content than the
underlying Unit J, primarily a 0.5-1-m thick silt layer interspersed with pebbles that dips toward the north.
The faulting observed within the trench is characterized by strands that dip between 45° and vertical
(90°), forming a 4-meter-wide main fault zone. The fault zone widens below a depth of -1.6 meters,
extending an additional meter on either side of the fault zone, resulting in a total width of approximately 6
meters. The fault zone exhibits a negative flower structure, indicative of combined extensional and strikeslip motion. Strike-slip displacement is primarily concentrated along the outer fault strands of the main
fault zone, notably between meters 10–11 and 13–14. These areas are marked by pervasively sheared

71
layers, demonstrating significant deformation within the fault system. Additionally, warped stratigraphic
layers, such as those within Unit E between meters 12–13 at a depth of -1.75 meters, are seen.
4.4 Interpretation of paleo-surface ruptures
EW-a
Located at m 10.2 and -1 m depth, Event EW-a is marked by a small fissure filled with gray sand, likely
sourced from unit A. A similar fissure, ~15 cm south of m 10 along the upper string line, also contains
gray sand. This event is interpreted as a very young seismic event, possibly the most recent event (MRE).
EW-b
Event EW-b spans m 10.5 to m 14 at depths of -1.25 m to -1 m, and is characterized by a fissure at m 11.3
and -1.2 m depth is filled with gray granules, from overlying unit B. Another fissure at m 13.5, between -
1 m and -1.25 m, contains sand and pebbles from unit B. Above this fissure, a dome-like structure (~5 cm
tall) with no internal stratigraphy is surrounded by horizontally laminated unit D. On the west wall, a
fissure fill (12–20 cm deep) composed of gray sand and small pebbles lacks bedding and is capped by an
angular unconformity with north-dipping sand and pebble layers. Cutting back the trench wall confirmed
the lateral continuity of features for at least 50 cm. Vertical separations along this horizon are
approximately 10 cm.
EW-c
Event EW-c, located between m 10.55 and m 11.15 at depths of -1.4 m to -1.25 m, is defined by a large
fissure fill at the base of a fault block of unit C. The fissure measures 25 cm deep and 15 cm wide and is
capped by unit C. An angular unconformity is also present, with bedding in unit D dipping more steeply
than in units C and B.

72
EW-d
Event EW-d is defined by a large fissure fill between m 12 and m 14 at a depth of -1 m. This fissure,
extending 30 cm deep, is filled with material derived from unit B and is three-dimensional in extent.
Vertical separations within the underlying unit D are less than 10 cm.
EW-e
Event EW-e is located between m 14.3 and m 14.8 at depths of -1.2 m to -1.25 m and is defined by
growth strata. Specifically, the top of a 20-cm-thick layer of alternating pebbles and coarse sand
containing carbonate clasts, which thins to 13 cm near a fault strand. Overlying this horizon is a
horizontally lying flat pebble layer. A granule layer above the horizon pinches out against a younger fold,
indicating localized deformation.
EW-f
Event EW-f extends from m 14.2 to m 15.22 at depths of -1.3 m to -1.45 m and is defined by an upward
fault termination, observed at m 14.3, -1.35 m, with overlying folded layers of alternating pebbles and
coarse sand. A liquefaction feature is present at m 14.6, extending from -1.45 m to -1.55 m depth.
Additionally, fault terminations at the base of a pebble unit are documented between m 14.15 and m 14.5
at -1.68 m depth.
EW-g
Event EW-g is characterized by an angular unconformity observed between m 14 and m 15 at depths of -2
m to -1.6 m. Unit H has been tilted approximately 10° to the north. Unit G unconformably overlies this
tilted unit.

73
EW-h
Event EW-h is characterized by a small colluvial wedge of Unit H, observed between m 9 and m 10 at
depths of -1.5 m.
EW-i
Event EW-i is characterized by a tilted Unit J compared to the overlying unit, observed between m 11 and
m 13. A fissure is filled with Unit H at m 12.2, at a depth of -2.5 m.
4.5 Future Work
In total, nine event horizons were identified from the trench exposures (Figure 4.1). However, direct age
constraints are currently unavailable, as infrared-stimulated luminescence (IRSL) dating results are still
pending. Without these age estimates, the timing and sequence of events cannot yet be fully constrained.
The Straw Peak Road site represents the easternmost paleoseismic site along the central Garlock Fault.
Establishing chronological constraints for the identified events will be crucial for determining whether a
rupture observed at the Koehn Lake site (Chapter 3) extends further east. This information will provide
valuable insights into the spatial distribution of Garlock fault ruptures.

74
CHAPTER 5 Combined Incremental Slip Rate and Paleo-Earthquake Records: An Earthquake-byEarthquake Record of Slip for the Garlock Fault
This chapter is based on the following manuscript in preparation:
Fougere, D., Dolan, J., Rhodes, E., and McGill S. (in prep.). Combined Incremental Slip Rate and PaleoEarthquake Records: An Earthquake-by-Earthquake Record of Slip for the Garlock Fault.
5.1 Abstract
Some plate boundary faults maintain relatively constant slip rates over time (e.g., Alpine fault, New
Zealand), while others exhibit significant temporal variations in slip behavior across multiple earthquake
cycles (e.g., Dead Sea Transform). Understanding these variations is critical for fully determining how
plate-boundary slip is partitioned among major faults in both time and space. However, few records exist
that allow a thorough examination of the “true” dated path of slip for an individual fault. The Garlock
fault, a major left-lateral strike-slip fault, is one such fault that demonstrates irregular slip behavior. In
this study, we present a detailed slip history for the central Garlock fault by integrating the updated
incremental slip rate record presented in this chapter with paleo-earthquake ages from Chapter 3. This
combined dataset generates one of the most comprehensive earthquake-by-earthquake records for any
major strike-slip fault globally. Our findings reveal a near-continuous record of surface-rupturing
earthquakes on the central Garlock fault extending back to the latest Pleistocene (~12 ka). Notably, the
results highlight distinct variations in slip behavior, with periods of accelerated slip accommodating
approximately 25 meters of displacement across clusters of four earthquakes, alternating with phases of
slower slip and reduced earthquake recurrence, contributing less than 10 meters of cumulative slip. These
data provide key insights into the temporal and spatial patterns of fault slip in complex plate boundary
systems, shedding light on the dynamic processes that influence seismic hazard in southern California.

75
5.2 Introduction
Documenting patterns of fault slip through time and space on major plate boundary faults is fundamental
for understanding the mechanisms governing fault behavior and seismic hazard assessment. While some
faults exhibit relatively constant slip rates over time (e.g., Berryman et al., 2012; Gold et al., 2011; Kozacı
et al., 2009; Noriega et al., 2006; Salisbury et al., 2018), others show significant temporal variations in
slip rates spanning multiple earthquake cycles (e.g., Dolan et al., 2007, 2016; Friedrich et al., 2004; Gold
& Cowgill, 2011; Hatem et al., 2020; Wallace, 1987; Weldon et al., 2004; Zinke et al., 2017, 2019, 2021).
Incremental slip rate records provide crucial insights into these temporal variations, helping to constrain
the time and displacement scales over which non-constant behavior occurs and illuminating the physical
mechanisms driving these variations. However, these insights remain data-limited due to the scarcity of
detailed incremental slip rate records that combine earthquake timing and displacements to generate a
comprehensive “dated path” of fault slip.
The Garlock fault, a major left-lateral strike-slip fault in the southern California plate boundary network,
exemplifies the importance of such records. Past studies of the Garlock fault have documented long-term
incremental slip rates, revealing irregular earthquake recurrence and temporal variations in slip behavior
(e.g., Dawson et al., 2003; Gath & Rockwell, 2018; Madden & Dawson, 2006). Currently, there are too
few comprehensive data sets of paired incremental fault slip rates and paleo-earthquake ages from fault
systems to allow the comparisons necessary to fully determine how plate-boundary slip is partitioned
amongst major faults in time and space. In this study, we present a detailed record of slip for the Garlock
fault by combining the updated incremental slip rate record with paleo-earthquake ages for the central
Garlock fault from Chapter 3 (Fougere et al., in prep), possibly one of the most comprehensive
earthquake-by-earthquake records for any major strike-slip fault globally, providing key insights into the
temporal and spatial patterns of fault slip in complex plate boundary systems.

76
5.3 New slip rate data from the central Garlock fault
To strengthen the current incremental slip rate record for the western and central Garlock fault (Figure
5.1), we use field mapping and analysis of high-resolution (0.5 m) GeoEarthScope lidar data collected
along the Garlock fault (data available at www.opentopography.org) to accurately measure tectonically
offset geomorphic features, combined with age estimates using post-IR infrared stimulated luminescence
(post-IR IRSL225) protocol (Rhodes, 2015) and radiocarbon dating of detrital charcoal. We document nine
new and refined slip rates from multiple sites on the central Garlock fault (Figure 5.2), adding to the
existing four slip rates [Clark Wash (Chapter 2), Christmas Canyon West (Dolan et al., 2016), Summit
Range East (Chapter 2), Summit Range West (Chapter 2)], and the age and single-event displacement of
the most recent event, significantly extending the record to ca. 12 ka (Table 5.1). The following sections
will describe each slip rate site from the westernmost presented here, the Koehn Lake Berm site on the
westernmost part of the central segment, to the easternmost site included in the incremental slip rate
record, the Pilot Knob Valley East site on the easternmost part of the central segment.
5.3.1 Koehn Lake berm
The Koehn Lake berm (KLb) site is located at the northeast corner of Koehn Lake, an ephemeral lake that
is at present largely dry, occupying an area of oblique extension caused by a left-releasing stepover in the
Garlock fault that varies in width from 2 to 4 km, forming a pull-apart basin between the western and
central segments of the Garlock fault, on the westernmost extent of the central segment of the fault
(Figure 5.3). The area of the pull-apart basin, Fremont Valley, is a floored, central basin surrounded by
the El Paso Mountains to the north comprising Mesozoic granitic rocks of pre-Tertiary to Cretaceous,
possibly late Jurassic age, and to the south by the Rand Mountains comprising Mesozoic granitic rocks
and the Rand Schist of Precambrian to Paleozoic age (Dibblee and Minch, 2008). Primarily coarsegrained Quaternary alluvial sediments are derived from the El Paso Mountains to the north, and the Rand
Mountains to the south of the basin are deposited in the Fremont Valley.

77
The KLb site, first identified by Clark (1973), and studied further by Madden and Dawson (2006,
unpublished), comprises an offset 1.8 km-long gravel berm. The berm was partially excavated in the early
1900s, resulting in the ridge crest of the berm north of the fault to no longer exist. However, the ridge
crest is preserved south of the fault. The offset of the Koehn Lake berm was measured by projecting the
estimated position of the ridge crest north of the fault to the mapped ridge crest south of the fault, yielding
a measurement of 90 +10
-5 m. The errors for this measurement include the widths of the excavated area
north of the fault.
To estimate the age of deposition for the Koehn Lake berm, three sediment samples were collected and
dated for IRSL analysis (KL22-IRSL-01, -02, and -03) from beach sands within the berm, as well as two
radiocarbon samples (KL22-C14-03 and -05) that were collected close together in the same stratum, a few
meters southeast of the IRSL sample location. The three IRSL samples, once corrected for fading using a
uniform 12% fading correction (Dolan et al., 2016; Rhodes, 2015), yielded ages of 2820 ± 350 years
before 2024 [yb2024 (KL22-IRSL-01)], 12450 ± 970 yb2024 (KL22-IRSL-02), and 9060 ± 750 yb2024
(KL22-IRSL-03). The deepest IRSL sample collected from the berm, KL22-IRSL-03, yielded a more
consistent age estimate (i.e., clear signals that indicate modes of well-bleached grains) of 12480 ± 920
yb2024 when a higher (25%) overdispersion was used in the IRSL calculation, rather than the typical
15%. Radiocarbon dating of KL22-C14-03 and KL22-C14-05 yielded calibrated ages of 14500-16250
years B.P. and 13330-16790 years B.P., respectively. The Koehn Lake berm formed when wave energy
was great enough to construct this feature, and because these conditions persisted over a finite period, the
berm has likely experienced multiple erosion and redeposition events. Therefore, the radiocarbon ages
above represent the maximum age of the berm construction, since the radiocarbon samples all have ages
greater than the IRSL age estimates, they were likely reworked. Therefore, I use an age of 12450 ± 970
yb2024 from KL22-IRSL-02 as the representative age of the final construction and abandonment of the
berm. Using the preferred offset value and uncertainty measurements of 90 ± +10
-5 m based on the
projected ridge crest north of the fault to the mapped ridge crest south of the fault, and the age estimate of

78
12450 ± 970 yb2024 as a representative age of the construction of the Koehn Lake berm, yields a slip rate
for the Koehn Lake berm of 7.2 ±1.3
0.9 mm/yr.
5.3.2 Searles Lake shoreline
Searles Lake is part of the Owens River system, a large, connected lake system that used to drain almost
the entire eastern Sierra Nevada and extended to Lake Manly in Death Valley. Intermittent Searles Lake
had occupied Searles Valley since ~3.2 Ma (Smith et al., 1983) up until the early Holocene (Bacon et al.,
2006; Smith, 2009). A late Pleistocene highstand shoreline of Searles Lake crosses the central Garlock
fault in many places. The Searles Lake Shoreline (SLS) site, originally studied by McGill and Sieh
(1993), is where this shoreline is best preserved (Figure 5.4). At this site, an abrasion platform and sea
cliff have been cut by wave action into older alluvium and older lacustrine sediments (McGill and Sieh,
1993). During the most recent highstand between 10 to 13.8 ka, approximately 0-2 m of lacustrine sands
were deposited on this abrasion platform and pinch out against the sea cliff. The shoreline angle, the angle
between the intersection of the seacliff and the abrasion platform, is left-laterally offset along the two
fault strands. The highstand sands have been buried by 0-1 m of colluvium, requiring McGill and Sieh
(1993) to excavate 20 trenches to locate and precisely map the shoreline.
The northern strand at the SLS site was determined by McGill and Sieh (1993) to be offset 42-46 m, with
a preferred offset of the maximum 46 m based on the trend of the shoreline likely modified by warping
near the fault (McGill and Sieh, 1993). The southern strand has been offset 37 ± 1 m, based on the
projection of the shoreline angle determined from trenches north and south of the southern strand (McGill
and Sieh, 1993). Just south of the northern strand is a zone of many closely spaced faults that left-laterally
offset the shoreline angle by 3-14 m, labeled the north central shear zone (Figure 5.4), the offsets in this
zone were measured using 3D trenching methods by McGill and Sieh (1993). The minimum offset of 3 m
represents slip on discrete faults within the north-central shear zone, whereas the maximum offset
measurement is based on possible warping of the shoreline due to faulting. A total offset across the two

79
sub-parallel fault strands as well as the north central shear zone is 82-106 m, with a preferred value of 90
m.
No direct age measurements were taken by McGill and Sieh (1993). Rather, the age of the offset was
inferred based on the age of the most recent lake highstand documented nearby in the Mojave desert
(Stuiver and Smith, 1979; Benson et al., 1990). However, beach sands exposed in the SLS trenches are
ideal for post-IR-IRSL225 dating. To directly constrain the age of the deposition of this shoreline, multiple
IRSL samples were collected from a re-excavated trench (southernmost 5 m of Trench 5 in McGill and
Sieh, 1993) and a hand-excavated pit (LOC20-01) between the fault strands to sample buried beach sands.
Six sediment samples were collected for IRSL analysis from the re-excavated Trench 5 (labeled Trench
21 from here on), two samples (SLS22-05 and -06) were collected from the overlying colluvium, and four
samples (SLS22-01, -02, -03, and -04) were collected from the underlying unit of well-sorted, coarsegrained sand. The age estimate for the deepest sample in the trench (SLS22-04) is secure at 12,500 ±
1,100 yb2024, and is used as the preferred age of the construction of the shoreline angle. Age estimates of
the shallower samples collected from the trench show an in-mixing of younger material (early to midHolocene). For example, sample SLS22-02 has two age populations, one at 12,000 ± 700 yb2024, like the
deepest trench sample age estimate, and a much younger population at 5300 ± 900. Additionally, some
IRSL samples likely also recorded age estimates of older material, presumably residual material from
previous highstands during cool wet episodes. For example, the two samples collected within the
colluvium have populations of wildly different ages, much older than the underlying beach sands. Three
sediment samples (GF20-01, -02, and -03) were also collected for IRSL analysis from a hand-excavated
pit (LOC20-01) between the fault strands. All three samples were collected within beach sands in a
vertical sequence between 22-48 cm depth. Age estimates for the samples are as follows, GF20-01 yields
an age estimate of 12,200 ± 1,300 yb2024, GF20-02 yields an age estimate of 12,300 ± 1,500 yb2024, and
GF20-03 yields an age estimate of 13,500 ± 1,200 yb2024.

80
The slip rate of the offset shoreline angle is calculated using the offset measurement across the two fault
strands of 90 ± 16
8 m, and using three different age estimates of construction of the shoreline angle for the
Garlock fault at this site. First, the oldest age and its uncertainty from Trench 21 (SLS22-04) of 12,500 ±
1,100 yb2024 yields a slip rate of 7.2 ± 2.0
1.2 mm/yr. Additionally, the combined ages and associated
uncertainties of the three samples collected at LOC20-01 (GF20-01, GF20-02, and GF20-03) have a
combined age of 12,745 ± 1525, yielding a slip rate of 7.1 ± 2.3
1.4 mm/yr. Lastly, I combine the deepest
sample age estimates (intersection) from Trench 21 (SLS22-04) and LOC20-01 (GF20-03), yielding a
combined age of 13,000 ± 1650 yb2024 to calculate a slip rate of 6.9 ± 2.3
1.4 mm/yr. All slip rates
calculated using various samples are very similar, with the slip rate values within 0.3 mm/yr of each
other, and share similar uncertainties, therefore, the preferred slip rate for the SLS site is based on the
tightest age constraint from SLS22-04, 12,500 ± 1,100 yb2024, and the cumulative offset across both
fault strands of 90 ± 16
8 m, yielding a preferred slip rate of 7.2 ± 2.0
1.2 mm/yr.
5.3.3 Pilot Knob Valley west
The Pilot Knob Valley west (PKVw) site is located at the southern end of the Slate Range (35.56°, -
117.2°; Figure 5.5). The PKVw site comprises a large alluvial fan complex along an 800-m stretch of the
central Garlock fault, comprising several terraces that get progressively older and higher in elevation
towards the east (Qal2-Qal7). The majority of the Slate Range is composed of pre-Cenozoic granitic and
metamorphic rocks, while the southern flank directly north of the Garlock fault comprises PlioPleistocene non-marine and Quaternary alluvium. Sediments from the Slate Range are transported
southwards and deposited south of the fault as alluvial fans and terraces. Along this stretch, the Garlock
fault is particularly well-defined in the geomorphology as a linear fault trace with well-defined left-lateral
offsets of southward-flowing drainages, and numerous shutter ridges. Drainages are labeled from D1
(westernmost) to D8 (easternmost; Figure 5.5). The north side of the fault has been uplifted and the
modern wash has incised deeply (>20 m) into the oldest alluvial fan surface (Qal8). Many shutter ridges
and offset channels are located throughout this extensive fan complex.

81
5.3.3.1 PKVw-A
This site, originally studied by Rittase et al. (2014), is located west of the modern drainage (D2; 35°
33.5985'N, 117° 12.16428'W). This site comprises late Holocene alluvial fans sourced from an abandoned
wash (D1, Figure 5.6). Here, Qal3b mapped by Rittase et al. (2014) is mapped as two vastly different
aged surfaces, Qal4a and Qal2. The eastern edge of Qal4a is defined by a 0.8-0.9 m tall riser that
separates this surface from the younger Qal2 surface. Qal4a has recognizable remnant bar and swale
topography, but incisions and pavement surfaces dominate, whereas Qal2 has preserved bar and swale
topography. Here, a ~50 m long shutter ridge has been progressively offset, buttressing alluvial deposits
against the northern edge of the shutter ridge. The modern channel (D2; Figure 5.6) flows south along the
eastern edge of the toe of the shutter ridge.
The offset measurement for PKVw-A is based on restoring the riser between Qal4a and Qal2 with the
eastern edge of the shutter ridge. Due to possible lateral erosion of the shutter ridge toe at its eastern
extent as it protrudes into D2 has likely been subject to erosion, therefore, this estimate is a minimum.
This measurement assumes that the Qal4a was deposited when some length of the shutter ridge protruded
into the drainage. Rittase et al. (2014) suggested a possible larger offset of 43-50 m based on restoring the
eastern end of the shutter ridge with the channel wall north of the fault and accounting for 7 m of lateral
erosion. However, as seen in the incremental slip rate record in Figure 5.6, this large offset of 43-50 m is
exceedingly unlikely considering all slip rate evidence presented. Specifically, the SRE and PKVe-D slip
rates have experienced similar offsets (38-46 m) since ca. 6 ka, whereas the large 43-50 m offset would
have had to occur since ca. 3.5 ka.
We excavated two pits into Qal4a and one pit into Qal2 to collect sediment samples for IRSL analysis.
From Qal4, we collected two samples from pit LOC20-03 (PKV20-01 and -02) and two samples from pit
LOC20-05 (PKV20-07 and -08). From Qal2, we collected three samples from pit LOC20-04 (PKV20-03,
-04, and -05). Sample ages from the Qal4a deposit yielded estimates of 3520 ± 190 yb2024 (PKV20-01),

82
3490 ± 160 yb2024 (PKV20-02), 2370 ± 120 yb2024 (PKV20-07), and 4090 ± 220 yb2024 (PKV20-08).
We calculate a preferred age estimate for Qal4a by combining PKV20-01 and PKV20-02 ages and their
uncertainties, yielding a combined age of 3505 ± 125 yb2024. These samples reveal a faster
sedimentation period than those samples collected from pit LOC20-05. Sample ages from the Qal2
deposit yielded estimates of 590 ± 40 yb2024 (PKV20-03), 760 ± 110 yb2024 (PKV20-04), 670 ± 70
yb2024 (PKV20-05). We calculate a preferred age estimate for Qal2 by combining PKV20-03, PKV20-
04, and PKV20-05, yielding a combined (union) age of 675 ± 45 yb2024.
Assuming Qal4a was deposited after some amount of offset of the shutter ridge, an upper terrace
reconstruction is used to calculate a slip rate of 8.3 - 10.9 mm/yr (PKVw-A) based on an offset
measurement of 30-37 m and a maximum age of Qal4a deposition of 3505 ± 125 yb2024. This scenario
requires Qal4a to be a degradational feature (i.e., fill-cut terrace).
5.3.3.2 PKVw-C
An older fan surface (Qal6b) is located at 35° 33.618'N, 117° 12.029'W (Figure 5.7). The Qal6b surface is
0.7-1.2 m higher in elevation than Qal5. Separating Qal5 and Qal6b is a 1-2 m deep drainage flowing
southwards across the fault from D3. North of the fault, D3 has been deflected by an 18-m-long shutter
ridge. A younger possible terrace has inset into Qal6b has formed in the channel separating Qal5 and
Qal6b. The offset measurement for PKVw-C is based on the restoration of the Qal6b western riser to the
eastern edge of D2, yielding a preferred offset of 95 m. A minimum offset of 92 m was determined based
on sedimentologically allowable geometry for these features. A maximum offset of 98 m was determined
by accounting for the width of the channel between Qal6b and Qal5.
Two natural exposures were cleaned off on the western riser of Qal6b, located south of the inset terrace in
D3 south of the fault. Four sediment samples for IRSL analysis were collected, two from LOC22-01
(PKV22-01 and -02) and two from LOC22-02 (PKV22-03 and -04). PKV22-01 and PKV22-02 were
collected within a red-brown sand unit from the exposure, overlying this is a grey-colored unit comprising

83
pebbles within a finer matrix containing a layer of carbonate cemented layers directly overlying the redbrown unit. Underlying the sampled unit is a layer of calcrete. PKV22-03 and PKV22-04 were collected
within a unit containing poorly sorted fine-to-cobble grain-sized sediment. IRSL analysis yielded ages of
13500 ± 1000 yb2024 (PKV22-01), 13700 ± 1200 yb2024 (PKV22-02), 8500 ± 900 yb2024 and 18600 ±
1300 yb2024 (PKV22-03), and 17100 ± 1200 yb2024 (PKV22-04). The combined (union) age of PKV22-
01 and PKV22-02 is used to define the age of Qal6b abandonment, yielding a preferred age of 13610 ±
770 yb2024. Age estimates for PKV22-03 are not included as this sample displays two populations of
ages.
The slip rate for PKVw-C is an upper terrace reconstruction based on an offset measurement of 95 ± 3 m
and an estimated age for Qal6b abandonment is 13610 ± 770 yb2024, yielding a slip rate of 7.0 ± 0.5
mm/yr.
5.3.3.3 PKVw-E
The PKVw-E – site, ~250 m east of PKVw-C, where most of this fan complex is composed of Qal6, with
Qal6b being the most extensive surface (extends laterally for ~350 m) with multiple alluvial fan lobes
emanating from the multiple drainages along this stretch of the fault (Figure 5.8). North of the fault, two
drainages (D5 and D6) flow southwards until they are deflected 90-170 m east by a large (~200-m long)
shutter ridge. The merged D5+D6 channel flows southeast where it has likely eroded some amount of the
northeastern corner of the shutter ridge. The coalesced channel flows south, incising deeply into an older
alluvial fan (Qal6c) by ~7 m. Qal6b and Qal6c surfaces are separated by a 1 to 2 m deep abandoned
channel that is incised through the shutter ridge from D5. Evidence for uplift towards the eastern end of
the PKVw site can be seen by a wind gap within the shutter ridge that formed from 2.2 m of cumulative
vertical displacement.

84
The offset measurement for PKVw-E is based on restoring the thalweg of the incised drainage between
Qal6b and Qal6c south of the fault to D5 north of the fault, yielding an offset measurement of 86 ± 2 m.
The widths of the channel north of the fault are incorporated into the uncertainty of this measurement.
Age estimates of the Qal6c alluvial fan are from three skrivelled IRSL samples (PKV20-12, -13, -14)
collected in a hand-excavated pit (LOC20-10) dug into the western riser of the eastern portion of the
Qal6c fan. LOC20-10 revealed a layered fan structure comprising three depositional events with varying
ages that get older with depth. IRSL analysis yielded ages of 9390 ±1500 yb2024 for PKV20-12 at 20-25
cm depth, 28310 ± 2140 yb2024 for PKV20-13 at 50-60 cm depth, and 33980 ± 2200 yb2024 for PKV20-
14 at 90-105 cm depth. The age constraint for PKVw-E is based on the age of the shallowest sample,
PKV20-12, likely a younger depositional event yielding an age of 9390 ±1500 yb2024. However, it is
possible that the age of incision between Qal6b and Qalc is slightly older than this (<13 ka) due to
extensive paleoclimate records in the region recording the end of a pluvial (e.g., Bacon et al., 2006; Orme
and Orme, 2008; Hoffman, 2009; Rosenthal et al., 2017). A minimum age estimate of 9.4 ka and
maximum estimate of 13 ka is used for the PKVw-E offset based on past research on the end of the last
major pluvial. Within LOC20-10, PKV20-13 likely samples the base of the Qal6c alluvial fan, and
PKV20-14 may sample an older (e.g., Qal7) alluvial fan composed of entirely colluvium.
The slip rate for PKVw-E is based on a thalweg offset by 86 ± 2 m occurring since 9.4-13 ka, yielding a
slip rate of 6.5-9.4 mm/yr.
5.3.4 Pilot Knob Valley east
The Pilot Knob Valley east (PKVe) site is a younger alluvial fan complex less than 2 km east of the
PKVw site, spanning an ~200 m stretch of the central segment of the fault (Figure 5.9). The fan complex
sourced from the modern channel flowing along the western edge of the terraces gets progressively older
and higher towards the east. Five terrace surfaces are mapped south of the fault and east of the modern
drainage (Qal0), from youngest to oldest, Qal1, Qal2, Qal3, Qal4a, and Qal4b. South of the fault, Qal1 is

85
a small wedge of recently abandoned alluvium separated from the modern channel by a 0.4 m tall riser
and exhibits vegetation as well as fresh bar and swale topography. The upstream side of the fault has
been uplifted and eroded more than the area south of the fault, however, risers north of the fault are still
noticeable and are used to correlate with risers south of the fault. North of the fault only two surfaces are
present, Qal3 and Qal4. West of the modern drainage, a remnant of Qal3 has inset into an older alluvial
fan, Qal7, which has had its eastern edge subsequently eroded where the drainage crosses the fault. Qal7
is much older than the alluvial fan complex east of the modern drainage, as seen by its much darker
appearance in satellite imagery.
At the PKVe site, the progressively higher terrace risers are used as piercing points to measure
displacement from the modern channel. The smallest offset measurement at this site, PKVe-A, is based on
the restoration of a 0.7-m-tall riser between Qal1 and Qal2 (also between Qal0 and Qal2 farther
downstream) south of the fault to the eastern edge of the modern drainage yielding 8 ± 1 m, uncertainty
based on the width of the riser. PKVe-B is based on the restoration of the projection of the <0.6 m tall,
west-facing riser separating Qal2 from Qal3 south of the fault, to the eastern edge of the modern drainage,
yielding a measurement of 19 ± 3 m. The uncertainty is based on the minimum and maximum projections
of the Qal2-Qal3 riser south of the fault, as the riser becomes wide and shallow near the fault. PKVe-C is
based on the restoration of the 0.5 to 1 m tall, west-facing riser of Qal4a south of the fault, to the westfacing riser of Qal4 north of the fault, yielding a measurement of 30 ± 2 m. The uncertainty is based on
the minimum and maximum projections of risers south of the fault. The largest offset, PKVe-D, is based
on the restoration of the easternmost edge of Qal4 south of the fault to with the east-facing riser of Qal4
north of the fault, giving a potential maximum offset since Qal4 was deposited, yielding a measurement
of 43 ± 3 m. The uncertainty is based on the minimum and maximum projection of the Qal4 east-facing
riser north of the fault. It appears that there has been ~2 m of uplift when comparing the elevation of the
alluvial fans north and south of the fault.

86
A total of seven pits were excavated into all relevant terraces at this site to collect sediment samples for
IRSL analysis. One IRSL sample (PKV19-05) was collected from Pit 13 from the Qal2 terrace, yielding
an age estimate of 860 ± 100 years for Qal2. Five IRSL samples were collected from Qal3, two north of
the fault from Pit Qal3N (PKV14-12 and PKV14-13), and three collected from the Qal3 terrace south of
the fault (PKV14-03, PKV14-04, and PKV14-05). North of the fault, both samples were collected at 42
cm depth, yielding age estimates of 2850 ± 330 yb2024 (PKV14-12) and 2010 ± 150 yb2024 (PKV14-
13), when combined, yields an age estimate for Qal3 north of the fault of 2160 ± 280 yb2024. Three IRSL
samples were from pit Qal3E collected in a vertical sequence south of the fault at 18 cm, 43 cm, and 68
cm depth. The shallowest sample (PKV14-03) yielded an age of 1920 ± 170 yb2024, the middle sample
(PKV14-04) yielded an age of 1620 ± 190 yb2024, and the deepest sample (PKV14-05) yielded an age of
1860 ± 130 yb2024. Combining these age estimates yields an average age of 1830 ± 180 yb2024 for the
deposition of Qal3 south of the fault. A pit (Qal4N) was excavated into Qal4 north of the fault and three
IRSL samples were collected in a vertical sequence. The shallowest sample (PKV14-09) was collected at
34 cm depth and yielded an age estimate of 4700 ± 410 yb2024. PKV14-10 was collected at 38 cm depth
and yielded an age estimate of 3160 ± 360 yb2024. The deepest sample (PKV14-11) was collected at 55
cm depth and yielded an age estimate of 5120 ± 450 yb2024. Combining these age estimates yields an
average age of 4160 ± (+390/-450) yb2024 for the deposition of Qal4 north of the fault. South of the fault,
the Qal4a surface appears to be a younger lobe of deposition associated with Qal4. Pit Qal4W was
excavated into Qal4a where two IRSL samples were collected at 36 cm depth (PKV14-01) and 66 cm
depth (PKV14-02), yielding age estimates of 3930 ± 330 yb2024 and 3990 ± 540 yb2024, respectively.
Combining these two age estimates yields an average age of deposition for Qal4a of 3940 ± 570 yb2024.
Three IRSL samples were collected from the easternmost lobe of Qal4 in a vertical sequence from Pit
Qal4E. The shallowest sample (PKV14-06) was collected at 24 cm depth and yielded an age estimate of
4740 ± 370 yb2024. The next sample collected at 38 cm depth (PKV14-07) yields an age estimate of
5680 ± 360 yb2024, and the deepest sample collected at 64 cm depth (PKV14-08) yields an age estimate
of 5570 ± 380 yb2024. The shallowest samples from pits Qal4N and Qal4E both reveal a ~4.7 ka

87
deposition event. For the age of Qal4, the age estimates of the two deepest samples likely represent this
terrace's average age of deposition. Combining PKV14-07 and PKV14-08 yields an age estimate of 5620
± 525 yb2024.
For the following slip rate calculations, an upper-terrace reconstruction model is used that assumes no
lateral erosion of the displaced risers. The slip rate for PKVe-A is based on 8 ± 1 m of slip occurring in
860 ± 100 yb2024, yielding a slip rate of 9.3 ± 1.6 mm/yr. The PKVe-B slip rate is based on a 19 ± 3 m of
slip occurring in 1830 ± 180 yb2024, yielding a slip rate of 10.2 ± 1.9 mm/yr. The slip rate for PKVe-C is
based on 30 ± 2 m of slip occurring in 3940 ± 570 yb2024, yielding a slip rate of 7.6 ± 1.2 mm/yr. The
slip rate for PKVe-D is based on 43 ± 3 m of slip occurring in 5620 ± 525 yb2024, yielding a slip rate of
7.6 ± 0.9 mm/yr.
5.4 Updated incremental slip rate record for the Garlock fault
A Bayesian framework for computing incremental fault slip rates is used to calculate the incremental slip
rate between each site, where age and displacements have been recorded to a finite uncertainty (Figure
10a). Specifically, the Rejection sampling for Incremental Slip Rate calculation (RISeR) program of
Zinke et al. (2017, 2019) is used to establish “geologically allowable” constraints on fault slip rate
behavior, such as assuming that the fault did not slip backward at any point in its history by discarding
values resulting in negative slip rates. Assuming the Garlock fault hasn’t slipped backward during its
history (i.e., right-laterally), we place tectonostratigraphic inferences on “geologically allowable” fault
slip behavior. RISeR’s Markov Chain Monte Carlo (MCMC) formulation uses the inverse transform
sampling method (see Zinke et al., 2017; 2019) to randomly sample the PDFs of age and displacement
measurements. A Bayesian condition is applied to the outputs to enforce the constraint of no negative slip
rates (e.g., Gold and Cowgill, 2011; Zinke et al., 2017; 2019). Slip rates are reported first as percentiles of
the viable picks and then based on analysis of a pseudo-continuous nonparametric function (PDF)
determined from the picks (Zinke et al., 2017; 2019). The results are output as PDFs (Figure 10b).

88
Our preferred model (Figure 10c) is based on (1) a combination of the incremental slip rate record shown
in Figure 10a and 10b, (2) the ages of the 14 paleo-earthquakes identified along the central Garlock fault
at Koehn Lake and El Paso Peaks, and (3) assumptions of single-event displacements. First, we note the
average displacement in the 14 earthquakes is ~6.4 m (85-90 m cumulative displacement), a value
consistent with measurements of single-event displacements in the most recent four earthquakes measured
on the central Garlock fault. Specifically, McGill and Sieh (1991) and Burns (2023) measured singleevent displacements along the central segment of 4.75 to 6.25 m at El Paso Mountains in the past four
earthquakes, and 4.3-7.3 m in the past three earthquakes at Pilot Knob Valley, with the most recent event
having a single-event displacement of 6.5 m. Using an average displacement of 6.5 m based on the
incremental slip record presented in this chapter, and the assumption that this displacement has been
relatively consistent through time through latest Pleistocene-Holocene time on the central Garlock fault.
We construct the preferred incremental slip record in the past 15 earthquakes on the central Garlock fault
shown in Figure 10c. In Figure 10c, we use time gaps between cluster 2 and cluster 3 (~1.2 ky-long gap)
as a zero slip constraint during that time interval (7.2-8.1 ka). Similarly, we use the absence of
earthquakes between cluster 1 and GF5 as a period of zero slip. We assume that potential slip in the
earthquakes constraining the beginning and endings of zero slip periods where the average displacement
is ±1.5 m (i.e., 5-8 m). Although it is possible that the single-event displacements could have been smaller
(but not likely larger) than the inferred 3 m displacement range, the consistency of displacements
observed in the past 4 earthquakes argues that slip has been relatively constant event to event.
5.5 Discussion
5.5.1 Correspondence of earthquake clusters with periods of faster-than-average slip
The new incremental slip rate record for the central Garlock fault indicates an average total displacement
of ~95 m since ca. 12.5 ka, yielding an average slip rate for this segment of the fault of ~7.5 mm/yr,
similar to the longest-term slip rate. The 14 published and newly presented incremental slip rates, when

89
combined with the timing of the past 14 earthquakes along the same section of the Garlock fault, reveal
periods of faster-than-average slip on the Garlock fault that match the timing of clusters of earthquakes
recorded at the Koehn Lake and El Paso Peaks paleoseismic trenches (Chapter 3) spanning the past ca. 12
ky. Specifically, the coincidence of periods of slip acceleration and clusters of earthquakes seen in the
youngest part of the record (i.e., 0-2 ka), is also observed in the combined record between 5 to 7 ka, as
well as between 8 to 10 ka (Figure 10c).
5.5.2 How much slip occurs per cluster?
The combined incremental slip rate and paleo-earthquake record reveals a near earthquake-by-earthquake
record of surface-rupturing events for the central Garlock fault extending back to the latest Pleistocene.
Interestingly, this record displays periods of faster-than-average slip on the Garlock fault encompassing
~20-25 m of total slip (during 4 earthquakes), whereas slower periods only encompass <10 m of slip in 0-
1 earthquakes (Figure 10d). Specifically, between 0 to 2 ka, the fault accommodated 26 m of slip during a
cluster of four earthquakes, coinciding with faster-than-average incremental slip at a rate of 13.6 +1.4/-1.2
mm/yr calculated between age-displacement pairs in Figure 10a (MRE and CCW). Between 2 to 5 ka, the
fault experienced slower-than-average slip at a rate of 3.2 ± 0.4 mm/yr, 3-5x slower than the younger,
succeeding period (0-2 ka) calculated between the time of the age-displacement pair for the MRE (Figure
10a) and the ca. 1.9 ka 26 m offset documented at the CCW site (Dolan et al., 2016). Notably, during this
period of slower-than-average slip, the Garlock fault only experienced one ground-rupturing earthquake,
which accommodated 7 ± 3 m of slip. From ~5 to 8 ka, the fault experienced another period of fasterthan-average slip at a rate of 12.4 +3.4/-2.4 mm/yr, calculated between the ages of the SRE and SRW
dated offsets, coinciding with ~25 m of cumulative fault slip. The period between 5 and 7 ka is marked by
a cluster of four earthquakes which collectively accommodated the 25 m of slip during this period of
faster-than-average slip. Preceding this cluster was a 1-ky-long period between 7 to 8 ka, with no paleoearthquakes recorded, during which we infer that the fault did not slip (i.e., a slip rate of 0 mm/yr).
Between 8 to 12 ka, the fault was slipping at an incremental rate of 6.6 +0.9-0.8 mm/yr, similar to

90
previous published longer-term slip rates documented in previous slip rate studies along the western and
central Garlock fault segments have recorded long-term (103) rates of 5-8 mm/yr (Clark and Lajoie, 1974;
McGill et al., 2009; Ganev et al., 2012), and the long-term slip rate for the Garlock fault calculated in this
chapter of 7.5 mm/yr. In the younger half of this 8-10 ka period, a cluster of four earthquakes was
recorded, accommodating a total of 25 m of slip in 2 ky. Between 10 and 12 ka, the fault experienced at
least one (likely two earthquakes), accommodating ~12 m of fault slip in 2 ky. The paleo-earthquake
record only extends back to 10.5 ka, however, so we have no information about the timing of earlier
earthquakes. Based on the incremental slip rate record, however, the total average amount of slip (~95 m)
recorded at multiple sites (KLb, PKVw-C, and SLS) along the fault requires ~5-10 m of additional slip,
which likely occurred in only one earthquake during the period from ca. 11.5 ka to ca. 13.5 ka. If this is
true, this period was marked by less frequent earthquake recurrence, comparable to the period of slow slip
on the Garlock fault between 2 to 5 ka.
5.5.3 Spatial observations
Previous studies along the Garlock fault suggest that slip along the fault decreases eastward (e.g., McGill
and Sieh, 1991). However, the similarity in age-displacement measurements from various sites along the
western and central Garlock fault segments, as well as individual event displacements inferred from the
paleo-earthquake record, show although the slip rate of the eastern segment of the Garlock fault is much
slower than to the west (Crane, 2014), this is not the case on the western and central segments of the
faults. For example, multiple geomorphic features on the central segment of the fault have experienced
displacements of 90-95 m since ca. 12.5 ka. Specifically, the age-displacement measurements for the
Koehn Lake berm site (90+10-5 m, 12,450 ± 970 yb2024) on the western end of the central segment is
commensurate with measurements from Searles Lake Shoreline (90+16-8 m, 12,500 ± 1100 yb2024) and
Pilot Knob Valley west (95 ± 3 m, 13,610 ± 770 yb2024) from the eastern end of the central segment, 55
km, and 65 km to the east of the Koehn Lake berm offset, respectively. Similarly, 65-70 m displacements
dated at ca. 8 ka have been documented from the Clark Wash site on the eastern end of the western fault

91
segment and Summit Range West from the middle of the central segment. These observations suggest the
incremental slip rate record is relatively similar along the central segment and at least the eastern end of
the western segment, within age limits of a few hundred years and displacement ranges of <5-10 m.
5.5.4 Fault interactions
The most recent event (MRE) on the Garlock fault is recorded at multiple paleoseismic sites along both
the western and central segments of the Garlock fault (Burke, 1979; Dawson et al., 2003; Madden and
Dawson, 2006; Madden and Dolan, 2008; McGill et al., 2009; Madden Madugo et al., 2012; Gath and
Rockwell, 2018; Peña, 2019; Peña et al., in prep). Since the MRE is seen at many sites, and based on the
large (~5-8 m) displacements inferred from geomorphic offsets to be a result of slip in the MRE (McGill
and Sieh, 1991; Burns, 2019), this was likely a single large-magnitude event that likely ruptured both the
central and western segments of the Garlock fault.
The MRE recorded on the Garlock fault occurred ca. 480 years ago matches the timing of a surface
rupture recorded on the Panamint Valley fault in the nECSZ at 330-485 cal. yrs ago (McAuliffe et al.,
2013), as well an event ca. 475 years ago recorded at many paleoseismic sites (i.e., Frazier Mountain,
Elizabeth Lake, Pallett Creek, Wrightwood, Pitman Canyon) along the Mojave section of the San Andreas
fault (Seitz et al., 1997; Weldon et al., 2004; Akçiz et al., 2010; Scharer et al., 2011; 2017; Bemis et al.,
2021). These data, showing near-simultaneous events that ruptured at least part of the NECSZ, the
western and central segments of the Garlock fault (GF), and the Mojave section of the San Andreas fault
(SAFm), allow for the possibility of a coordinated multi-fault rupture. If this was a multi-fault rupture, it
could have caused an Mw > 8.0 earthquake (assuming >550 km in rupture length and a seismogenic depth
of 15 km; Zeng et al., 2022).
More likely than a Mw > 8.0 multi-fault rupture, however, is that these three faults ruptured separately
during a brief period of time. The similar timing of events on such kinematically disparate faults suggests
the influence of regional fault interactions. The east-west Garlock fault intersects multiple faults,

92
including the San Andreas fault, and well as several large right-lateral ECSZ faults. It is worth noting,
however, that the available age constraints on the timing of past events on all of these faults precludes
distinguishing a possible large multi-fault rupture or ruptures of the individual faults in sequence.
Earthquakes on any of these faults can influence how stress is redistributed along the fault after an
earthquake along the Garlock fault, as suggested by McAuliffe et al. (2013). To explore regional
interactions between faults within the southern California plate boundary fault network, McAuliffe et al.
(2013) used Coulomb failure function change (ΔCFF) models to model changes in static stress on a fault
imposed by an event on a nearby fault. Specifically, ΔCFF modeling suggests that an event on the
Panamint Valley fault in the NECSZ has the potential to increase the Coulomb failure stress on the central
Garlock fault, which in turn could cause the triggering of a large event (Mw ≤7.8) on the Garlock fault
(McGill and Sieh, 1991; McAuliffe et al., 2013). Similarly, the western Garlock fault has been shown to
be driven by conjugate slip with the SAF to accommodate north-south shortening and east-west extrusion
from the mechanically inefficient strike of the SAFm relative to plate-boundary motion (Hill and Dibblee,
1953; King et al., 2004; Stuart, 1991; Hatem and Dolan, 2018). These results demonstrate that regional
fault interactions documented between the SAFm-GF-NECSZ subsystem could explain the SAFm,
nECSZ, and Garlock fault MRE ca. 475 years ago.
5.5.5 Driving mechanisms
The data presented here, including ~26 m of cumulative displacement in cluster 1 (0.5–2 ka) during a
four-earthquake cluster, ~25–30 m of slip during four-earthquake cluster 2 (5–7 ka), and ~25 m during
four-earthquake cluster 3 (8–10 ka), add to a growing body of evidence from strike-slip faults around the
world suggesting that displacements during such fast periods are all relatively similar at 25 ± 5 m (e.g.,
Weldon et al., 2004; Dolan et al., 2016; 2024; Zinke et al., 2017; 2019; 2021; Hatem et al., 2020; Fougere
et al., 2024). These data place firm constraints on how much slip occurs during fast periods, which in turn
places constraints on what processes might be controlling the observed fast-slow behavior of faults within

93
complex regional fault systems (Dolan et al., 2016, 2024; Gauriau and Dolan, 2021, 2024; Cawood and
Dolan, 2024). For example, processes that may operate at these displacement and time scales include
alternating strengthening and weakening fault-zone rocks within either the upper seismogenic or lower
ductile crust (Dolan et al., 2007; 2024; Mildon et al., 2022; Cawood and Dolan, 2024).
To change the relative strength of mechanically complementary faults (e.g., SAFm-GF-NECSZ), the
driving mechanisms must operate over time and displacement scales larger than those of an individual
seismic cycle. One explanation for this is strain hardening and annealing processes, where the ductile
roots of a fault system harden during periods of rapid slip (e.g., during an earthquake cluster) at rates that
temporarily overwhelm the counteracting annealing processes leading to a lull in lower crustal ductile
shear, and subsequently lowering upper crustal strain accumulation and earthquake occurrence (Dolan et
al., 2007; 2016), or possibly a feedback mechanism in which coseismic slip introduces fluids into the
ductile roots of fault zones (Oskin et al., 2008). Recently, Cawood and Dolan (2024) proposed a possible
model for the strength evolution of a fault zone, explaining the potentially reversible strengthening and
weakening microstructural mechanisms that could account for the behavior of alternating fast periods of
slip spanning multiple earthquakes and slow periods of few earthquakes and little fault slip. However, it is
difficult to determine the exact mechanism responsible for temporal clusters of earthquakes and many of
these processes may be mechanically complementary.
5.5.6 Probabilistic seismic hazard analysis
These data have important implications for probabilistic seismic hazard analysis (PSHA). Specifically,
our identification of periods of faster-than-average slip occurring during clusters of 4-5 earthquakes, and
the accommodation of ~25 m of slip in each fast period suggest that these values may provide a potential
path forward to better estimate time-dependent, near-future hazard. For example, if 25 m of slip on a fault
reflects an underlying process, we may be able to infer from paleoseismology that after a cluster of
earthquakes that resulted in ~25 m of slip the fault has completed a fast mode and is now entering a slow

94
mode. As an example from the Garlock fault incremental slip rate record, and following the previous
reasoning, after 0.5 ka, we suggest that the Garlock fault may have entered a slow mode following the
four-event cluster between 0.5-2 ka during which the fault accumulated about 26 m of slip. This inference
is also supported by the apparent very slow accumulation current rate of elastic strain accumulation on the
Garlock fault (~0-3 mm/yr; Evans et al., 2016; Gan et al., 2000; McClusky et al., 2001; Meade & Hager,
2005; Miller et al., 2001; Peltzer et al., 2001; Savage et al., 1981, 1990, 2001), relatively to our 12.5 ka
average rate of ~7.5 mm/yr This pattern has been observed on other strike-slip faults (e.g., Gauriau and
Dolan, 2024). Slow geodetic rates on faults that may have recently completed a cluster of earthquakes and
~25 m of slip may be a hallmark of faults that have recently entered a slow mode. These faults might then
be reasonably inferred to have a lower-than-average near-future probability of earthquake recurrence.
5.6 Conclusions
The combined incremental slip rate and paleo-earthquake record presented here reveal a near earthquakeby-earthquake record of surface-rupturing events for the central Garlock fault spanning back to the latest
Pleistocene (ca. 12 ka). These data provide key insights into the temporal and spatial patterns of fault slip
on a major strike-slip fault embedded in a tectonically complex plate boundary fault system. Specifically,
the results highlight distinct periods of variability in slip behavior, with periods of faster-than-average slip
spanning clusters of four earthquakes accommodate ~25 m of slip alternating with periods of slower-thanaverage slip and reduced earthquake recurrence accommodating <10 m of slip. The identification of
periods of faster-than-average slip occurring during clusters of four earthquakes, and the accommodation
of ~25 m of slip in each fast period suggest that these values may provide a potential path forward to
better estimation of time-dependent, near-future hazard.

95
CHAPTER 6 Conclusions
The Earth's surface is shaped by dynamic tectonic plates that continuously shift and interact, with faults
serving as the primary structures accommodating the motion between plates. Faults can exhibit diverse
behaviors, some faults slip in large, frequent earthquakes such as the Alpine fault, a major plate boundary
fault in New Zealand that experiences large Mw 7.0-8.0 earthquakes approximately every 300 years
(Berryman et al., 2012). Conversely, many other plate boundary faults experience irregular recurrence
intervals between earthquakes, such as the Jordan Valley section of the Dead Sea transform fault (Marco
et al., 1996). Recently, Gauriau and Dolan (2021) demonstrated that fault behavior is influenced by the
complexity of the surrounding fault system in which an individual fault is embedded. Understanding how
faults store and release energy over time and space, as well as how they interact within a mechanically
complementary fault system is necessary for understanding the mechanisms controlling these behaviors
and accurately assessing seismic hazard. In southern California, irregular fault behaviors have been
observed on fault systems accommodating strain within the Pacific-North America plate boundary. For
example, extreme (3x) changes in slip rate spanning multiple earthquake cycles have been observed on
the Mojave section of the San Andreas fault (SAFm) at the Wrightwood site (Weldon et al., 2004), as
well as temporal clusters of earthquakes identified on western Mojave eastern California shear zone
(ECSZ) faults by Rockwell et al. (2000). Additionally, most models using all of the available geodetic
data suggest there are persistent discrepancies between geologic and geodetic rates in southern California
(e.g., Evans, 2016). Specifically, the Garlock fault reveals geodetic rates much slower (~0-3 mm/yr) than
the ~7.5 mm/yr averaged over ca. 12 ka, suggesting strain accumulation on the Garlock is currently
slower than in the past. Conversely, Peltzer et al. (2000) suggested that elastic strain accumulation rates
along the westernmost part of the ECSZ are currently much faster than its long-term average rate, and
20+ years after this research was published, the 2019 Ridgecrest ruptured near a location of concentrated
elastic strain accumulation identified by Peltzer et al. (2000) in synthetic aperture radar data. To better
characterize these fault behaviors, I have documented a detailed record of slip for the Garlock fault by

96
combining the updated incremental slip rate record with paleo-earthquake ages for the central Garlock
fault, possibly one of the most comprehensive earthquake-by-earthquake records for any major strike-slip
fault, offering new insights into the temporal and spatial patterns of fault slip within a complex plate
boundary system.
The new and refined incremental slip rate record presented in this dissertation enhances and bolsters the
previously sparse mid-to-early Holocene record of slip along the Garlock fault. The average long-term
slip rate for the central Garlock fault suggests an average displacement of ~95 m since approximately
12.5 ka, yielding an average slip rate of ~7.5 mm/yr, aligning with previous long-term slip rate estimates.
Interestingly, incremental slip rates for the Garlock fault varied by a factor of two to five-fold during
Holocene-latest Pleistocene time, confirming that the period of rapid slip documented since ca. 1.9 ka by
Dolan et al. (2016) is a common pattern of behavior for this fault. Periods of accelerated slip are separated
by seismic “lull” periods where the fault slipped significantly slower relative to the “fast” periods. In
addition to the temporal behavior of the fault, this dissertation suggests that different segments of the
Garlock fault (western vs. central) have similar incremental slip rates averaged over the same time
intervals. Previous studies along the Garlock fault suggest that slip along the fault decreases eastward
(e.g., McGill and Sieh, 1991). However, the similarity in age-displacement measurements from various
sites along the western and central Garlock fault segments show that these segments slip at commensurate
rates, at least since ca. 8 ka, where 65-70 m displacements dated at ca. 8 ka have been documented from
the Clark Wash site on the eastern end of the western fault segment and Summit Range West from the
middle of the central segment.
Past paleoseismic studies along the Garlock fault revealed evidence of temporal clustering of earthquakes.
For example, Dawson et al. (2003) identified a cluster of four earthquakes between 0.5–2 ka at the El
Paso Peaks paleoseismic site, which raised questions about whether this cluster-and-lull behavior
persisted further back in time and over what spatial and temporal scales it operated. To address those
questions, in this dissertation I present a new paleo-earthquake record for the central Garlock fault,

97
extending the earthquake history back to early Holocene-latest Pleistocene time (ca. 12 ka). Using
paleoseismic trenching and an improved luminescence dating technique (Rhodes, 2015), the study
presented in Chapter 3 expanded and refined the timing of paleo-earthquakes at the Koehn Lake
paleoseismic site. The combined Koehn Lake + El Paso Peaks record revealed fourteen paleo-earthquakes
over the past ca. 12 ka, confirming that cluster-and-lull behavior was a persistent pattern of behavior on
the central Garlock fault. Specifically, three distinct temporal clusters, each comprising four earthquakes,
were documented, occurring between 0.5-2 ka, 5-7 ka, and 8-10 ka. This variability in earthquake
recurrence suggests that the central Garlock fault was not solely driven by steady tectonic loading but was
also influenced by interactions with nearby faults, which may have triggered earthquakes during lulls.
To more fully constrain the possible driving mechanisms behind these behaviors, in Chapter 5, a
combined incremental slip rate and paleo-earthquake record was generated for the Garlock fault to
provide insight into fault behaviors and help determine how plate-boundary slip is partitioned amongst
major faults in time and space. In this dissertation, I have presented a detailed record of slip for the
Garlock fault by combining the updated incremental slip rate record with paleo-earthquake ages for the
central Garlock fault, possibly one of the most comprehensive earthquake-by-earthquake records for any
major strike-slip fault globally, providing key insights into the temporal and spatial patterns of fault slip
in complex plate boundary systems. This record of slip is based on (1) the incremental slip rate record
from thirteen individual age-displacement measurements, (2) the ages of the fourteen paleo-earthquakes
identified along the central Garlock fault at Koehn Lake and El Paso Peaks, and (3) assumptions of
single-event displacements (6.5 ± 1.5 m) documented by Burns (2023) and McGill and Sieh (1991) in the
last four events on the central Garlock fault. The results highlight distinct periods of variability in slip
behavior, with periods of faster-than-average slip spanning clusters of four earthquakes that accommodate
~25 m of slip, alternating with periods of slower-than-average slip and reduced earthquake recurrence
accommodating <5-10 m of slip.

98
The results presented in Chapter 5 add to the growing evidence that ~25 m of slip accommodated in
clusters of approximately four earthquakes, as seen on the Garlock fault, is a common fault behavior in
mechanically complementary fault systems. The San Andreas fault at the Wrightwood site recorded ≥20
m of slip during a 250-year-long period of exceptionally fast slip (Weldon et al., 2004). Similarly, in New
Zealand, displacements during “fast” periods on the three northern faults of the Marlborough fault zone
all equal ~20–25 m (Zinke et al., 2017; 2019; 2021). These behaviors documented on various plate
boundaries around the world provide critical insights into the driving mechanisms behind fault slip and
earthquake clustering. At the plate-boundary scale, slip is often shared across mechanically
complementary faults in complex structural settings (Peltzer et al., 2001; Dolan et al., 2007; 2016). Such
faults can trade off slip while maintaining a system-level rate that matches relative plate motion (Dolan et
al., 2024), with elastic strain accumulation waxing and waning across subsystems (Gauriau and Dolan,
2024). The data presented here align with the model of Hatem and Dolan (2018), in which plate boundary
systems alternate strain accumulation between subsystems operating over timescales longer than
individual earthquake cycles. Similarly timed events on mechanically complementary faults, such as the
ca. 500-year-old events recorded on the Mojave section of the San Andreas fault, the Garlock fault, and
parts of the northern ECSZ, further imply regional fault interactions. McAuliffe et al. (2013)
demonstrated this through Coulomb failure stress modeling, showing how earthquakes on nearby faults
(e.g., the Panamint Valley fault in the northern ECSZ) can increase stress on the central Garlock fault,
potentially triggering large events (Mw ≤7.8). The observed variability in slip rates and clustering
requires mechanisms operating on the scale of a fault zone to sustain faster-than-average fault slip over
multiple earthquake cycles. One proposed explanation involves strengthening and weakening processes
within the ductile roots of faults (Dolan et al., 2007), supported by a recent model for the strength
evolution of a fault zone proposed by Cawood and Dolan (2024) where the ductile roots of a fault system
harden during periods of rapid slip at rates that temporarily overwhelm the counteracting annealing
processes. Alternatively, Mildon et al. (2022) suggest changes in differential stress due to fault zone
interactions alter viscous strain rates, however, it is unclear whether the magnitudes of stress change are

99
sufficient to drive slip-rate variations over earthquake clustering timescales. More recently, Dolan et al.
(2024) suggested that similar ~25 m slip values on multiple mechanically complementary plate boundary
faults may represent the maximum amount of resolved elastic strain energy that is “stored” on each fault,
where a fault may slip faster for 25 m using up the elastic strain energy on that fault. However, the
processes driving fault behaviors described here are likely controlled by a myriad of factors that are
difficult to quantify. The near-earthquake by earthquake slip record presented for the Garlock fault in this
dissertation, together with other similar records from other major plate boundary-scale strike-slip faults,
provide tight constraints over the displacement (~25 m) and time (millennial) scales at which whatever
mechanical processes control these behaviors operate.

100
TABLES
Table 2.1. Results of single-grain post-IR IRSL dating from Summit Range East (SRE), Summit Range
West (SRW), and Clark Wash (CW). *denotes younger population within sample and is not used in
calculation of average age of stratigraphic unit.
Site Field code Lab code Stratigraphic unit Depth (m) Age (years) 1-σ
SRE SRE14-01* J0733 Qfc2 0.25 1890 ± 150
SRE SRE14-01 J0733 Qfc2 0.25 5560 ± 370
SRE SRE14-02* J0734 Qfc2 0.45 1820 ± 180
SRE SRE14-02 J0734 Qfc2 0.45 5390 ± 410
SRE SRE14-03 J0735 Qfc2 0.65 5640 ± 260
SRE SRE14-04 J0736 Qfc2 0.95 6110 ± 370
CW GF16-01 J1385 Qfy 0.34 8490 ± 490
CW GF16-02 J1386 Qfy 0.52 8420 ± 460
CW GF16-03 J1387 Qfy 0.7 7100 ± 560
CW GF16-04 J1388 Qfy 1.05 12500 ± 900
SRW GF16-05 J1389 Qf2 0.41 7840 ± 760

101
SRW GF16-06 J1390 Qf2 0.63 9050 ± 740
SRW GF16-07 J1391 Qf2 0.84 6710 ± 840

102
Table 2.2. Results of radiocarbon dating from the Clark Wash site. Calibrations based on IntCal20
calibration curve using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020).
Field code Lab code Uncalibrated age
(years B.P.)
Mean (cal.
years B.P.)
2-σ
uncertainty
Stratigraphic
unit
H14-6N-56 AA-14555 6812 ± 91 7667 ± 86 Hoa
H14-6N-17 AA-14551 6888 ± 88 7737 ± 87 Hoc
H14-6N-133 AA-14552 6918 ± 71 7760 ± 75 Hoc
H14-6N-50B AA-14553 6968 ± 109 7802 ± 100 Hoc
H14-6S-31 AA-14549 7040 ± 94 7857 ± 91 Hoc
H14-6S-1 AA-14550 7056 ± 103 7872 ± 101 Hoc
H14-6N140/141/142
Beta74108
7170 ± 140 7997 ± 146 Hoc
H14-6N-44 AA-16339 7240 ± 59 8066 ± 68 Hoa

103
Table 3.1. Summary of event evidence seen on the Koehn Lake trench exposures
Label Type of evidence Wall Terrace West (m) East (m) Description of evidence
a
Upward
terminations and
filled fissures Both Upper 2.5 5
4 upward terminations and a fissure fill,
faults terminate at the base of a gravel
channel
b
Upward
terminations and
filled fissure Both Upper 4 5.5
5 upward terminations and a fissure fill,
faults terminate at the base of a gravel
channel
c Large fissure fill Both Lower 12.5 14.5
70 cm deep fissure filled with faulted and
warped lacustrine units, capped by 5-15 cm
brown clay and 1-2 cm brown sand, layers
are slightly arcuate
d
Fissure fills OR
SSDs Both Upper 18 19
West wall fissure fills within a thick clay
layer and have a thinner clay layer that is the
same color and texture above which has
filled these fissures, faults that rupture
through have displacements of 0.5-1 cm.
East wall 30 cm deep fissure with faults
bounding the structure that extend close to
the surface
e
Angular
unconformity Both Upper 18 19
Thin (cm-scale) dipping clay beds
unconformable against thick clay layer above
by ~20-30°
f
Upward
terminations Both Upper 19.9 20.3
West wall fault terminates between the base
of a gravel layer and the top of a coarse sand
layer. Base of friable sediments (where the
fault may have propagated through but a
fault trace is no longer present). Adjacent to
main fault trace. Vertical displacement of 15
cm. East wall fault splays into three, where
all three are terminated between the base of
the friable sediment. One is main fault
between the south and graben sections.

104
Vertical displacements are difficult to
measure
g
Upward
terminations West Upper 20.5 -
Adjacent to the main fault, west wall 2-3
faults terminate at the base of a thin (~2 cm)
gravel-coarse sand layer and the top of a thin
(~2 cm) clay layer).
h Fissure fill West Upper 21.5 -
At the base of a faulted gravel unit within the
graben. This gravel is underlain by a 4 cm
thick clay unit. A 25 cm deep fissure fill with
an underlying fault is observed at 21.35 m
and -0.4 m, filled with the overlying gravel.
25 cm north, another possible fissure fill
(wider and shallower) is observed with an
underlying fault. 20 cm farther to the north
from the smaller fissure fill, a fault
terminates at the base of the gravel/top of the
4 cm clay layer.
i Paleoseismite Both Lower 23 23
Clay deformed more than overlying sand,
more deformation near edges of graben and
less deformed in the middle. Multiple faults
propagating through this section, possibly
one upward termination at m 24.4, -1.7 m.
Faults have vertical displacements of 1-10
cm
j
Upward
terminations,
growth strata West Upper 24 -
Faults terminate within a gravel layer, at the
base of a 1-2 cm thick cemented layer within
the gravel where no displacements can be
observed above this. Vertical displacements
of 1-2 cm are observed on these fault strands
which turns into cracking just below EH 6.
To be conservative, EH 6 is the first
unbroken (and uncracked) layer above the
faults.
k
Angular
unconformity,
upward
terminations Both Upper 24.6 25.8
Base of a gravel channel and the top of a
dipping clay layer. This upside-down, coneshaped feature is bounded on either side by
faults that extend upwards. Within this
“cone” there is a block of broken clay and
sand layers to the north and dipping (and

105
slightly deformed) clay and sand layers to the
south. There has been 60 cm of displacement
on the southernmost fault in this “cone”,
where a clay and sand block match up with
the edge of a swale on the lower west wall.
On the east wall a similar but larger “cone”
feature than the WW. The unconformity is
overlain by fine sand (rather than gravel) and
the top of the shallowest clay layer, displaced
from a clay layer on the lower east wall.
Vertical displacements 3-60 cm (60 cm seen
at the north edge of graben). Some cracking
above these faults but no significant
displacements were observed. Faults
terminate at the base of a gravel/coarse sand
unit. The gravel/coarse sand unit is most
likely a channel deposit that may have
eroded the upper section of these two faults
and may explain why event horizon is
observed deeper on the west wall compared
with the east wall. Faults have Terminate at
the base of a gravel/coarse sand unit, likely
erosional. Faults from north to south have
vertical displacements of 2-3 cm, 1-2 cm, 1
cm, and 15 cm. The northern and southern
faults appear to match displacements on the
WW, both having displacements of ~3 cm
and 15 cm
l
Upward
terminations Both Lower 27.2 26.5
On the west wall faults terminate at the base
of a gravel channel, which may have eroded
any fault traces that extended shallower. 1-3
cm of vertical displacement. The
southernmost fault (if it is a fault), which
terminates at the deepest part of the channel,
appears to displace a brown clay layer by
about 3 cm. On the east wall faults terminate
at the base of the same gravel
channel/massive sand, similar to the west
wall but not exactly the same faults. Vertical
displacements of <5-10 mm. An additional
four faults propagate through this gravel
channel with vertical displacements of 1-4
cm.

106
m
Fissure fill,
upward
termination West Lower 26.6 -
Lined by soft-sediment likely deformed by a
tectonic event. Above silt/deformed clay
layers and below an eroded channel feature
comprising coarse sand/pebbles, grading
finer upwards
n
Upward
terminations Both Lower 29.5 29
Similar feature to l, faults terminate a the
base of a thick gravel channel. Some faults
may terminate within the gravel, however
they don’t appear to displace any small scale
cemented gravel layers near the top of this
channel. Vertical displacements range 1-10
cm.
o
Angular
unconformity,
upward
termination Both Lower 29.5 29
Slightly deformed (soft-sediment
deformation) clay-fine sand layers which
buttress against a sand and clay layer above
p
Tilted blocks,
soft-sediment
deformation West Lower 29.8 -
Faulted blocks (10-15 cm wide) with a swale
to the south are filled in with clay and sand
layers, EH above faulted blocks and below
lowest clay swale layer, less soft sediment
deformation is seen on this wall compared
with the WW
q
Upward
terminations East Upper - 36
Four faults terminate within (but near the
base of) a draped clay layer that thins out to
the north. These faults break the base of this
layer but don’t propagate to the top of the
draped clay. Displacements range from
negligible to a few centimeters.
r
Upward
terminations East Upper - 28
Terminate at the base of a gravel/coarse sand
unit, likely erosional. Faults from north to
south have vertical displacements of 2-3 cm,
1-2 cm, 1 cm, and 15 cm. The northern and
southern faults appear to match
displacements on the WW, both having
displacements of ~3 cm and 15 cm

107
s Fallen block West Upper 35.5 -
Fallen into a channel and clay/mud has filled
around it related to a tectonic event, or
alternatively may have fallen due to gravity
and slope shallowing processes. Within
“triangle” of clay deposition between the
block and the wall to the north it fell from,
clay and silt beds are discontinuous and
broken up. Above this “triangle” the beds
appear to be deposited sub-horizontally in a
mud swale smiley. Stratigraphy seen within
the fallen block matches the clay and sand
layers of the wall it fell from.
t
Soft sediment
deformation Both Upper 18 19
Multiple soft sediment deformation features
along a horizon
u
Upward
terminations East Upper - 19.5
three faults terminate at the base of clay layer
and top of gravel layer. The two north
strands have <5 mm of vertical displacement
and the strand at m 19.4, +0.15 m has 5 cm.
The southernmost strand doesn’t have
obvious vertical displacement, but broken
clay layers are observed.

108
Table 3.2. IRSL sample age estimates (variable OD) using a fading correction of 12%, modeled water
content that is more representative of actual conditions over the lifetime of burial of the sediment taking
into account modern, measured water content AND a calculated saturated water content (from bulk
density and porosity) for each sample, and overdispersion values of between 15 and 25% are used based
on clear signals that indicate modes of well-bleached grains.
Field code Age (ka) Distance from south end of trench Depth Collected Sediment type
KL-01 14.34 17.87 -2.42 gray lake bed
KL-03 12.92 17.44 -1.92 gray lake bed
KL-05 12.84 17.3 -1.43 gray lake bed
KL-06 12.82 17.32 -1.2 gray lake bed
KL-07 9.54 17.46 -0.88 sand/clay
KL-08 9.32 17.15 -0.63 sand/clay
KL-09 9.9 17.35 -0.55 sand/clay
KL-12 8.68 17.5 0.04 sand fine
KL-14 6.18 23.5 -2.26 sand/clay
KL-15 6.37 23.64 -2.12 sand/clay
KL-16 5.82 23.47 -2.04 sand fine
KL-17 6.22 22.7 -1.86 sand/clay
KL-18 6.29 22.87 -1.62 sand/clay

109
KL-21 5.49 22.2 -1.25 sand/clay
KL-31 8.13 29.26 -2.02 sand fine
KL-32 7.32 29.44 -1.95 sand med
KL-33 6.61 29.2 -1.8 gravel
KL-34 7.69 27.3 -1.77 sand med
KL-35 7.54 27.09 -1.67 sand coarse
KL-36 6.42 21.45 -0.45 silt
KL-37 4.24 21.74 -0.29 sand coarse
KL-38 9.88 18.64 -0.56 sand/clay
KL-39 7.78 18.37 -0.44 sand/clay
KL-41 6.14 19.63 0.05 sand coarse
KL-42 3.24 20.11 0.3 sand coarse
KL-44 0.68 20.33 0.39 sand coarse
KL-45 8.92 29.28 -2.81 sand/clay
KL-46 7.65 29.3 -2.56 silt
KL-47 8.58 29.07 -2.33 sand med

110
KL-48 7.13 26.64 -2.28 sand fine
KL-49 7.12 26.57 -2.25 sand fine
KL-50 5.91 26.94 -2.3 sand coarse
KL-51 6.87 26.41 -2.13 sand med
KL-52 10.37 16 -1.13 sand/clay
KL-53 3.68 23.73 -0.5 sand med
KL-54 2.9 23.71 -0.24 sand fine
KL-55 6.35 24.54 -1.02 sand/clay
KL-56 5.63 24.47 -0.69 sand coarse
KL-57 6.82 22.85 -2.25 sand/clay
KL-58 5.49 23.17 -1.98 sand/clay
KL-59 8.27 15.1 -0.75 sand fine
KL-60 5.24 27.66 -0.35 sand fine
KL-61 1.56 28.17 -0.13 sand coarse
KL-63 5.28 35.15 -0.44 sand fine
KL-64 3.07 35.22 0.1 sand fine

111
KL-65 7.18 26.83 -2.45 sand fine
KL-66 8.05 2.62 -0.46 sand/clay
KL-68 1.83 2.22 -0.2 sand coarse
KL-70 7.9 5.73 -0.24 sand/clay
KL-71 7.46 5.78 0.08 sand/clay

112
Table 3.3. Radiocarbon dates and calibrated ages from the Koehn Lake 2021 paleoseismic trench
Sample no.
UCIAMS
no.
Conventional
14C age (yr
B.P.)
Calibrated
median (yr
B.P.)
Calibrated
range (yr
B.P.) Wall Terrace
Sediment
Type
KL2021-C14-1 248432 7950 ± 20 8829 8984-8646 west upper silt mix
KL2021-C14-12 249880 2770 ± 15 2861 2931-2786 east upper sand
KL2021-C14-14 248433 2575 ± 15 2733 2748-2720 east upper sand
KL2021-C14-17 248434 2545 ± 15 2717 2743-2523 east upper clay
KL2021-C14-18 248435 7835 ± 30 8610 8720-8541 west upper silt mix
KL2021-C14-19 248436 7965 ± 20 8855 8989-8651 west upper silt mix
KL2021-C14-20 248437 14490 ± 40 17655 17875-17441 east lower clay
KL2021-C14-21 249881 5085 ± 20 5806 5908-5749 east lower silt
KL2021-C14-25 248438 7660 ± 140 8471 8978-8178 east lower clay/silt
KL2021-C14-29 248439 7520 ± 20 8354 8388-8217 west lower silt
KL2021-C14-33 249882 2525 ± 15 2619 2729-2513 west upper clay
KL2021-C14-34 249883 5095 ± 20 5804 5912-5751 west upper silt
KL2021-C14-53 248440 10190 ± 70 11850 12431-11403 east lower clay

113
KL2021-C14-54 248441 8250 ± 35 9220 9406-9032 east upper silt
KL2021-C14-56 248442 8190 ± 25 9124 9270-9023 west upper silt
KL2021-C14-60 248443 11860 ± 420 13905 15223-12994 west upper silt
KL2021-C14-61 248444 24140 ± 530 28386 29543-27337 west upper silt
KL2021-C14-64 248445 7745 ± 20 8519 8591-8450 east lower clay
KL2021-C14-67 248446 4775 ± 25 5526 5585-5471 west upper sand
KL2021-C14-68 248447 6055 ± 20 6908 6977-6800 west lower gravel
KL2021-C14-71 249884 7535 ± 30 8360 8408-8215 west upper silt
KL2021-C14-79 249885 8180 ± 80 9146 9425-8986 west upper silt
KL2021-C14-80 249886 5880 ± 20 6703 6747-6653 west upper silt
KL2021-C14-81 248448 2490 ± 15 2581 2714-2493 west upper silt/sand
KL2021-C14-88 249887 2475 ± 15 2595 2707-2469 west upper sand/gravel
KL2021-C14-92 248453 7290 ± 120 8112 8363-7873 west lower silt
KL2021-C14-93 248469 8290 ± 140 9261 9543-8810 west lower silt
KL2021-C14-100 248454 5730 ± 210 6554 7156-6009 west upper silt
KL2021-C14-110 248455 2955 ± 20 3117 3208-3005 west upper gravel

114
KL2021-C14-117 248456 6040 ± 60 6888 7156-6735 west lower sand
KL2021-C14-120 249888 3575 ± 20 3873 3965-3779 east upper clay
KL2021-C14-121 248457 6080 ± 20 6937 7150-6857 west lower silt
KL2021-C14-126 248458 5845 ± 45 6659 6780-6501 west lower fine sand
KL2021-C14-154 248459 8475 ± 20 9502 9533-9467 east lower fine sand
KL2021-C14-160 248460 5010 ± 20 5737 5889-5656 west upper silt
KL2021-C14-161 249889 2495 ± 20 2582 2720-2492 west upper fine sand
KL2021-C14-165 248461 5025 ± 15 5831 5894-5662 west lower fine sand
KL2021-C14-174 248462 9110 ± 130 10291 10655-9893 west lower clay
KL2021-C14-178 248463 9430 ± 220 10720 11270-10185 west lower silt/clay
KL2021-C14-179 248464 7485 ± 20 8313 8371-8202 east upper sand
KL2021-C14-182 249890 3640 ± 40 3956 4085-3846 east upper
medium
sand
KL2021-C14-183 249891 4295 ± 20 4852 4871-4834 east upper
medium
sand
KL2021-C14-184 248465 7890 ± 130 8744 9078-8415 east lower fine sand
KL2021-C14-188 248466 6070 ± 20 6924 6992-6807 west lower
medium
sand

115
KL2021-C14-189 249892 6135 ± 20 7016 7158-6945 east lower fine sand
KL2021-C14-190 248467 8220 ± 180 9171 9534-8648 east upper silt
KL2021-C14-192 248468 5125 ± 45 5858 5990-5745 west upper fine sand

116
Table 3.4. Age constraints on paleoseismic slip events at the Koehn Lake site calculated on OxCal (V3)
Trench Section Event label
Event
horizon
label
Average age -
median (years
before 2024 CE)
Minimum 95%
age (years before
2024 CE)
Maximum 95%
age (years before
2024 CE)
Central/Southern KL1 f 470 220 730
Central KL2 j 3300 2760 3910
Central KL3 i 6250 5990 6570
Northern KL4 l & m 6930 6870 7000
Northern KL5 m 7080 6970 7180
Northern KL6 o 8400 8050 8860
Northern KL7a p 8610 8330 9020
Southern KL7b t 8930 8400 9520
Southern KL8 d 9350 8820 9920
Southern KL9 e 9580 9020 10150
Southern KL10 c 10920 10360 11560
Possible
correlations
Northern
KL2, EPPR, EPP-Q r 3370 1560 5190

117
Northern KL2, EPP-K s 4010 2780 5500
Northern KL2 q 2800 2790 2820
Central
KL3, EPPK, EPP-F h 5290 4110 6570
Southern
KL3, KL4,
KL5, EPPK, EPP-F g 6130 4780 7190
Central/Northern KL3, EPP-K k 5810 5080 6580
Southern KL4, EPP-F b 7180 6230 8080
Southern KL5, EPP-F a 7570 6660 8410
Southern
KL6, KL7a,
KL7b, KL8,
KL9 u 9320 8420 10240

118
Table 3.5. Minimum paleo-earthquake record for the central Garlock fault between the KL and EPP
trench sites
Event
label
Garlock
label
Average age - median
(years before 2024 CE)
Minimum 95% age
(years before 2024 CE)
Maximum 95% age
(years before 2024 CE)
Interevent
time
EPP-W GF1 480 270 650 -
EPP-U GF2 1210 915 1465 730
EPP-R GF3 1650 1400 1850 440
EPP-Q GF4 1860 1590 2090 210
KL2 GF5 3300 2760 3910 1440
EPP-K GF6 5160 4730 5550 1860
KL3 GF7 6250 5990 6570 1090
KL4 GF8 6930 6870 7000 680
KL5 GF9 7080 6970 7180 150
KL6 GF10 8400 8050 8860 1320
KL7 GF11 8770 8330 9520 370
KL8 GF12 9350 8820 9920 580
KL9 GF13 9580 9020 10150 230

119
KL10 GF14 10920 10360 11560 1340

120
Table 5.1. Summary of slip rate data used in the Garlock fault incremental slip rate record. *MRE-most
recent event
Segment Site
Slip Rate
Label Offset (m) Age (years) References
All - MRE* 6.5 ± 1.5 270-650
Burns (2023), McGill and Sieh
(1991), Dawson et al. (2003)
Western Clark Wash CW 66 ± 6 8010 [+1115 -210]
McGill et al. (2009), Fougere
et al. (2024)
Central
Koehn Lake
berm KLb 85 - 100 12450 ± 970 this dissertation
Central
Summit Range
West SRW 70 ± 7 7960 ± 450
Ganev et al. (2012), Fougere et
al. (2024), this dissertation
Central
Summit Range
East SRE 38 ± 1 5570 ± 190
Fougere et al. (2024), this
dissertation
Central
Christmas
Canyon West CCW 26[+3.5 -2.5] 1870 ± 150 Dolan et al. (2016)
Central
Searles Lake
Shoreline SLS 90 (+16/-8) 12500 ± 1100
McGill and Sieh (1993), this
dissertation
Central
Pilot Knob
Valley west
PKVwA1 30-37 3505 ± 125
Rittase et al. (2014), this
dissertation
Central
Pilot Knob
Valley west PKVw-C 95 ± 3 13610 ± 770 this dissertation
Central
Pilot Knob
Valley west PKVw-E 86 ± 2 9400-12500 this dissertation

121
Central
Pilot Knob
Valley east PKVe-A 8 ± 1 860 ± 100 Crane (2014), this dissertation
Central
Pilot Knob
Valley east PKVe-B 19 ± 3 1860 ± 180 Crane (2014), this dissertation
Central
Pilot Knob
Valley east PKVe-C 30 ± 2 3940 ± [+575,-555] Crane (2014), this dissertation
Central
Pilot Knob
Valley east PKVe-D 43 ± 3 5630 ± [+530,-520] Crane (2014), this dissertation

122
FIGURES
Figure 2.1. Map of the Garlock fault and other active faults around the Mojave region of southern
California (gray), including the Mojave section of the San Andreas fault. Yellow star is the location of the
Clark Wash (CW) site of this study, blue star is the location of the Summit Range West (SRW) and
Summit Range East (SRE) sites of this study. White circles show the locations of past slip rate studies;
CCW is the Christmas Canyon West site of Dolan et al. (2016), KL is the Koehn Lake site of Clark &
Lajoie (1974), and PKV is the Pilot Knob Valley site of Rittase et al. (2014). A gray square shows the
location of the El Paso Peaks (EPP) paleoseismic site of Dawson et al. (2003). Quaternary fault traces
sourced from US Geological Survey & California Geological Survey (2023)

123
Figure 2.2. (a) Annotated lidar-derived hillshade of the Summit Range West (SRW) study site originally
studied by Ganev et al. (2012) at N 35.479116° W 117.560319°, and the Summit Range East (SRE) at N
35.485943° W 117.530138°. (b) Topographic map (2-m contour interval) derived from the lidar dataset
of the same areas as in (a). Garlock fault trace shown by the red line.

124
Figure 2.3. (a) Interpreted lidar-derived hillshade of the Summit Range East (SRE) site. Colors show
various mapped alluvial fan surfaces (Qfa, Qfb, Qfc1, Qfc2, and Qfd) and other mapped rock units (Tss
and Mzg). IRSL sample pit is shown by a blue circle and the fault trace is shown by red lines. Inset shows
a topographic profile marked by a white dotted line (A-A’) across Qfc1 and Qfc2 that suggests that the
two fan segments may be the same fan. 38-m back-slipped restoration of the SRE site shown with (b)
interpreted lidar-derived hillshade and (c) lidar-derived hillshade. The 38-m preferred offset measurement
is based on the most plausible configuration of the channel (Ch-1) incised into the Qfc2 fan.

125
Figure 2.4. Age estimates for IRSL samples SRE14-01, SRE14-02, SRE14-03, and SRE14-04 that were
collected from the Qfc2 fan at the Summit Range East site. (a) Cross-section diagram of Qfc2 sample pit
showing depths at which samples were collected and general stratigraphy. Sample SRE14-04 is not
included in the final slip rate calculation, (b) Single-grain potassium-feldspar post-IR infrared stimulated
luminescence distribution data and preferred age estimates for each sample. (c) Preferred estimate of the
age of Qfc2 deposition by combining three shallowest sample calendar dates with Gaussian errors using a
chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020).

126
Figure 2.5. (a) Interpreted lidar-derived hillshade of the Summit Range West (SRW) site. Colors show
various mapped alluvial fan surfaces (Qf1, Qf2, Qf3, and Qf4) and other mapped rock units (QTc). IRSL
sample pit is shown by a blue circle and the fault trace is shown by red lines. 70-m back-slipped
restoration of the SRW site shown with (b) interpreted lidar-derived hillshade and (c) lidar-derived
hillshade. The 70-m preferred offset measurement is based on the restoration of the incised edge of the
Qf2 fan surface on the western margin of the channe

127
Figure 2.6. Age estimates for IRSL samples GF16-05, GF16-06, and GF16-07, which were collected
from the Qf2 fan at the Summit Range West site. (a) Cross-section view of Qf2 sample pit showing
depths at which samples were collected. (b) Single-grain K-feldspar IRSL distribution data and preferred
age estimates for each sample. (c) Preferred estimate of the age of Qf2 deposition by combining all three
calendar dates with Gaussian errors using a chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009;
Reimer et al., 2020).

128
Figure 2.7. Regional topographic map of the Clark Wash site with a contour interval of 100 m. The Clark
Wash site is located at the northeastern extent of the western Garlock fault segment adjacent to the 2-3 km
wide step over at Koehn Lake at N 35.214608° W 118.075739°. The black dashed box outlines the
location of Figure 2.8a.

129
Figure 2.8. (a) Interpreted lidar-derived hillshade of the Clark Wash (CW) site. Colors overlying lidarderived hillshade show various mapped surfaces (disturbed surface, modern channel, various aged
alluvium, and colluvium). IRSL sample pit is shown by a blue circle and the fault trace is shown by a red
line. Eight of the ten trenches that were excavated by S. F. McGill et al. (2009) fall within the area shown
in this figure and are marked by gray boxes. The location beneath which radiocarbon samples from Hoa
were collected is indicated by the asterisk within one of the trenches. The 66-m back-slipped restoration
of the CW site is shown with (b) interpreted lidar-derived hillshade and (c) lidar-derived hillshade. The
66-m preferred offset measurement is based on the restoration of the northeastern channel wall (see
Figure S1 for minimum and maximum restorations).

130
Figure 2.9. Age estimates for IRSL samples GF16-01, GF16-02, GF16-03, and GF16-04 collected at the
Clark Wash site. (a) Cross-section view of IRSL sample pit showing depths that samples were collected.
GF16-01, GF16-02, and GF16-03 were collected from Qfy, whereas GF16-04 was collected from Qfo. (b)
Single-grain K-feldspar post-IR IRSL distribution data and preferred age estimates for each sample. (c)
Preferred estimate of the age of Qfy deposition by combining all four IRSL dates with Gaussian errors
using a chi-squared test using OxCal 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020). (d) Probability
density functions for the eight radiocarbon charcoal samples ages collected from the Clark Wash site
calculated using OxCal version 4.4 (Bronk Ramsey, 2009; Reimer et al., 2020). Average age of all eight
samples was calculated using the OxCal combine function. (e) Age estimate of initial onset of fault offset
calculated by finding the PDF representing the interval between (c) and (d).

131
Figure 2.10. (a) Incremental slip history for the central and western Garlock fault including sites: CCW -
Christmas Canyon West slip rate site (Dolan et al., 2016), CW - Clark Wash (this paper, building on S. F.
McGill et al., 2009), MRE - most recent earthquake recorded by Dawson et al. (2003), SR - Slate Range
slip rate site (Rittase et al., 2014), SRE - Summit Range East slip rate site (this paper), SRW - Summit
Range West slip rate site (this paper, building on Ganev et al., 2012). (b) Incremental slip rates for the
central Garlock fault calculated using RISeR (Zinke et al., 2017, 2019) using only the most tightly
constrained rates from a 16-km-long section of the central part of the fault.

132
Figure 3.1. Map of the Garlock fault (red) and other active faults around the Mojave region of southern
California (gray), including the Mojave section of the San Andreas fault. The yellow circle is the location
of the Koehn Lake paleoseismic site (this paper). White circles show the locations of past paleoseismic
sites along the Garlock fault; CT - Campo Teresa (Gath and Rockwell, 2018), TL - Twin Lakes (Madden
Madugo et al., 2012), CW - Clark Wash (McGill et al., 2009), EPP - El Paso Peaks (Dawson et al, 2003),
CCW - Christmas Canyon West (McGill et al., in prep), EP - Echo Playa (Kemp et al., 2016), SPR –
Straw Peak Road (Chapter 4). The Garlock fault has been divided into three sections, western, central,
and eastern, based on changes in its strike or the presence of structural discontinuities (McGill, 1992).

133
Figure 3.2. (a) Fremont Valley map with geology sourced from Macrostrat (Peters et al., 2018), (b)
Annotated satellite image of the Koehn Lake paleoseismic and slip rate site (Google Earth, accessed
December 2024), and (c) Annotated photograph of the Koehn Lake trench looking eastwards.

134
Figure 3.3. Generalized schematic of the Koehn Lake trench stratigraphy and average ages for the
various units within the three sections of the trench

135
Figure 3.4. Annotated southern trench logs, (A) annotated west wall trench log, (B) annotated east wall
trench log, and (C) relative positions of event horizons (black lines), IRSL samples, and radiocarbon
samples from both walls.

136
Figure 3.5. Annotated northern trench logs, (A) annotated west wall trench log, (B) annotated east wall
trench log, and (C) relative positions of event horizons (black lines), IRSL samples, and radiocarbon
samples from both walls.

137
Figure 3.6. Annotated northern trench logs, (A) annotated west wall trench log, (B) annotated east wall
trench log, and (C) relative positions of event horizons (black lines), IRSL samples, and radiocarbon
samples from both walls. This section is dominated by alluvium and highly channelized with many
laterally discontinuous units along the length of this section of the trench exposure.

138
Figure 3.7. Paleo-earthquake age data. (A) Koehn Lake minimum paleo-earthquake record (V3), (B) El
Paso Peaks paleo-earthquake record, and (C) combined central Garlock fault preferred paleo-earthquake
record for the central Garlock fault (V4).

139
Figure 4.1. Annotated Straw Peak Road trench log for the east wall exposure, shown from m 7 to m 1

140
Figure 5.1. Map of the Garlock fault (red) and other active faults around the Mojave region of southern
California (gray), including the Mojave section of the San Andreas fault. Squares along the Garlock fault
are slip rate (white squares) and paleoseismic (purple squares) study sites. Sites that have both slip rate
and paleoseismic studies are shown with half white-half purple squares. CT – Campo Teresa (Gath and
Rockwell, 2018), TL – Twin Lakes (Madden and Dolan, 2008); Madden Madugo et al., 2012), CW –
Clark Wash (McGill et al., 2009; Fougere et al., 2024), KL – Koehn Lake (Clark & Lajoie, 1974; Madden
and Dawson, 2006, this chapter), EPP – El Paso Peaks (Dawson et al., 2003), SRW – Summit Range
West (Ganev et al., 2012; Fougere et al., 2024), SRE – Summit Range East (Fougere et al., 2024), CCW –
Christmas Canyon West (Dolan et al., 2016; Pena, 2019; McGill et al, in prep), SLS - Searles Lake
Shoreline (McGill and Sieh, 1993, this chapter), SPR – Straw Peak Road (chapter 4), PKVw – Pilot Knob
Valley West (Crane, 2014; Rittase et al., 2014, this chapter), PKVe – Pilot Knob Valley East (Crane,
2014, this chapter). Quaternary fault traces sourced from US Geological Survey & California Geological
Survey (2023).

141
Figure 5.2. Lidar-derived hillshaded images showing overview maps for the Koehn Lake berm site,
Searles Lake shoreline site, and Pilot Knob Valley.

142
Figure 5.3 Koehn Lake berm slip rate site, luminescence (IRSL) sample locations and ages, and preferred
restoration based on the projected and estimated ridge crest.

143
Figure 5.4 Searles Lake Shoreline slip rate site, luminescence (IRSL) sample locations and ages, and
preferred restoration.

144
Figure 5.5 Pilot Knob Valley west slip rate site overview.

145
Figure 5.6 Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations and ages,
preferred restoration for PKVw-A, and trench log from Rittase et al. (2014).

146
Figure 5.7 Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations and ages, and
preferred restoration for PKVw-C.

147
Figure 5.8 Pilot Knob Valley west slip rate site, luminescence (IRSL) sample locations and ages, and
preferred restoration for PKVw-E.

148
Figure 5.9 Pilot Knob Valley east slip rate site and preferred restorations for PKVe-A, PKVe-B, PKVeC, and PKVe-D

149
Figure 5.10 (A) Age-displacement measurements used in incremental slip rate record for the western and
central Garlock fault, (B) updated in incremental slip rate record calculated using RISeR (Zinke et al.,
2017; 2019), (C) combined records showing earthquake-by-earthquake slip, and (D) summary of periods
of fault behavior.

150
REFERENCES
Akçiz, S. O., L. G. Ludwig, J. R. Arrowsmith, and O. Zielke. 2010. Century-long average time intervals
between earthquake ruptures of the San Andreas fault in the Carrizo Plain, California. Geology 38: 787–
790.
Andrew, J. E., Walker, J. D., and Monastero, F. C. (2015). Evolution of the central Garlock fault zone,
California: A major sinistral fault embedded in a dextral plate margin. Geological Society of America
Bulletin, 127(1-2), 227-249. doi: 10.1130/B31027
Atwater, T., & Stock, J. (1998). Pacific-North America plate tectonics of the Neogene southwestern
United States: an update. International Geology Review, 40(5), 375-402.
Bacon, S. N., Burke, R. M., Pezzopane, S. K., & Jayko, A. S. (2006). Last glacial maximum and
Holocene lake levels of Owens Lake, eastern California, USA. Quaternary Science Reviews, 25(11-12),
1264-1282.
Bemis, S. P., Scharer, K., & Dolan, J. F. (2021). The San Andreas Fault paleoseismic record at Elizabeth
Lake: Why are there fewer surface‐rupturing earthquakes on the Mojave section?. Bulletin of the
Seismological Society of America, 111(3), 1590-1613.
Bennett, R. A., Davis, J. L., & Wernicke, B. P. (1999). Present-day pattern of Cordilleran deformation in
the western United States. Geology, 27(4), 371-374.
Berryman, K. R., Cochran, U. A., Clark, K. J., Biasi, G. P., Langridge, R. M., and Villamor, P. (2012).
Major earthquakes occur regularly on an isolated plate boundary fault. Science, 336(6089), 1690-1693.
doi: 10.1126/science.1218959.1
Blythe, A. E., and Longinotti, N. (2013). Exhumation of the southern Sierra Nevada–eastern Tehachapi
Mountains constrained by low-temperature thermochronology: Implications for the initiation of the
Garlock fault. Lithosphere, 5(3), 321-327. doi: 10.1130/L252.1
Brady III, R. H., Clayton, J., Troxel, B. W., Verosub, K. L., Cregan, A., & Abrams, M. (1989). Thematic
Mapper and field investigations at the intersection of the Death Valley and Garlock fault zones,
California. Remote Sensing of Environment, 28, 207-217.

151
Bronk Ramsey, C. (2009). Bayesian analysis of radiocarbon dates. Radiocarbon, 51(1), 337–360. doi:
10.1130/L252.1
Burbank, D. W., and Whistler, D. P. (1987). Temporally constrained tectonic rotations derived from
magnetostratigiraphic data: Implications for the initiation of the Garlock fault, California. Geology,
15(12), 1172-1175
Burns, James Eric, "LATE HOLOCENE SLIP HISTORY OF THE CENTRAL GARLOCK FAULT,
MOJAVE DESERT, CALIFORNIA" (2023). Electronic Theses, Projects, and Dissertations. 1656.
Carter, B. (1994). Neogene offsets and displacement rates, central Garlock fault, California, in McGill,
S.F., and Ross, T.M., eds., Geological investigations of an active margin: Geological Society of America,
Cordilleran Section Annual Meeting Guidebook, v. 27, p. 348-356.
Cawood, T., & Dolan, J. (2024). An exploration of potentially reversible controls on millennial-scale
variations in the slip rate of seismogenic faults: Linking structural observations with variable earthquake
recurrence patterns. Seismica, 3(2).
Chuang, R. Y., and Johnson, K. M. (2011). Reconciling geologic and geodetic model fault slip-rate
discrepancies in Southern California: Consideration of nonsteady mantle flow and lower crustal fault
creep. Geology, 39(7), 627-630. doi: 10.1130/G32120.1
Clark, M. M. (1973). Map showing recently active breaks along the Garlock and associated faults,
California. United States Geologic Survey Report, No. 741. doi: 10.3133/i741
Clark, M. M., and Lajoie, K. R. (1974). Holocene behavior of the Garlock fault. Geological Society of
America Abstracts with Programs, 6(7), 156-157.
Crane, Thomas M., "QUALITATIVE COMPARISON OF OFFSET SURFACES BETWEEN THE
CENTRAL AND EASTERN GARLOCK FAULT" (2014). Electronic Theses, Projects, and
Dissertations. 119.

152
Daëron, M., Klinger, Y., Tapponnier, P., Elias, A., Jacques, E., & Sursock, A. (2007). 12,000-year-long
record of 10 to 13 paleoearthquakes on the Yammouneh fault, Levant fault system, Lebanon. Bulletin of
the seismological society of America, 97(3), 749-771.
Davis, G. A., and Burchfiel, B. C. (1973). Garlock fault: An intracontinental transform structure, southern
California. Geological Society of America Bulletin, 84(4), 1407-1422.
Dawson, T. E., McGill, S. F., and Rockwell, T. K. (2003). Irregular recurrence of paleo-earthquakes
along the central Garlock fault near El Paso Peaks, California. Journal of Geophysical Research: Solid
Earth, 108(B7). doi: 10.1029/2001JB001744
Del Vecchio, J., Lang, K. A., Robins, C. R., McGuire, C. P., and Rhodes, E. J. (2018). Storage and
weathering of landslide debris in the eastern San Gabriel Mountains, California, USA: Implications for
mountain solute flux. Earth Surface Processes and Landforms, 43(13), 2724-2737. doi: 10.1002/esp.4427
DeMets, C., Gordon, R. G., Stein, S., & Argus, D. F. (1987). A revised estimate of Pacific‐North America
motion and implications for western North America plate boundary zone tectonics. Geophysical Research
Letters, 14(9), 911-914.
Dibblee, T. W. (2008). Geologic Map of the Cross Mountain & Saltdale 15 Minute Quadrangles: Kern
County, California. Santa Barbara Museum of Natural History.
Dixon, T. H., Miller, M., Farina, F., Wang, H., & Johnson, D. (2000). Present‐day motion of the Sierra
Nevada block and some tectonic implications for the Basin and Range province, North American
Cordillera. Tectonics, 19(1), 1-24.
Dokka, R. K., & Travis, C. J. (1990). Late Cenozoic strike‐slip faulting in the Mojave Desert, California.
Tectonics, 9(2), 311-340.
Dolan, J. F., and Meade, B. J. (2017). A comparison of geodetic and geologic rates prior to large strike‐
slip earthquakes: A diversity of earthquake‐cycle behaviors? Geochemistry, Geophysics, Geosystems,
18(12), 4426-4436. doi: 10.1002/2017GC007014

153
Dolan, J. F., Bowman, D. D., and Sammis, C. G. (2007). Long-range and long-term fault interactions in
Southern California. Geology, 35(9), 855-858. doi: 10.1130/G23789A.1
Dolan, J. F., McAuliffe, L. J., Rhodes, E. J., McGill, S. F., and Zinke, R. (2016). Extreme multimillennial slip rate variations on the Garlock fault, California: Strain super-cycles, potentially timevariable fault strength, and implications for system-level earthquake occurrence. Earth and Planetary
Science Letters, 446, 123-136. doi: 10.1016/j.epsl.2016.04.011
Dolan, J.F., Van Dissen, R.J., Rhodes, E.J., Zinke, R., Hatem, A.E., McGuire, C., Langridge, R.M. and
Grenader, J.R. (2024). One tune, many tempos: Faults trade off slip in time and space to accommodate
relative plate motions. Earth and Planetary Science Letters, 625, 118484.
Evans, E. L., Thatcher, W. R., Pollitz, F. F., and Murray, J. R. (2016). Persistent slip rate discrepancies in
the eastern California (USA) shear zone. Geology, 44(9), 691-694. doi: 10.1130/G37967.1
Evans, E. L. (2017). Persistent slip rate discrepancies in the eastern California (USA) shear zone: Reply.
Geology, 45(9), e426-e426.
Evans, E. L. (2018). A Comprehensive analysis of geodetic slip-rate estimates and uncertainties in
California. Bulletin of the Seismological Society of America, 108(1), 1-18. doi: 10.1785/0120170159
Faulds, J. E., Henry, C. D., Spencer, J. E., & Titley, S. R. (2008). Tectonic influences on the spatial and
temporal evolution of the Walker Lane: An incipient transform fault along the evolving Pacific–North
American plate boundary. Ores and orogenesis: Circum-Pacific tectonics, geologic evolution, and ore
deposits: Arizona Geological Society Digest, 22, 437-470.
Faulds, J. E., Henry, C. D., & Hinz, N. H. (2005). Kinematics of the northern Walker Lane: An incipient
transform fault along the Pacific–North American plate boundary. Geology, 33(6), 505-508.
Field, E. H., Biasi, G. P., Bird, P., Dawson, T. E., Felzer, K. R., Jackson, D. D., Johnson, K.M., Jordan,
T.H., Madden, C., Michael, A.J., Milner, K.R., Page, M.T., Parsons, T., Powers, P.M., Shaw, B.E.,
Thatcher, W.R., Weldon, R.J., and Zeng, Y. (2015). Long‐term time‐dependent probabilities for the third
Uniform California Earthquake Rupture Forecast (UCERF3). Bulletin of the Seismological Society of
America, 105(2A), 511-543. doi: 10.1785/0120140093

154
Field, E. H., Jordan, T. H., Page, M. T., Milner, K. R., Shaw, B. E., Dawson, T. E., Biasi, G.P., Parsons,
T., Hardeback, J.L., Michael, A.J., Weldon, R.J., Powers, P.M., Johnson, K.M., Zeng, Y., Felzer, K.R.,
van der Elst, N., Madden, C., Arrowsmith, R., Werner, M.J., and Thatcher, W.R. (2017). A synoptic
view of the third Uniform California Earthquake Rupture Forecast (UCERF3). Seismological Research
Letters, 88, 1259–1267. doi:10.1785/0220170045
Fougere, D., Dolan, J., Rhodes, E., & McGill, S. (2024). Refined Holocene slip rate for the western and
central segments of the Garlock fault: record of alternating millennial-scale periods of fast and flow fault
slip. Seismica, 3(2).
Frankel, K.L., Glazner, A.F., Kirby, E., Monastero, F.C., Strane, M.D., Oskin, M.E., Unruh, J.R., Walker,
J.D., Anandakrishnan, S., Bartley, J.M. and Coleman, D.S. (2008). Active tectonics of the eastern
California shear zone. Field Guide. https://doi.org/10.1130/2008.fld011(03)
Friedrich, A. M., Lee, J., Wernicke, B. P., and Sieh, K. (2004). Geologic context of geodetic data across a
Basin and Range normal fault, Crescent Valley, Nevada. Tectonics, 23(2). doi: 10.1029/2003TC001528
Gan, W., Svarc, J. L., Savage, J. C., and Prescott, W. H. (2000). Strain accumulation across the Eastern
California Shear Zone at latitude 36 30′ N. Journal of Geophysical Research: Solid Earth, 105(B7),
16229-16236. doi: 10.1029/2000JB900105
Ganev, P. N., Dolan, J. F., Blisniuk, K., Oskin, M., & Owen, L. A. (2010). Paleoseismologic evidence for
multiple Holocene earthquakes on the Calico fault: Implications for earthquake clustering in the Eastern
California shear zone. Lithosphere, 2(4), 287-298.
Ganev, P. N., Dolan, J. F., McGill, S. F., and Frankel, K. L. (2012). Constancy of geologic slip rate along
the central Garlock fault: Implications for strain accumulation and release in southern California.
Geophysical Journal International, 190(2), 745-760.
Garfunkel, Z. (1974). Model for the late Cenozoic tectonic history of the Mojave Desert, California, and
for its relation to adjacent regions. Geological Society of America Bulletin, 85(12), 1931-1944.
Gath, E., & Rockwell, T. (2018). Paleoseismology of the Western Garlock Fault at Campo Teresa, Tejon
Ranch, Southern California. Active Tectonics and Archeoseismology Meeting. Report.

155
Gauriau, J., and Dolan, J. F. (2021). Relative structural complexity of plate‐boundary fault systems
controls incremental slip‐rate behavior of major strike‐slip faults. Geochemistry, Geophysics, Geosystems,
22(11). doi: 10.1029/2021GC009938
Gauriau, J., & Dolan, J. (2024). Comparison of geodetic slip-deficit and geologic fault slip rates reveals
that variability of elastic strain accumulation and release rates on strike-slip faults is controlled by the
relative structural complexity of plate-boundary fault systems. Seismica, 3(1). doi:
10.26443/seismica.v3i1.1119
Goes, S. D. B., & Ward, S. N. (1994). Synthetic seismicity for the San Andreas fault. Annals of
Geophysics, 37(6). http://doi.org/10.4401/ag-4150
Gold, R. D., Cowgill, E., Arrowsmith, J. R., Chen, X., Sharp, W. D., Cooper, K. M., and Wang, X. F.
(2011a). Faulted terrace risers place new constraints on the late Quaternary slip rate for the central Altyn
Tagh fault, northwest Tibet. Geological Society of America Bulletin, 123(5-6), 958-978. doi:
10.1130/B30207.1
Gold, R. D., and Cowgill, E. (2011b). Deriving fault-slip histories to test for secular variation in slip, with
examples from the Kunlun and Awatere faults. Earth and Planetary Science Letters, 301(1-2), 52-64. doi:
10.1016/j.epsl.2010.10.011
Guest, B., Pavlis, T. L., Golding, H., and Serpa, L. (2003). Chasing the Garlock: A study of tectonic
response to vertical axis rotation. Geology, 31(6), 553-556.
Hanks, T. C. (1975). Strong ground motion of the San Fernando, California, earthquake: ground
displacements. Bulletin of the Seismological Society of America, 65(1), 193-225.
Harvey, A. M., and Wells, S. G. (2003). Late Quaternary variations in alluvial fan sedimentologic and
geomorphic processes, Soda Lake basin, eastern Mojave Desert, California. Geological Society of
America Special Paper 368, Boulder, Colorado, 368. doi: 10.1130/0-8137-2368-X.207
Hatem, A. E., and Dolan, J. F. (2018). A model for the initiation, evolution, and controls on seismic
behavior of the Garlock fault, California. Geochemistry, Geophysics, Geosystems, 19(7), 2166-2178. doi:
10.1029/2017GC007349

156
Hatem, A. E., Dolan, J. F., Zinke, R. W., Langridge, R. M., McGuire, C. P., Rhodes, E. J., Brown, N., and
Van Dissen, R. J. (2020). Holocene to latest Pleistocene incremental slip rates from the east-central Hope
fault (Conway segment) at Hossack Station, Marlborough fault system, South Island, New Zealand:
Towards a dated path of earthquake slip along a plate boundary fault. Geosphere, 16(6), 1558-1584. doi:
10.1130/GES02263.1
Hearn, E. H., Pollitz, F. F., Thatcher, W. R., & Onishi, C. T. (2013). How do “ghost transients” from past
earthquakes affect GPS slip rate estimates on southern California faults?. Geochemistry, Geophysics,
Geosystems, 14(4), 828-838.
Hearn, E. (2022). “Ghost transient” corrections to the southern California GPS velocity field from San
Andreas fault seismic cycle models. Seismological Society of America, 93(6), 2973-2989.
Hill, M. L., & Dibblee Jr, T. W. (1953). San Andreas, Garlock, and Big Pine faults, California: a study of
the character, history, and tectonic significance of their displacements. Geological Society of America
Bulletin, 64(4), 443-458.
Jahns, R. H., & Wright, L. A. (1960). The Garlock and Death Valley fault zones in the Avawatz
Mountains, California. Geological Society of America Bulletin, 71, 2063.
Kirby, M. E., Heusser, L., Scholz, C., Ramezan, R., Anderson, M. A., Markle, B., Rhodes, E., Glover,
K.C., Fantozzi, J., Hiner, C., and Rangel, H. (2018). A late Wisconsin (32–10k cal a BP) history of
pluvials, droughts and vegetation in the Pacific south‐west United States (Lake Elsinore, CA). Journal of
Quaternary Science, 33(2), 238-254. doi: 10.1002/jqs.3018
Klinger, Y., M. Le Béon, and M. Al-Qaryouti. 2015. 5000 yr of paleoseismicity along the southern Dead
Sea fault. Geophysical Journal International 202: 313–327.
Kondo, H., Özaksoy, V., Yıldırım, C., Awata, Y., Emre, Ö., & Okumura, K. (2004). 3D trenching survey
at Demir Tepe site on the 1944 earthquake rupture, North Anatolian fault system, Turkey. Annual Report
on Active Fault and Paleoearthquake Research, 4, 231-242.
Kondo, H., Özaksoy, V., & Yıldirim, C. (2010). Slip history of the 1944 Bolu‐Gerede earthquake rupture
along the North Anatolian fault system: Implications for recurrence behavior of multisegment
earthquakes. Journal of Geophysical Research: Solid Earth, 115(B4).

157
Kozacı, Ö., Dolan, J. F., and Finkel, R. C. (2009). A late Holocene slip rate for the central North
Anatolian fault, at Tahtaköprü, Turkey, from cosmogenic 10Be geochronology: Implications for fault
loading and strain release rates. Journal of Geophysical Research: Solid Earth, 114(B1). doi:
10.1029/2008JB005760
LaViolette, J. W., Christenson, G. E., Stepp, J. C., Fife, D. L., & Brown, A. R. (1980). Quaternary
displacement on the western Garlock fault, southern California. Geology and mineral wealth of the
California desert, 449-456.
Libby, W. F., Anderson, E. C., & Arnold, J. R. (1949). Age determination by radiocarbon content: worldwide assay of natural radiocarbon. Science, 109(2827), 227-228.
Loomis, D. P., and Burbank, D. W. (1988). The stratigraphic evolution of the El Paso basin, southern
California: Implications for the Miocene development of the Garlock fault and uplift of the Sierra
Nevada. Geological Society of America Bulletin, 100(1), 12-28.
Madden, C., and Dawson, T. (2006). High-resolution dating of a complete Holocene earthquake record
for the central Garlock fault: Toward an understanding of earthquake recurrence, fault segment interaction
and regional earthquake clustering in southern California. SCEC Report.
Madden Madugo, C., Dolan, J. F., and Hartleb, R. D. (2012). New paleo-earthquake ages from the
western Garlock fault: Implications for regional earthquake occurrence in southern California. Bulletin of
the Seismological Society of America, 102(6), 2282-2299. doi: 10.1785/0120110310
Marco, S., Stein, M., Agnon, A., and Ron, H. (1996). Long‐term earthquake clustering: A 50,000‐year
paleoseismic record in the Dead Sea Graben. Journal of Geophysical Research: Solid Earth, 101(B3),
6179-6191.
McAuliffe, L. J., Dolan, J. F., Kirby, E., Rollins, C., Haravitch, B., Alm, S., and Rittenour, T. M. (2013).
Paleoseismology of the southern Panamint Valley fault: Implications for regional earthquake occurrence
and seismic hazard in southern California. Journal of Geophysical Research: Solid Earth, 118(9), 5126-
5146.

158
McClusky, S. C., Bjornstad, S. C., Hager, B. H., King, R. W., Meade, B. J., Miller, M. M., Monastero,
F.C., and Souter, B. J. (2001). Present day kinematics of the eastern California shear zone from a
geodetically constrained block model. Geophysical Research Letters, 28(17), 3369-3372. doi:
10.1029/2001GL013091
McGill, S., and Rockwell, T. (1998). Ages of late Holocene earthquakes on the central Garlock fault near
El Paso Peaks, California. Journal of Geophysical Research: Solid Earth, 103(B4), 7265-7279. doi:
10.1029/97JB02129
McGill, S. F., and Sieh, K. (1991). Surficial offsets on the central and eastern Garlock fault associated
with prehistoric earthquakes. Journal of Geophysical Research: Solid Earth, 96(B13), 21597-21621. doi:
10.1029/91JB02030
McGill, S., and Sieh, K. (1993). Holocene slip rate of the central Garlock fault in southeastern Searles
Valley, California. Journal of Geophysical Research: Solid Earth, 98(B8), 14217-14231. doi:
10.1029/93JB00442
McGill, S. F., Wells, S. G., Fortner, S. K., Kuzma, H. A., and McGill, J. D. (2009). Slip rate of the
western Garlock fault, at Clark Wash, near Lone Tree Canyon, Mojave Desert, California. Geological
Society of America Bulletin, 121(3-4), 536-554. doi: 10.1130/B26123.1
McGill, S. H. F. (1992). Paleoseismology and neotectonics of the central and eastern Garlock fault,
California [Doctoral dissertation], California Institute of Technology.
Meade, B. J., and Hager, B. H. (2005). Block models of crustal motion in southern California constrained
by GPS measurements. Journal of Geophysical Research: Solid Earth, 110(B3). doi:
10.1029/2004JB003209
Mildon, Z.K., Roberts, G.P., Faure Walker, J.P., Beck, J., Papanikolaou, I., Michetti, A.M., Toda, S.,
Iezzi, F., Campbell, L., McCaffrey, K.J. and Shanks, R. (2022). Surface faulting earthquake clustering
controlled by fault and shear-zone interactions. Nature Communications, 13(1), 7126.
Miller, M. M., Johnson, D. J., Dixon, T. H., and Dokka, R. K. (2001). Refined kinematics of the Eastern
California shear zone from GPS observations, 1993–1998. Journal of Geophysical Research: Solid Earth,
106(B2), 2245-2263. doi: 10.1029/2000JB900328

159
Monastero, F. C., Sabin, A. E., and Walker, J. D. (1997). Evidence for post-early Miocene initiation of
movement on the Garlock fault from offset of the Cudahy Camp Formation, east-central California.
Geology, 25(3), 247-250.
Moulin, A., & Cowgill, E. (2023). The Mojave Section of the San Andreas Fault (California), 2:
Pleistocene Records of Near‐Field Transpression Illuminate the Atypical Evolution of the Restraining
“Big Bend”. Geochemistry, Geophysics, Geosystems, 24(10), doi: 10.1029/2023GC010897
Ninis, D., Little, T. A., Van Dissen, R. J., Litchfield, N. J., Smith, E. G., Wang, N., Rieser, U., and
Henderson, C. M. (2013). Slip rate on the Wellington fault, New Zealand, during the late Quaternary:
Evidence for variable slip during the Holocene. Bulletin of the Seismological Society of America, 103(1),
559-579. doi: 10.1785/0120120162
Noriega, G. R., Arrowsmith, J. R., Grant, L. B., and Young, J. J. (2006). Stream channel offset and late
Holocene slip rate of the San Andreas fault at the Van Matre ranch site, Carrizo Plain, California. Bulletin
of the Seismological Society of America, 96(1), 33-47. doi: 10.1785/0120050094
Okumura, K., Rockwell, T. K., Duman, T., Tokay, F., Kondo, H., Yoldirim, C., & Ozaksoy, V. (2003).
Refined slip history of the North Anatolian fault at Gerede on the 1944 rupture. Eos (Transactions,
American Geophysical Union), 84(46), F715.
Onderdonk, N. W., McGill, S. F., and Rockwell, T. K. (2015). Short-term variations in slip rate and size
of prehistoric earthquakes during the past 2000 years on the northern San Jacinto fault zone, a major
plate-boundary structure in southern California. Lithosphere, 7(3), 211-234. doi: 10.1130/L393.1
Oskin, M., and Iriondo, A. (2004). Large-magnitude transient strain accumulation on the Blackwater
fault, Eastern California shear zone. Geology, 32(4), 313-316. doi: 10.1130/G20223.1
Oskin, M., Perg, L., Shelef, E., Strane, M., Gurney, E., Singer, B., and Zhang, X. (2008). Elevated shear
zone loading rate during an earthquake cluster in eastern California. Geology, 36(6), 507-510. doi:
10.1130/G24814A.1
Owen, L. A., Frankel, K. L., Knott, J. R., Reynhout, S., Finkel, R. C., Dolan, J. F., and Lee, J. (2011).
Beryllium-10 terrestrial cosmogenic nuclide surface exposure dating of Quaternary landforms in Death
Valley. Geomorphology, 125(4), 541-557. doi: 10.1016/j.geomorph.2010.10.024

160
Peltzer, G., Crampé, F., Hensley, S., and Rosen, P. (2001). Transient strain accumulation and fault
interaction in the Eastern California shear zone. Geology, 29(11), 975-978.
Pena, K. (2019). Paleoseismology of the central Garlock fault in Searles Valley, California [Master's
thesis], California State University, San Bernardino.
Platt, J. P., and Becker, T. W. (2013). Kinematics of rotating panels of E–W faults in the San Andreas
system: what can we tell from geodesy? Geophysical Journal International, 194(3), 1295-1301. doi:
10.1093/gji/ggt189
Plattner, C., Malservisi, R., Furlong, K. P., & Govers, R. (2010). Development of the Eastern California
Shear Zone—Walker Lane belt: The effects of microplate motion and pre-existing weakness in the Basin
and Range. Tectonophysics, 485(1-4), 78-84.
Plescia, J. B., & Henyey, T. L. (1982). Geophysical character of the proposed eastern extension of the
Garlock fault and adjacent areas, eastern California. Geology, 10(4), 209-214.
Ramos, M. D., Neo, J. C., Thakur, P., Huang, Y., & Wei, S. (2020). Stress changes on the Garlock fault
during and after the 2019 Ridgecrest earthquake sequence. Bulletin of the Seismological Society of
America, 110(4), 1752-1764.
Reimer, P., Austin, W., Bard, E., Bayliss, A., Blackwell, P., Bronk Ramsey, C., Butzin, M., Cheng, H.,
Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajdas, I., Heaton, T., Hogg, A., Hughen, K.,
Kromer, B., Manning, S., Muscheler, R., Palmer, J., Pearson, C., van der Plicht, J., Reimer, R., Richards,
D., Scott, E., Southon, J., Turney, C., Wacker, L., Adolphi, F., Büntgen, U., Capano, M., Fahrni, S.,
Fogtmann-Schulz, A., Friedrich, R., Köhler, P., Kudsk, S., Miyake, F., Olsen, J., Reinig, F., Sakamoto,
M., Sookdeo, A., and Talamo, S. (2020). The IntCal20 Northern Hemisphere radiocarbon age calibration
curve (0–55 cal kBP). Radiocarbon, 62. doi: 10.1017/RDC.2020.41
Rhodes, E. J. (2011). Optically stimulated luminescence dating of sediments over the past 200,000 years.
Annual Review of Earth and Planetary Sciences, 39(1), 461-488.
Rhodes, E. J. (2015). Dating sediments using potassium feldspar single-grain IRSL: initial
methodological considerations. Quaternary International, 362, 14-22. doi: 10.1016/j.quaint.2014.12.012

161
Rittase, W. M., Kirby, E., McDonald, E., Walker, J. D., Gosse, J., Spencer, J. Q., and Herrs, A. J. (2014).
Temporal variations in Holocene slip rate along the central Garlock fault, Pilot Knob Valley, California.
Lithosphere, 6(1), 48-58. doi: 10.1130/L286.1
Rockwell, T. K., Lindvall, S., Herzberg, M., Murbach, D., Dawson, T., & Berger, G. (2000).
Paleoseismology of the Johnson Valley, Kickapoo, and Homestead Valley faults: Clustering of
earthquakes in the eastern California shear zone. Bulletin of the Seismological Society of America, 90(5),
1200-1236.
Rockwell, T. K., Dawson, T. E., Young Ben-Horin, J., & Seitz, G. (2015). A 21-event, 4,000-year history
of surface ruptures in the Anza seismic gap, San Jacinto Fault, and implications for long-term earthquake
production on a major plate boundary fault. Pure and Applied Geophysics, 172, 1143-1165.
Rosenthal, J.S., Meyer, J., Palacios-Fest, M.R., Young, D.C., Ugan, A., Byrd, B.F., Gobalet, K. and
Giacomo, J. (2017). Paleohydrology of China Lake basin and the context of early human occupation in
the northwestern Mojave Desert, USA. Quaternary Science Reviews, 167, 112-139.
Saha, S., Moon, S., Brown, N. D., Rhodes, E. J., Scharer, K. M., McPhillips, D., McGill, S.F., and
Castillo, B. A. (2021). Holocene depositional history inferred from single‐grain luminescence ages in
southern California, North America. Geophysical Research Letters, 48(15). doi: 10.1029/2021GL092774
Salisbury, J. B., Arrowsmith, J. R., Brown, N., Rockwell, T., Akciz, S., and Ludwig, L. G. (2018). The
age and origin of small offsets at Van Matre Ranch along the San Andreas fault in the Carrizo Plain,
California. Bulletin of the Seismological Society of America, 108(2), 639-653. doi: 10.1785/0120170162
Savage, J. C., Prescott, W. H., Lisowski, M., and King, N. E. (1981). Strain accumulation in southern
California, 1973–1980. Journal of Geophysical Research: Solid Earth, 86(B8), 6991-7001. doi:
10.1029/JB086iB08p06991
Savage, J. C., Lisowski, M., and Prescott, W. H. (1990). An apparent shear zone trending north‐northwest
across the Mojave Desert into Owens Valley, eastern California. Geophysical Research Letters, 17(12),
2113-2116. doi: 10.1029/GL017i012p02113
Savage, J. C., Gan, W., and Svarc, J. L. (2001). Strain accumulation and rotation in the Eastern California
Shear Zone. Journal of Geophysical Research: Solid Earth, 106(B10), 21995-22007. doi:
10.1029/2000JB000127

162
Scharer, K. M., Biasi, G. P., & Weldon, R. J. (2011). A reevaluation of the Pallett Creek earthquake
chronology based on new AMS radiocarbon dates, San Andreas fault, California. Journal of Geophysical
Research: Solid Earth, 116(B12).
Scharer, K., Weldon, R., Biasi, G., Streig, A., & Fumal, T. (2017). Ground‐rupturing earthquakes on the
northern Big Bend of the San Andreas fault, California, 800 AD to present. Journal of Geophysical
Research: Solid Earth, 122(3), 2193-2218.
Seitz, G., Weldon II, R., & Biasi, G. P. (1997). The Pitman Canyon paleoseismic record: A re-evaluation
of southern San Andreas fault segmentation. Journal of Geodynamics, 24(1-4), 129-138.
Sieh, K., and Wallace, R. E. (1987). The San Andreas fault at Wallace Creek, San Luis Obispo County,
California. Cordilleran Section of the Geological Society of America, Centennial Field Guide. doi:
10.1130/0-8137-5401-1.233
Smith, G. I., and Ketner, K. B. (1970). Lateral displacement on the Garlock fault, southeastern California,
suggested by offset sections of similar metasedimentary rocks. United States Geological Survey
Professional Paper, 700.
Smith, G. I. (1962). Large Lateral Displacement on Garlock Fault, California, as Measured from Offset
Dike Swarm. American Association of Petroleum Geologists Bulletin, 46(1), 85-104.
Smith, G.I., (1964). Geology and volcanic petrology of the Lava Mountains, San Bernardino County,
California. United States Geological Survey Professional Paper, 457, p.97.
Smith, G. I., Barczak, V. J., Moulton, G. F., & Liddicoat, J. C. (1983). Core KM-3, a surface-to-bedrock
record of late Cenozoic sedimentation in Searles Valley, California (No. 1256). USGPO.
Smith, G. I. (2009). Late Cenozoic geology and lacustrine history of Searles Valley, Inyo and San
Bernardino Counties, California (No. 1727). US Geological Survey.
Smith, G. I., & Stuiver, M. (1979). Subsurface stratigraphy and geochemistry of late Quaternary
evaporites, Searles Lake, California, with a section on radiocarbon ages of stratigraphic units (No. 1043).
US Govt. Print. Off.

163
Toda, S., & Stein, R. S. (2020). Long‐and short‐term stress interaction of the 2019 Ridgecrest sequence
and Coulomb‐based earthquake forecasts. Bulletin of the Seismological Society of America, 110(4), 1765-
1780.
Wallace, R. E. (1987). Grouping and migration of surface faulting and variations in slip rates on faults in
the Great Basin province. Bulletin of the Seismological Society of America, 77(3), 868-876. doi:
10.1785/BSSA0770030868
Weldon, R., Scharer, K., Fumal, T., and Biasi, G. (2004). Wrightwood and the earthquake cycle: What a
long recurrence record tells us about how faults work. Geological Society of America today, 14(9), 4-10.
Wesnousky, S. G. (2005). The San Andreas and Walker Lane fault systems, western North America:
Transpression, transtension, cumulative slip and the structural evolution of a major transform plate
boundary. Journal of Structural Geology, 27(8), 1505-1512.
Zeng, Y., Petersen, M., & Boyd, O. (2022). Lower seismogenic depth model for western US earthquakes.
Seismological Society of America, 93(6), 3186-3204.
Zinke, R., Dolan, J. F., Rhodes, E. J., Van Dissen, R., & McGuire, C. P. (2017). Highly variable latest
Pleistocene-Holocene incremental slip rates on the Awatere fault at Saxton River, South Island, New
Zealand, revealed by lidar mapping and luminescence dating. Geophysical Research Letters, 44, 11,301–
11,310. doi: 10.1002/2017GL075048
Zinke, R., Dolan, J. F., Rhodes, E. J., Van Dissen, R., McGuire, C. P., Hatem, A. E., Brown, N.D., and
Langridge, R. M. (2019). Multimillennial incremental slip rate variability of the Clarence fault at the
Tophouse Road site, Marlborough fault system, New Zealand. Geophysical Research Letters, 46(2), 717-
725. doi: 10.1029/2018GL080688
Zinke, R., Dolan, J. F., Rhodes, E. J., Van Dissen, R. J., Hatem, A. E., McGuire, C. P., Brown, N.D., and
Grenader, J. R. (2021). Latest Pleistocene–Holocene Incremental Slip Rates of the Wairau Fault:
Implications for Long‐Distance and Long‐Term Coordination of Faulting Between North and South
Island, New Zealand. Geochemistry, Geophysics, Geosystems, 22(9). doi: 10.1029/2021GC009656

Abstract (if available)

Abstract Documenting the behavior of faults and fault systems is fundamental to understanding the driving mechanisms that govern their activity. Mechanisms operating on faults and fault systems are complex, interconnected, and poorly understood. Currently, there are too few comprehensive data sets of paired incremental fault slip rates and paleo-earthquake ages from fault systems to allow the comparisons necessary to fully determine how plate-boundary slip is partitioned amongst major faults in time and space. This dissertation focuses on developing a comprehensive slip rate and paleo-earthquake record
spanning the Holocene for the Garlock fault, a major strike-slip fault within the southern California plate boundary network that exhibits irregular earthquake recurrence and slip rate through time. Expanding and resolving the incremental slip rate and paleo-earthquake record for the Garlock fault provides valuable insights into fault behavior, enhancing our understanding of fault mechanics, potential interactions between faults, and provides data necessary to facilitate the refinement of next-generation seismic hazard assessments.

To generate a paired incremental fault slip rate and paleo earthquake record for the Garlock fault, I use lidar digital topographic data to measure offset landforms, infrared stimulated luminescence dating (IRSL), 14C radiocarbon dating, and paleoseismic trenching methods. In this dissertation, I add 12 new and revised slip rates for the Garlock fault and constrain the timing of 14 paleo-earthquakes on the central Garlock fault. These results yield one of the longest and most detailed records of combined incremental fault slip and paleo-earthquakes on any strike-slip fault in the world. These new slip rate data support the idea that the Garlock fault experiences significant temporal variations (two-to-five fold) in slip rate spanning multiple earthquake cycles, corresponding with periods of multi-millennial earthquake clustering punctuated by periods of reduced earthquake recurrence and slower slip rate. New paleo-earthquake data from the Koehn Lake trench confirm that earthquake clustering is a common pattern of earthquake recurrence over the past ~13 ky, with two new temporal four-earthquake clusters documented at 6-7 ka and 8-10 ka. The results of this dissertation provide additional constraints that facilitate a more complete understanding of fault system phenomena, provide insights into long-term fault behavior, and highlight the role of the Garlock fault in the plate boundary system.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

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

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

Observations of temporal and spatial patterns of strain accommodation and earthquake occurrence along strike-slip faults of New Zealand and southern California, USA

PDF

Implications of new fault slip rates and paleoseismologic data for constancy of seismic strain release and seismic clustering in the eastern California shear zone

PDF

Constancy of strain release rates along the North Anatolian fault

PDF

The impact of faulting and folding on earthquake cycles in collisional orogens

PDF

Seismicity distribution near strike-slip faults in California

PDF

Multi-scale damage signatures across major strike-slip faults

PDF

Direct observation of fault zone structure and mechanics in three dimensions: a study of the SEMP fault system, Austria

PDF

Fault slip rates, constancy of seismic strain release, and landscape evolution in the eastern California shear zone

PDF

Rupture synchronicity in complex fault systems

PDF

Volumetric interactions between major ruptures and fault zones illuminated by small earthquake properties

PDF

Microseismicity, fault structure, & the seismic cycle: insights from laboratory stick-slip experiments

PDF

From laboratory friction to numerical models of fault dynamics

PDF

Multi-scale imaging of major fault zones in Southern California

PDF

Heterogeneity of earthquake stress drops, focal mechanisms and active fault zones

PDF

Seismic velocity structure and spatiotemporal patterns of microseismicity in active fault zones of Southern California

PDF

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

Tectonic control on landsliding revealed by the 2015 Gorkha earthquake

Asset Metadata

Creator Fougere, Dannielle (author)

Core Title Revealing spatiotemporal patterns in fault-system behavior: insights from multi-millennial earthquake clustering and slip rate variability on the Garlock fault

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Degree Conferral Date 2025-05

Publication Date 02/06/2025

Defense Date 01/22/2025

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag earthquake clustering,fault,Garlock fault,slip rate variation

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Dolan, James (committee chair), Nutt, Steve (committee member), Vidale, John (committee member)

Creator Email dfougere.personal@gmail.com,dfougere@usc.edu

Unique identifier UC11399GTHX

Identifier etd-FougereDan-13819.pdf (filename)

Legacy Identifier etd-FougereDan-13819

Document Type Dissertation

Format theses (aat)

Rights Fougere, Dannielle

Internet Media Type application/pdf

Type texts

Source 20250211-usctheses-batch-1241 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu