Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Symmetry properties, pulverized rocks and damage architecture as signatures of earthquake ruptures
(USC Thesis Other)
Symmetry properties, pulverized rocks and damage architecture as signatures of earthquake ruptures
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
i
SYMMETRY PROPERTIES, PULVERIZED ROCKS AND DAMAGE
ARCHITECTURE IN FAULT ZONES AS SIGNATURES OF EARTHQUAKE
RUPTURES
by
Ory Dor
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)
August 2007
Copyright 2007 Ory Dor
ii
Epigraph:
anitche
iii
Dedication:
This thesis is dedicated with love to my parents, Chaya and David Dor, who made
my education possible. They most certainly have the largest effect on the choices I
made.
iv
Acknowledgements:
I would like to express my deep appreciation to my advisor, Yehuda Ben-Zion,
who was always attentive to any scientific, administrative or personal concern I
raised. As a mentor, his primary concern was always my professional and personal
development and success. I want to thank Tom Rockwell, my geologist advisor, for
his consistent generosity in help and support, and for sharing with me his wide and
deep knowledge of Earth sciences, while always keeping a high spirit. Jim Brune and
Judi Chester were both accessible for consultation, providing constructive comments
in many instances during my 5 year long work. I also thank Charlie Sammis for a
fruitful collaboration and Toshi Shimamoto for discussions and advice. My
colleagues Cengiz Yildirim and Matt Sisk became my friends during our
collaborative work. My special thanks to Cindy White, the department’s academic
advisor, for her care which went far beyond my academic standing.
At least half of this work’s credit should go to my wife, Rachel. Besides
significant help with field work, her wise advice on matters related and not related to
geology together with unconditional love through joyful and tough times carried me
in my work.
v
Table of contents
Epigrap ..……………………………………………………………………………..ii
Dedication ................................................................................................................... iii
Acknowledgements..................................................................................................... iv
List of Tables ............................................................................................................ viii
List of Figures ............................................................................................................. ix
Abstract ..................................................................................................................... xiii
Introduction.................................................................................................................. 1
Chapter 1: Geologic observations of damage asymmetry in the structure of the
San Jacinto, San Andreas and Punchbowl faults in southern
California: A possible indicator for preferred rupture propagation
direction.................................................................................................... 8
Abstract.................................................................................................................. 8
Introduction............................................................................................................ 9
Approach and methodology................................................................................. 16
Field Observations ............................................................................................... 22
The San Jacinto Fault........................................................................................ 23
The Mojave Section of the San Andreas Fault ................................................. 32
The Punchbowl Fault........................................................................................ 51
Discussion............................................................................................................ 63
Asymmetry of structural properties in light of velocity structure .................... 64
Historic earthquake behavior that might be explained by preferred rupture
propagation direction. ....................................................................................... 68
Possible Related Observations.......................................................................... 72
Possible Interpretation Problems ...................................................................... 75
Conclusions.......................................................................................................... 81
Chapter 2: Pulverized Rocks in the Mojave section of the San Andreas Fault
Zone........................................................................................................ 84
Abstract................................................................................................................ 84
Introduction.......................................................................................................... 85
Observations ........................................................................................................ 90
vi
Approach........................................................................................................... 90
Spatial distribution of PFZR in the San Andreas damage zone, Mojave ......... 95
Classes of pulverization and fault-zone scale mapping results ........................ 99
Pulverization of sedimentary rocks ................................................................ 104
Discussion.......................................................................................................... 105
Geologic observations related to the possible depth of pulverization ............ 106
Correlation of mapping results with geophysical observations...................... 110
Relations of the mapping results to theoretical predictions............................ 110
Comparison between the structures of the active SAF and exhumed faults of
the SAF system ............................................................................................... 112
Summary............................................................................................................ 114
Chapter 3: Geologic and geomorphologic asymmetry across the rupture zones
of the 1943 and 1944 earthquakes on the North Anatolian Fault:
possible signals for preferred earthquake propagation direction.......... 117
Abstract.............................................................................................................. 117
Introduction........................................................................................................ 119
Observations ...................................................................................................... 123
Approach and methodology............................................................................ 123
Geologic observations .................................................................................... 130
Geomorphologic observations ........................................................................ 152
Discussion.......................................................................................................... 167
Summary and synthesis of mapping results.................................................... 167
Possible implications of structural asymmetry ............................................... 175
Chapter 4: Damage characterization in sandstones along the Mojave section of
the San Andreas Fault with a new method: initial results and
implications for the depth and mechanism of dynamic rock
pulverization ......................................................................................... 180
Abstract.............................................................................................................. 180
Introduction........................................................................................................ 182
Geologic setting and sampling locations ........................................................... 185
Research approach ............................................................................................. 188
Methodology...................................................................................................... 190
Samples extraction and preparation................................................................ 190
Image analysis ................................................................................................ 192
Fracture orientation measurements................................................................. 194
Observations ...................................................................................................... 197
Description of the Juniper Hills host rock (Based on a sample taken 670 m
from the fault) ................................................................................................. 197
Common damage features present in all/most samples.................................. 198
FIPL measurements ........................................................................................ 201
vii
Orientation of microfractures ......................................................................... 209
Discussion.......................................................................................................... 210
Precision of measurements ............................................................................. 210
Structure of the damage zone (on the southwest side of the SAF)................. 215
Mechanical interpretation of damage fabric ................................................... 216
Implications of results for dynamic rock failure............................................. 219
Summary............................................................................................................ 221
Chapter 5: The geometry of slip surfaces in the hanging-wall of the Sierra
Madre fault, La-Canada, California: evidence for Mohr-Coulomb
failure induced by a dynamic off-fault stress field............................... 223
Abstract.............................................................................................................. 223
Introduction........................................................................................................ 224
Regional geological setting................................................................................ 227
Local geological setting ..................................................................................... 228
Methodology...................................................................................................... 229
Observations ...................................................................................................... 231
Mesoscale observations .................................................................................. 231
Microscale observations ................................................................................. 236
Discussion.......................................................................................................... 243
Activation of cm-scale slip surfaces during earthquake ruptures................... 243
Microfractures and pulverization.................................................................... 250
Damage accumulation over many seismic cycles .......................................... 253
Summary............................................................................................................ 255
Bibliography............................................................................................................. 257
Appendices............................................................................................................... 279
Appendix 1: Shapiro-Wilk and Shapiro-Francia W tests for normality of the
distribution of FIPL results in Chapter 4. .......................................................... 279
Appendix 2: Bitmap images used for FIPL measurements in Chapter 4........... 281
viii
List of Tables
Table 1.1: Devil’s Punchbowl Linear fracture density. .............................................62
Table 2.1: Cumulative area of outcrops with different damage levels SW and
NE of the fault...........................................................................................97
Table 3.1: Definitions of morphometric parameters................................................130
Table 3.2: measured length of river valley sections that flow north and south
with respect to the fault along the 1943 (43) and 1944 (44) rupture
sections....................................................................................................156
Table 3.3: comparison of variables that may control and affect erosion
between the two terrains (figures 3.10, 3.11). Values correspond to
the entire terrain and are not the mean values for all the basins. ............159
Table 3.4: results of morphometric analysis of two correlative terrains
displaced by the 1944 rupture section. ....................................................163
Table 4.1: Area-weighted average FIPL and standard deviation for the 104
grains population analyzed from sample 8E-b and for each of the
40-grains sets chosen from this population.............................................203
Table 4.2: summary of measurements done on the JHF samples. ...........................204
Table 4.3: summary of measurements done on samples 8E. ...................................205
Table 4.4: Rose diagrams showing the distribution of orientations for the
various fracture types from samples 8E-b and 31 with
corresponding statistics. ..........................................................................210
Table 5.1: Properties of measured fracture sets in samples 183 and 4. ...................241
ix
List of Figures
Figure 1.1: Schematic illustration of rupture modes of small to moderate (M <
6.5) and large (M>7) earthquakes on a strike-slip fault..........................10
Figure 1.2: properties of the wrinkle-like pulse on a material interface. .....................10
Figure 1.3: Simulations of rupture along biomaterial interface by Ben-Zion and
Shi (2005)................................................................................................14
Figure 1.4: Location map. ............................................................................................23
Figure 1.5: Structural units and fracture density in the gouge of the San Jacinto
fault zone south of the Ramona Indian Reservation. ..............................26
Figure 1.6: Fault core of the SAF at the Paleoseismicity site south of the town
of Little Rock. .........................................................................................37
Figure 1.7: Locations of the Little Rock Creek study site and features of the
SAF near by. ...........................................................................................40
Figure 1.8: Trench log from the Little Rock Creek......................................................41
Figure 1.9: A trench log displaying 45 m long cross section of the San Andreas
fault zone near Palmdale.........................................................................45
Figure 1.10: Geologic map of the Punchbowl fault in the South Fork and
Devil’s Chair study areas (after Dibblee 2002). .....................................54
Figure 1.11: Fracture density profile across the Punchbowl fault near South
Fork.........................................................................................................56
Figure 1.12: Geomorphologic expression of the damage pattern across the
Punchbowl fault. .....................................................................................59
Figure 1.13: Major geographic features in the Devil’s Chair area and locations
of the FD measurement sites marked with white dots ............................61
x
Figure 1.14: The south-central San Andreas Fault system in California and the
known extent of the partially overlapping 1812 and 1857 ruptures .......69
Figure 2.1: The south central San Andreas Fault (SAF) system with its major
strands. ....................................................................................................91
Figure 2.2: Map of crystalline plutonic rocks in the damage zone of the SAF,
classified according to their damage pattern...........................................94
Figure 2.3: A large scale distribution pattern of crystalline rocks and their
damage pattern between Tejon Pass and Cajon Pass............................101
Figure 2.4: A detailed distribution of damaged fault zone rocks in Quail Lake –
Sawmill Mtn Ranch ..............................................................................102
Figure 2.5: A detailed distribution of damaged fault zone rocks in a site
northwest of Lake Hughes ....................................................................103
Figure 3.1: Ruptures along the North Anatolian Fault in the 20th Century...............120
Figure 3.2: Rupture radiation field and damage generation about a biomaterial
interface.................................................................................................121
Figure 3.3: Geomorphologic map of the area around the trenching site near
Ladik. ....................................................................................................131
Figure 3.4: Photomosaic and corresponding maps of a trench wall from Ladik........134
Figure 3.5: Fault core in Ladik...................................................................................138
Figure 3.6: Trench log of the fault core in Bademci. .................................................141
Figure 3.7: A system of six trenches excavated near Celtikci....................................145
Figure 3.8: Geological map of a fault section near Hamamle....................................150
Figure 3.9: Location of fault trace with respect to river valleys. ...............................154
xi
Figure 3.10: Slope-aspect analysis. ............................................................................158
Figure 3.11: Automatically delineated drainages in the two correlative terrains.......160
Figure 3.12: Correlations between morphometric parameters. ..................................162
Figure 3.13: Bad-land topography in two sites along the 1943 rupture. ....................166
Figure 4.1: Working area and sampling locations......................................................188
Figure 4.2: Exposure of the JHF 10 m southwest of the SAF in the Littlerock
paleoseismicity site. ..............................................................................192
Figure 4.3: Intermediate and final products of the analysis process ..........................193
Figure 4.4: Fracture types...........................................................................................196
Figure 4.5: Microscale damage features.....................................................................200
Figure 4.6: Distribution of FIPL values for 104 grains in sample 8E-b.....................201
Figure 4.7: FIPL as a function of original grain area for 104 grains in sample
8E-b.......................................................................................................202
Figure 4.8: Photomicrographs of Hungry Valley formation ......................................207
Figure 4.9: Possible artifacts shown on a single grain ...............................................212
Figure 4.10: SEM images of frames from the grain in Figure 4.9 .............................214
Figure 4.11: A schematic illustration of sample 8E ...................................................219
Figure 5.1: Model predictions ....................................................................................225
Figure 5.2: Location of the working site near La-Canada, California........................227
Figure 5.3: Geology of the Arroyo Seco canyon at its mouth....................................229
xii
Figure 5.4: kinematic indicators used in this study ....................................................230
Figure 5.5: Local geological settings in the mapping site..........................................232
Figure 5.6: Meso-scale structure of the Sierra Madre fault near JPL.........................233
Figure 5.7: Small-scale slip surfaces. .........................................................................234
Figure 5.8: Distribution of strike and dip of slip surfaces, 115 data. .........................235
Figure 5.9: Lower hemisphere equal area projections of normals and slip
vectors...................................................................................................236
Figure 5.10: A transmitted light photomicrograph of sample 0 .................................238
Figure 5.11: Transmitted light photomicrographs of samples 183 and 4...................240
Figure 5.12: Rose diagrams........................................................................................241
Figure 5.13: Photomicrographs of single grains.........................................................242
Figure 5.14: Projection of model prediction on the Sierra Madre fault geometry .....244
Figure 5.15: The contoured data set of 115 normals to slip surfaces projected
over the prediction of Figure 5.14c.......................................................246
Figure 5.16: Rotated predicted data compared with observed data............................248
xiii
Abstract
Structural symmetry properties were mapped across faults of the San Andreas
Fault (SAF) and North Anatolian Fault (NAF) systems, at various scales and several
sites on each fault. Fractures on a fault-core scale, subsidiary faults and fault rocks
on a fault-zone scale and pulverized rocks on a damage-zone scale show
systematically asymmetry. On the SAF, San Jacinto and Punchbowl faults the
northeast side is more damaged. On the NAF 1943 and 1944 rupture sections the
south and north sides, respectively, are more damaged. Asymmetric erosion patterns
along the NAF including locations of river valleys with respect to the fault and
contrast in drainage density and other morphometric parameters across the fault, are
consistent with the geologically mapped structural asymmetry. These asymmetric
patterns are compatible with preferred rupture directions northwestward on faults of
the SAF system, and eastward and westward on the 1943-1944 rupture sections of
the NAF, respectively (as occurred in these two earthquakes). Tomographic studies
show that the northeast side of the SAF and the San Jacinto fault have faster seismic
velocities at depth. Significant damage content in sedimentary rocks of the Juniper
Hills formation near the SAF in the central Mojave section indicates that dynamic
generation of damage can occur close to the Earth surface, in agreement with other
indications for minimal exhumation of damaged fault zone rocks. An asymmetric
shallow damage structure correlated with the velocity structure at depth is a
predicted outcome for rupture along a bimaterial interface (Ben-Zion and Shi, 2005).
xiv
Microfractures in the Juniper Hills rocks near the fault, orientated preferably normal
to its strike, are compatible with the transient stress field associated with seismic slip
events on frictional rough surfaces (Chester and Chester, 2000). The damage fabric
is anisotropic, rich with compressional features, and therefore not compatible with an
absolute tension.
Structural analysis of orientation and slip data of 115 slip surfaces in the
hanging-wall of the Sierra Madre fault near JPL shows that their geometry and
kinematics are compatible with Mohr-Coulomb failure associated with the stress
field of propagating mode II ruptures with slip weakening (Rice et al. 2005).
1
Introduction
The architecture and composition of Fault Zone (FZ) in seismo-tectonic
environments are shaped by processes occurring during the earthquake cycle,
primarily during the propagation of earthquake ruptures. This is associated, along
with other phenomena, with the progressive removal of fault jogs and other
geometrical complexities with cumulative slip (e.g. Wesnousky, 1988; Ben-Zion and
Sammis, 2003) and with dynamic generation of off-fault damage during rupture
propagation as part of the rupture’s process zone (e.g. Andrews, 2005; Ben-Zion and
Shi, 2005; Rice et al., 2005). Furthermore, earthquake mechanics is affected by the
structure of FZs. This involves geometrical perturbations in fault structures that
interact with the rupture and allow its initiation, arrest, and the change of its speed or
direction (Poliakov et al., 2002), the influence of fluids and FZ hydrology on rupture
properties (Wibberly and Shimamoto, 2003) and the influence of material contrast
across the fault on dynamic weakening and preferred rupture direction (e.g.
Weertman, 1980; Shi and Ben-Zion, 2006). Those examples illustrate the tight
interactions that earthquakes have with FZ structures and require that the study of the
earthquakes process will include a considerable attention to the geometrical,
mechanical and compositional properties of FZs.
Several theoretical works that discuss unresolved problems in geophysics
presented clear predictions about the structure of FZs and motivated this body of
work. Simulations of mode-II ruptures along bimaterial interfaces show that in such
cases ruptures tend to propagate in the motion direction of the slower velocity
2
medium. With such rupture direction, the tensional quadrant of the radiated seismic
field is on the faster velocity medium, where more damage is expected to develop at
shallow depth (e.g. Ben-Zion, 2001; Ben-Zion and Shi, 2005). Over geologic time
this behavior will lead to an asymmetric pattern of rock damage across the interface.
The work of Rice et al. (2005) showed that a slip-weakening rupture front with a
self-healing pulse can cause an inelastic respond of the wall rock exceeding a zero-
cohesion Mohr-Coulomb failure criterion, leading to the activation of favorably
oriented conjugate set of slip surfaces with a specific geometry over a length scale
proportional to some rupture properties. Laboratory work and modeling by Brune et
al. (1993) and others showed that the passage of ruptures are associated with normal
interface vibrations leading to dynamic loading and un-loading of the fault, possibly
producing an opening mode. Such behavior may be associated with a complete
reduction of normal stress and the likely dilation of the wall-rock. These models
predict several diagnostic signatures of rock damage which are tested by detailed
field studies of FZ structures in this thesis.
Valuable information about FZ structure is obtained from studies of exhumed
faults (e.g. Chester et al., 2004), although those are associated with interpretation
problems due to uncertainties regarding the past faulting style (e.g. seismic vs.
aseismic), the exact exhumation history and alteration of FZ fabric. Studies of faults
in deep mines (e.g. McGarr et al., 1979) allow correlation of FZ elements with
available seismic data of the earthquakes during which they were generated, but the
mining environment is drained from groundwater, includes many unnatural stress
3
concentrations and is able of generating up to moderate size earthquakes only.
Drilling into faults (e.g. SAFOD) allows the extraction of highly detailed data sets
from the fault core and damage zone, but the information is approximately one
dimensional. Complimentary to those direct observations are high resolution seismic
imaging of faults (e.g. McGuire and Ben-Zion, 2005; Lewis et al., 2005) in which the
fault interface and the damage structure are sometimes imaged at a resolution of 10s
of meters or less.
In this work observations regarding the symmetry of FZ structural properties,
the geometry of slip surfaces, the distribution of pulverized rocks and the microscale
fabric of fault rocks were made along major active and exhumed faults of the
southern San Andreas and North Anatolian fault systems while using information
obtained in other studies by tomographic methods. Rocks within those FZs record, in
addition to the imprints of earthquake ruptures, signature of previous deformation
phases, surface weathering and the possible expressions of various site effects. To
overcome the effects that those sources of noise may have on the interpretation of the
collected data, multi-scale observations were made in many locations along several
major faults. The observations span length scales from the micron to the regional
through thin sections analysis, detailed fracture mapping in natural and artificial
exposures, mapping of subsidiary faults and fault rocks across FZs and regional
mapping of pulverized rocks. This study utilized several types of observations, some
of them were developed during its course, including electron and optical microscopy,
image analysis, structural analysis techniques, outcrop mapping, trenching and GIS.
4
In the first chapter of this manuscript, “Geologic observations of damage
asymmetry in the structure of the San Jacinto, San Andreas and Punchbowl faults in
southern California: A possible indicator for preferred rupture propagation
direction” (Dor et al., PAGEOPH 2006), observations are made in several sites in the
centimeter to meter fault core scale and meters to 10’s of meters FZ scale. The
observations show that the northeast side of these faults is systematically more
damaged. The same sense of asymmetry is observed by seismic imaging of a shallow
low velocity trapping structure southeast of the small scale mapping sites along the
San Jacinto fault (Lewis et al., 2005). This sense of asymmetry suggests that
earthquake ruptures propagate (or propagated, on the exhumed Punchbowl),
preferably to the northwest. The northeast more damaged sides of the San Jacinto
and San Andreas faults are on the blocks with faster seismic velocities.
In the second chapter, “Pulverized Rocks in the Mojave section of the San
Andreas Fault Zone” (Dor et al., EPSL 2006), mapping of pulverized crystalline
rocks along the Mojave section of the San Andreas Fault (SAF) shows that they are a
systematic damage product of the fault, forming a ~100 m wide tabular zone parallel
to its slipping zone. This zone is apparently shifted to the northeast side of the fault,
in agreement with the sense of asymmetry of the smaller scale damage elements
discussed in the first chapter. A line of evidence suggests that those rocks, with a
reduction of the grain size and apparent lack of original fabric distortion, were
pulverized at a relatively shallow depth (the top ~3 km of the crust). The properties
of the pulverized rocks layer is compatible with characteristics of low velocity
5
trapping structures that were seismically imaged elsewhere.
In the third chapter, “Geologic and geomorphologic asymmetry across the
rupture sections of the 1943 and 1944 earthquakes on the North Anatolian Fault,
Turkey: possible signals for a preferred rupture direction.”, geologic observations
made in trenches and natural exposures imply more damage accumulation on the
south side of the 1943 rupture zone and on the north side of the 1944 rupture zone.
The same sense of structural asymmetry was inferred from geomorphological
observations of erosion patterns including the location of river valleys with respect to
the fault, analysis of drainage density and other morphometric parameters on
correlative terrains across the fault and comparison of gully networks in bad-land
topography between the two sides of the fault. The inferred damage asymmetry is
compatible with preferred eastward and westward rupture directions on the 1943 and
1944 fault sections, respectively, in agreement with the propagation directions of
these two recent earthquakes.
In the forth chapter, “Damage characterization in sandstones along the Mojave
section of the San Andreas Fault with a new method: initial results and implications
for the depth and mechanism of dynamic rock pulverization”, an implementation of a
new image analysis method delineates a damage zone at the order of about a 100 m
on the southwest side of the SAF in sandstones that were never deeply buried while
displaced along the fault. Damage fabric is found to be anisotropic with
microfractures near the fault preferably oriented normal to its strike. The
observations suggest that dynamic generation of damage occurs close to the surface
6
of the Earth and that this type of damage is not associated with isotropic dilation of
the rock.
In the fifth chapter, “The geometry of slip surfaces in the hanging-wall of the
Sierra Madre fault, La-Canada, California: an expression for Mohr-Coulomb failure
induced by dynamic off-fault stress field.”, detailed mapping of 115 small scale slip
surfaces and the kinematic indicators found on them in the granitic hanging-wall of
the active Sierra Madre fault (on the foothills of the San Gabriel Mountains near Los
Angeles), shows that they are systematically arranged in a conjugate set with their
strike oblique in an acute angle to that of the fault and their inclinations steeper than
that of the fault surface. A layer of rock above the fault surface of one m width is
pulverized with the igneous fabric largely undisturbed. Crystals in a sample
immediately above the fault plane are intensely shattered but their shape is mostly
preserved.
The observations discussed in chapters 1 to 3 show persistent structural
asymmetry across the structure of major strike slip faults. In two of the cases, that of
the SAF and the San Jacinto fault, available information on the local and regional
velocity structure can be correlated with the sense of structural asymmetry. Various
observations described in chapters 2 and 4 are compatible with generation of the
mapped damage elements at shallow depth. These properties of the damage are
compatible with predictions of the theory of rupture along a bimaterial interface
(Ben-Zion and Shi, 2005). Initial results of fracture analysis described in chapter 4
suggest that the maximum transient compressive stress within the fault damage zone
7
during slip events was normal to the fault, in agreement with wear models and stress
concentrations due to slip along frictional rough surfaces (Chester and Chester,
2000). The anisotropic damage fabric in those rocks is inconsistent with dynamic
opening and complete reduction of normal stress, but compatible with the southwest
side been on the compressional quadrant of northwestward propagating ruptures, the
preferred direction inferred for earthquakes along this fault section. The geometry
and kinematics of the slip surfaces mapped in the Sierra Madre fault match
predictions for favorably oriented slip surfaces activated when the Mohr-Coulomb
yielding criterion is exceeded in a region affected by the stress field of mode II
ruptures with slip weakening (Rice et al., 2005). Shattering and dilation of grains
immediately above the fault plane can possibly be explained by an opening mode
operating at different depth or at a different stage of the fault development.
This work illustrates the possible feedback between theory and observations in
the field of earthquake sciences. Further scientific and social implications of the
findings are discussed at various places throughout the manuscript.
8
Chapter 1: Geologic observations of damage asymmetry in the structure
of the San Jacinto, San Andreas and Punchbowl faults in
southern California: A possible indicator for preferred
rupture propagation direction.
Published at PAGEOPH, 2006
Co-authors: Tom Rockwell
1
and Yehuda Ben-Zion
2
1. Department of Geological Sciences, San Diego State University, San
Diego, CA 92182-1020, USA.
2. Department of Earth Sciences, University of Southern California, Los
Angeles, CA 90089-0740, USA.
Abstract
We present new in-situ observations of systematic asymmetry in the pattern of
damage expressed by fault zone rocks along sections of the San Andreas, San
Jacinto, and Punchbowl faults in southern California. The observed structural
asymmetry has consistent manifestations at a fault core scale of millimeters to
meters, a fault zone scale of meters to tens of meters and related geomorphologic
features. The observed asymmetric signals are in agreement with other geological
and geophysical observations of structural asymmetry in a damage zone scale of tens
to hundreds of meters. In all of those scales, more damage is found on the side of the
fault with faster seismic velocities at seismogenic depths. The observed correlation
9
between the damage asymmetry and local seismic velocity structure is compatible
with theoretical predictions associated with preferred propagation direction of
earthquake ruptures along faults that separate different crustal blocks. The data are
consistent with a preferred northwestward propagation direction for ruptures on all
three faults. If our results are supported by additional observations, asymmetry of
structural properties determined in field studies can be utilized to infer preferred
propagation direction of large earthquake ruptures along a given fault section. The
property of a preferred rupture direction can explain anomalous behavior of historic
rupture events, and may have profound implications for many aspects of earthquake
physics on large faults.
Introduction
Small earthquakes expanding in two directions on a fault are in general a
mixture of modes II and III shear ruptures. However, moderate and large earthquakes
on strike-slip faults (e.g., events with magnitude larger than about M6.5) become,
once they saturate the seismogenic zone, predominantly mode II ruptures (Figure
1.1). Theoretical works indicate that mode II ruptures on a fault that separates
different media (e.g., Weertman, 1980; Adams, 1995; Andrews and Ben-Zion, 1997;
Cochard and Rice, 2000; Ben-Zion and Huang, 2002; Shi and Ben-Zion, 2006) tend
to evolve with continued propagation to a narrow “wrinkle-like” pulse that
propagates preferentially in the direction of slip on the compliant side of the fault
(Figure 1.2).
10
Figure 1.1: Schematic illustration of rupture modes of small to moderate (M < 6.5) and large
(M>7) earthquakes on a strike-slip fault. Earthquakes smaller than about M6.5 propagate as mixed
modes II and III rupture (bottom), while ruptures of M > 7 earthquakes saturate the seismogenic zone
and then propagate predominantly as mode II rupture (top).
Figure 1.2: properties of the wrinkle-like pulse on a material interface. (a) Particle velocities at a
given time generated by a wrinkle-like rupture pulse (small red bar within the white box) propagating
to the right along a right-lateral strike-slip fault between different elastic solids (thin red horizontal
line). The slower velocity block is above the fault as manifested by the fronts of the radiated seismic
waves. (b). Enlarged view of the white box in (a) showing asymmetric particle velocities around the
rupture pulse. After Ben-Zion (2001).
11
Characteristic features of the wrinkle-like pulse include: (1) strong dynamic
reduction of normal stress at the propagating tip, (2) asymmetric particle motion on
the different sides of the fault, (3) self-sharpening with propagation distance, and (4)
preferred direction of rupture propagation and associated directivity effects.
Properties (1)-(3) can produce tensile components of particle motion, leading in
some conditions to "opening modes" of rupture. Property (4) can lead to strong
directivity effects and associated asymmetric ground motion. Anooshehpoor and
Brune (1999) observed features (1)-(4) of the wrinkle-like rupture in laboratory
experiments of sliding on an interface between two different foam rubber blocks. If
these properties characterize ruptures on natural faults, they would have fundamental
consequences for many aspects of earthquake dynamics, including effective
constitutive laws, suppression of branching, the heat flow paradox, short rise-time of
earthquake slip, and expected seismic shaking hazard (Ben-Zion and Andrews, 1998;
Ben-Zion, 2001). It is thus important to test the above theoretical predictions with
detailed in-situ observations.
Large fault zones that accommodated significant slip have interfaces that
separate different media (Ben-Zion and Sammis, 2003, and references therein) due to
the production of damaged fault zone material and the juxtaposition of different rock
bodies across the fault. Such material interfaces are seen directly in exhumed fault
zones (e.g. Chester and Logan, 1986, Evans and Chester, 1995; Sibson, 2003) and
were imaged in the structure of large strike-slip faults using direct- and coda-wave
tomography (e.g., Eberhart-Phillips and Michael, 1993; Scott et al., 1994; Shapiro et
12
al., 2005), reflection/refraction seismology (e.g., Fuis et al., 2001; Lutter et al.,
2004), seismic fault zone head waves (Ben-Zion and Malin, 1991; Ben-Zion et al.,
1992; McGuire and Ben-Zion, 2005) and geodetic data (Le Pichon et al., 2005).
Testing whether mode II ruptures on faults that separate different solids have a
preferred propagation direction, which correlates with the velocity structure as
predicted by theory, should ideally be done with data associated with a large
population (for high statistical significance) of earthquakes that are large enough to
saturate the seismogenic zone (of strike-slip faults) and propagate predominantly as
mode II ruptures. Tests associated with smaller events should especially employ
large data sets, as the bimaterial effect is not fully developed for small to moderate
mixed-mode events. High resolution locations of several thousand small events on
the San Andreas Fault (SAF) show directional asymmetry that is compatible with a
preferred propagation direction associated with the local velocity structure (Rubin
and Gillard, 2000; Rubin 2002). However, the employed events are small and it is
important to obtain additional evidence associated with large earthquakes behavior.
McGuire et al. (2002) and Henry and Das (2001) analyzed rupture properties of over
100 large global earthquakes and found that most are predominantly unidirectional.
However, the examined earthquakes are associated with different faults and therefore
do not necessarily indicate a preferred propagation direction of large events on a
given fault section. Testing the latter can, at present, be done only with geological
studies that search for predicted signals that might reflect the cumulative effect of
many large ruptures.
13
Poliakov et al. (2002) examined the stress field near the tip of a semi-infinite
crack propagating in one direction with constant velocity, and found that off-fault
damage is predicted to occur primarily in the tensional quadrant of the assumed
propagation direction, in a zone that depends on the principal pre-stress directions,
rupture velocity and frictional parameters. Rice et al. (2005) extended the results for
the case of a steady unidirectional slip pulse. Yamashita (2000), Dalguer et al. (2003)
and Andrews (2005) performed numerical calculations of dynamic ruptures on a
fault in a homogenous solid with spontaneous generation of damage in the material
off the fault. In such cases, ruptures propagate as symmetric bilateral cracks and
damage in a single event is produced primarily in anti-symmetric triangular zones,
with width proportional to rupture length, in the two tensional quadrants (Figure
1.3a). If earthquakes on a given fault section propagate predominately as such
bilateral ruptures, or as unilateral ruptures without a preferred propagation direction,
the cumulative pattern of rock damage generated by many events will be
approximately symmetric across the fault. On the other hand, if earthquakes on a
given fault section have a preferred propagation direction, due to any mechanism, the
cumulative damage pattern will be asymmetric, with more damage in the tensional
quadrant associated with the preferred propagation direction.
14
Figure 1.3: Simulations of rupture along biomaterial interface by Ben-Zion and Shi (2005). (a)
Plastic strain (color scale) generated by a bilateral crack-like rupture on a fault at y = 0 in a
homogeneous solid (b) Plastic strain (color scale) generated by a wrinkle-like rupture pulse
propagating to the right on a fault separating a compliant material at y > 0 from stiffer material at y <
0. The stiffer side of the fault is in the tensional quadrant of the radiation pattern for the preferred
propagation direction of the wrinkle-like pulse. The dashed box marks the nucleation zone.
15
Ben-Zion and Shi (2005) performed simulations of ruptures along a material
interface with spontaneous generation of damage in the bulk and showed that in such
cases damage is generated primarily on the stiffer side of the fault, which for the
preferred propagation direction is persistently in the tensional quadrant of the
radiated seismic field. The damage is generated in a strip of approximately constant
width that is related to the pulse width (Figure 1.3b). The cumulative damage
generated by many such ruptures is expected to be similar to the asymmetric pattern
of Figure 1.3b, with significantly more damage on the side of the fault with higher
seismic velocity. The above results provide clear predictions that can be tested in the
field by performing geologic observations of symmetry properties of rock damage
across large strike-slip faults.
In the following sections, we present systematic in-situ observations of
symmetry properties of rock damage in the structure of large strike-slip faults of the
San Andreas system in southern California. The observations indicate overall a
consistent damage asymmetry across the Principal Slip Surface of a given fault
section at several scales. The observed sense of damage asymmetry at these scales
correlates with available information on the seismic velocity structure as predicted
by the theory of rupture along a material interface. The results open up the possibility
of inferring preferred directions of rupture propagation of large earthquakes on large
fault sections from geologic field observations.
16
Approach and methodology
We study in-situ structural properties of fault zones that may reflect the long
term signature of a large population of earthquake ruptures. The examined properties
include the intensity and style of wall-rock and gouge damage, the orientation and
density of fractures, the presence and orientation of subsidiary faults, and
geomorphologic features. Because these properties can be affected by local fault
geometry and complexity, the recognition of persistent asymmetry that is
independent of local site effects requires documentation of multiple signals at
multiple scales at several locations along a relatively straight fault segment.
We utilized both natural and excavated exposures for our studies. In general,
active faults generate detritus that tends to obscure or bury the fault and immediate
damage zone, and we found this to be the case along the San Andreas and San
Jacinto faults. In contrast, the Punchbowl fault, an older inactive strand of the San
Andreas system, is incised and easy to study in natural exposures.
Scales of observations: We discuss primarily two types of observations
associated with different scales. The first type includes observations on the near-field
damage of millimeters to meter fault core scale within gouge zones and their
immediately adjacent wall-rocks. We refer to gouge as crushed and ground up rock
produced by shear. Increasing displacement tends to reduce the grain size (Engelder,
1974; Biegel et al., 1989) and may enhance the development of shear fabric, possibly
due to the competency contrast between the wall-rocks and the gouge material
(Goodwin and Tikoff, 2002). Well-developed gouge zones, like most of those
17
described in this paper, do not display primary structures, probably due to the
accommodation of large shear strain. The composition of a gouge zone is derived
from the adjacent wall-rocks but it often also contains translocated (pedogenic) clay
and other material that migrate from the surface environment into the fault,
especially if it is close to the ground surface. Such alien material often has blackish
organic appearance. The gouge zones of the active faults studied here are typically
dominated by claylike un-lithified material, which behaves plastically when
moistened and contains very small amounts of visible porphyroclasts (unless
otherwise stated). Hence, some of the described gouge elements match the definition
of Scholz (2002) for foliated gouge, modified after Sibson (1977), although we also
commonly observe cohesive and stiff gouge elements that do not fall into any of
Scholz’s categories. The examined gouge layers of the active faults are the zones that
accommodate most of the displacement along the fault, and are therefore the surface
analogue of the fault core, with concentrated slip, in the structure of several exhumed
faults (e.g. Chester et al., 1993). We use the term ‘proto-gouge’ to describe a fault
zone layer within or close to the fault gouge or core in which the fabric and the
composition are in an intermediate stage between a typical gouge and the local wall-
rock.
The second type of observations is made on the distribution of secondary
faulting and associated features in the meters to tens of meters fault zone scale. A
fault zone includes mesoscopic fractures and small secondary faults, typically
extending out to a distance of about 50 to 70 m from the fault core (e.g., Schulz and
18
Evans, 2000). We discuss also a third type of observations, made on the distribution
of damage to the bounding rocks in the tens to hundreds of meters damage zone
scale. This type of observation includes zones of pulverized rocks and seismic fault
zone trapping structures, with considerably more damage than the background
damage of the country rocks.
Frame of reference for symmetry properties: Our frame of reference for
descriptions and measurements of symmetry properties is the currently active or
most recently active Principal Slip Surface (PSS). In the fault core scale, it is the
major through-going slip surface within the fault core that appears to accommodate
most of the displacement during the current phase of faulting, or the life time of the
evolving gouge. In the fault zone scale, the PSS is the most dominant and most
recently active strand among the secondary faults that compose the fault zone
(indicated mostly by the geomorphology). The PSS usually, but not necessarily,
coincides with the major lithological boundary. In the damage zone scale, symmetry
is examined with respect to the zone that accommodates the majority of the long
term displacement. We use the active (or most recently active) slip zone because the
composition and structure of faults generally evolve with time, and we are interested
in structural features that were produced in the current or most recent phase of
faulting, which may be correlated with available information on the current velocity
structure. We can not always distinguish between different generations of damage
and this is another reason to perform observations on several scales and establish a
consistency along a given fault section.
19
The symmetry of structural properties is evaluated by comparing intensity and
style of deformation between the two sides of the PSS. In most of the gouge
exposures, there is usually one continuous surface that becomes clearly visible when
pieces and fragments of gouge are peeled away. When more than a single significant
slip surface is present within the fault core, we describe symmetry properties with
respect to the PSS that is the most continuous and dominant. We assume that every
faulting event includes motion on the currently-active coherent slip surfaces, perhaps
with ongoing destruction or overprinting of older, previously-active slip surfaces.
Evidence for such behavior can be frequently seen in trench logs of paleoseismic
studies, where the relative timing of events is known and one can see the destruction
or overprinting of an older slip surface by a younger trace (e.g. Rockwell et al.
1990).
Overview of investigation sites: For our first set of sites along the San Jacinto
fault, we worked along a stretch of fault with deeply incised geomorphology
resulting from a recent stream capture. In that place the fault is poorly exposed but
could be accessed with minimal work. We mapped the damage in a road-cut and in
backhoe and hand-excavated pits, focusing on the mm to m fault core scale damage.
For the San Andreas Fault zone, we studied several exposures at the core scale along
recently active traces, and also opened trenches in bedrock at a number of localities
to explore the tens of meters wide fault zone scale damage. In a parallel study, we
examined the broader damage zone scale that may extend hundreds of meters out
from the principal slip surface and includes such characteristics as pulverization of
20
rocks (Dor et al., 2006b; Brune, 2001; Wilson et al. 2005) and chemical
alterations. In the Punchbowl Fault within and in the vicinity of the Devil’s
Punchbowl County Park, we studied symmetry properties of the fault core with
respect to the ultracataclasite layer (Chester and Chester, 1998) as well as fault zone
scale distribution of damage. Observations of symmetry properties in the microscale
are also a natural component of a study of this kind, but are beyond the scope of this
paper and are left for a future work.
Methods: The exposures were first cleaned and flattened, then moistened and
finally scraped gently so as not to disturb the natural fabric. We found the natural
gouge fabric to be better defined after several hours, when the exposures had
sufficiently dried. In some cases, additional cleaning with a soft brush and
compressed air were used to better express the gouge fabric. Using optimal natural
light conditions, high resolution digital photos were taken. For most of the smaller
scale study sites, we mapped the fractures on the images and digitized them for
quantification of the fracture density in the different gouge and wall-rock layers on
both sides of the PSS. For some of the other sites we described the overall pattern of
structural features without quantifying them. Fault and damage zone rocks in the
exhumed Punchbowl fault are well exposed and composed of hard clastic and
metamorphic rocks that require just a light brushing and dusting before
documentation. Study of fabric in soft fault rocks was done only in exposures that
are at a minimum distance of 1-1.5 m below the original ground surface (mainly
21
excavated by us), to avoid bias of the results by fabric that results from thermal
cycling, wet-dry cycles, or other surface phenomena.
The intensity of deformation was evaluated in several ways. First, we measure
Fracture Density (FD) across the fault-core or the fault-zone on digitized fracture
maps in rectangular ‘digital samples’ employing a “box-count” method, which was
similarly used by Dershowitz and Herda (1992) and others, involving counting the
cumulative length of fractures and dividing by the sample area. Alternatively, we
apply those measurements to an entire digitized gouge or wall-rock layer. We also
describe the mode of fractures, i.e., joints (mode I) or shear fractures, when such can
be typified, along with their length and orientation. A second method to evaluate
damage was to count fractures along scan-lines perpendicular to the fault strike,
similar to the linear fracture density method used by Wilson et al. (2002) and Priest
(1993). For some exposures, we compare relative strike-slip activity across the PSS
either qualitatively or by counting secondary strike-slip faults that were identified
using kinematic slip indicators. Finally, we describe symmetry properties of
geomorphic features that characterize the various sites.
Distinguishing fabrics: For several cases, we discuss relative activity time for
the different gouge elements within a gouge zone. The discussion is not based on an
absolute dating method, but rather on qualitative characteristics of the shear fabric.
The most recently active principal shear zone, as defined by the surface
geomorphology, tends to break out and be expressed with a scaly fabric (Agar et al.,
1989), with highly sheared and polished anastomosing facoidal lenses of gouge, and
22
abundant evidence of shear in the form of striae and mullions on curviplanar
surfaces. In contrast, what we interpret as older shears tend to be much more
cohesive and either do not have as easily recognized parting surfaces or the surfaces
are compressed and welded. We have found that gouge with high moisture content
appears to behave plastically, whereas lower moisture content gouges tend to have
more brittle characteristics (undoubtedly fluctuates with the seasons). Nevertheless,
the PSS along the active fault core expresses highly sheared parting surfaces even
when the moisture content is high, while older shears tend to develop good parting
surfaces only when relatively dry. Overall, the more cohesive fractures were found
away from the active zone of the gouge when the active zone could be identified
independently. In contrast, the less cohesive shear fractures were found within the
more moistened and more plastic gouge elements. In the presence of several gouge
layers embedded within the fault zone, the one that can be correlated with the
geomorphically active trace of the fault is typically the more moistened/plastic,
which presents relatively incohesive fractures. We use these criteria when the active
or the most recently active zone of the gouge can not be identified independently.
Field Observations
We discuss our field observations by fault zones, starting with the exposures
along the San Jacinto Fault near Anza. We then describe the several sites we studied
along the San Andreas Fault and the Punchbowl Fault, an inactive ancestral strand of
the San Andreas fault system (Figure 1.4).
23
Figure 1.4: Location map. Faults of the southern San Andreas system included in this study and
location of investigation sites (white circles). The town of Anza and city of Palmdale are indicated by
black circles.
The San Jacinto Fault
The San Jacinto Fault (SJF) is a primary strand of the San Andreas Fault south
of the Transverse Ranges (Figure 1.4) and is currently the most active fault zone in
southern California (Crowell, 1962; Allen et al., 1965; Sanders and Kanamori,
1984). The 20 km long Anza section of the SJF is known as the Anza Seismicity gap
(Brune, 1968; Sanders and Kanamori, 1984) and includes the study area. This section
of the fault is well expressed by geomorphic features and the fault is relatively
simple with only one major active strand, the Clark fault (Rockwell et al., 1990), that
separates surficial tonalite and other leucocratic plutonic rocks on the northeast
against gabbroic and more melanocratic rocks to the southwest (Sharp, 1967). These
lithologies probably do not extend very deep below the surface, as the local velocity
structure indicates faster basement velocities at seismogenic depths on the
northeastern side of the fault (Scott et al., 1994). The total right-lateral displacement
24
across all strands of the SJF southeast of Hemet is estimated to be 29 km (Sharp,
1967). The surface slip rate was estimated by trench studies in the Anza area, a few
km southeast of our study area, to be on the order of 12-18 mm/y (Rockwell et al.
1990; Merifield et al., 1991). More recently, in a new set of trenches across the SJF
in Hog Lake, Rockwell et al. (2005) documented the occurrence of 15 surface
ruptures during the past 3500 years, with the most recent event at approximately
A.D. 1760.
The study area is on the edge of the Ramona Indian Reservation, northeast of the
Anza Valley. The surface expression of the fault within the study area indicates that
the fault is highly linear and localized, with one narrow principal strand; in places it
has a well-developed meter-wide gouge zone that juxtaposes surficial middle
Pleistocene alluvial deposits on the southwest against late Pleistocene alluvial
deposits on the northeast. The following sections describe the structure and
symmetry properties of three fault core scale exposures that we excavated and
utilized along a 140 m section of the fault (Figure 1.5). The exposures are between
1.5 to 5 m from the original ground surface, but during the last earthquake
(approximately 245 years ago) their depth was about twice the current one based on
estimates of local erosion rates.
The Road Cut and Trench Exposure - We utilized a steep road cut, facing
towards the southeast, where we exposed the meter-wide gouge zone and adjacent
damaged wall-rocks with a backhoe. The exposed fault can be divided into five
structural domains (Figure 1.5b). To the west, the southwest wall-rock is a well-
25
cemented, sand-rich middle-Pleistocene conglomerate with sparse cm-sized pebbles.
The conglomerate appears to be macroscopically intact even in the immediate
vicinity of its sharp contact with the gouge zone. The western-most gouge element is
a 15 to 20 cm wide, highly cohesive, stiff and massive, gray-black and clayey, with a
relatively large amount (compare to other gouge elements in this exposure) of sandy
porphyroclasts. There are very few macroscopic fractures in this layer and the lack of
foliation or any other evidence for shear fabric suggests that the layer is inactive. The
contact with the south west wall-rock is well defined. The “massive” gouge is bound
on the northeast by the principal slip zone and a PSS.
26
Figure 1.5: Structural units and fracture density in the gouge of the San Jacinto
fault zone south of the Ramona Indian Reservation. The colors of the different frames
correspond to the colors of the circles in the small location map (panel e).
Red frame: (a) Panoramic view showing the upper and lower road cut exposures with the inferred
location of the PSS (green) and the analyzed frames (yellow boxes) presented to the right (panels b
and c). In the center of the panel the fault is covered with debris. (b) A cross fault view of the gouge
and damage zone, with a division to 5 structural domains. The red graph shows the trend of fracturing
intensity across the fault. Each yellow dot indicates the cumulative fracture length (CFL) in cm
measured in the rectangle frame right below it in the middle panel. (c) A digitized fracture map of the
27
upper panel. Light blue spots mark zones with insufficient visibility for detailed mapping. Dashed line
marks the PSS. The rectangles are samples of CFL. The numbers on the lower horizontal bar give
fracture density (FD) calculated for each of the structural domains from the upper panel by dividing
the CFL from all the samples in the domain with the cumulative area of the used samples. A sharp
contrast exists between the FD of the sheared and massive gouge. Note also that the FD of the NE
wall-rock is higher than the FD of the massive gouge. (d) The gouge 3 m below the upper road cut
narrows down to 15 cm but maintains its structural division. Massive, gray, 1 cm wide gouge appears
on the SW and highly foliated (sheared) gouge appears on the NE.
Blue frame: (f) The “Cliff” exposure, 100 m south east of the road cut. The picture on the left shows
the artificial exposure with difference between the dark cohesive gouge on the left to the fractured
gouge on the right. The framed zone is 70 cm wide. The structural division in the digitized map on the
right is consistent with the structural division shown for the road cut (red frame), with similar trend of
the FD indicated by the numbers at the colored bar at the bottom of the panel.
Green frame: (g) Digital fracture map in artificial exposure (inset) 140 m SE of the road cut. The
gouge and adjacent wall-rocks were divided into 5 structural domains that can be correlated to the
structural division of the road cut exposure (red frame). The PSS is not clear here but based on the
other two exposures with similar structure, the PSS should be between the massive and the sheared
gouge. The FD index is shown for each domain on the lower colored bar. Higher damage is on the NE
side of the fault.
There is a sharp transition in the nature of the gouge that takes place across the
PSS. The 40 cm wide gouge layer northeast of the PSS has a dark-brownish color, is
highly sheared, and includes numerous small slip surfaces (shear fractures). These
curviplanar, facoidal slip surfaces are polished, shiny and contain low-rake slip
striations. Most of the surfaces are oriented parallel to the fault. The cohesion on the
slip surfaces is low as they are easy to separate. The slabs of the material between the
fractures are claylike, flexible and lack significant amount of visible porphyroclasts.
Farther northeast, the next gouge layer is about 40 to 50 cm wide and provides a
gradual transition from the sheared gouge on the southwest to the northeastern wall-
rock. This gouge layer is also rich in clay over porphyroclasts but slightly more stiff
and cohesive with respect to the adjacent sheared gouge layer. It has a mixed-
population of fracture modes, dominated by joints (mode I fractures) with just minor
evidence of shearing; it therefore appears to be mostly shattered. The fracture
28
orientation is more diverse and exhibits an overall higher angle to the PSS compare
to the fractures in the sheared gouge layer. Finally, the northeastern wall-rock is a
late-Pleistocene conglomerate that is similar in texture to the SW wall-rock but is
more damaged with distributed macro-scale fractures. It is important to note that
these cemented sediments are much younger than the less-damaged middle
Pleistocene sediments on the southwest side of the fault.
We created a highly detailed digital image framing a 180 cm x 50 cm rectangle
across the fault. Figure 1.5c shows 900 fractures that were mapped on the image. The
gray spots mark zones where visibility on the image was not good enough for
accurate mapping. The fracture pattern in the digitized image delineates the
structural divisions. In order to quantify the fracture density gradient, we defined 29
rectangular 4 cm x 12 cm frames in a continuous array across the image. The
distribution of the rectangles on the image was dictated by the zones of low
visibility. The Cumulative Fracture Length (CFL, in cm) within each of these frames
was calculated, and the results are shown by the red curve against distance across the
fault in Figure 1.5b.
The CFL in the southwest wall-rock is practically zero at this mapping
resolution. In the massive gouge, the CFL rises to several cm/frame, whereas in the
sheared gouge it jumps across the PSS to between 20 and 50 cm/frame. The CFL
stays at a level of 30-40 cm/frame in the shattered gouge and drops to between 0 to
20 cm/frame in the NE wall-rock.
29
To correspond with the structure of the fault zone, the CFL curve was divided
along the length axis into domains corresponding to the above structural divisions.
Each structural domain was assigned a FD index in units of cm/cm
2
based on the
sum of the cumulative fracture length from all the frames within a domain, divided
by the cumulative area of all the frames within the domain. The FD index values for
the structural domains are as follows: SW wall-rock: 0; massive gouge: 0.08; sheared
gouge: 0.82; shattered gouge: 0.58; and NE wall-rock: 0.17. The results indicate that
the structural division can be reasonably also considered as fracture density division.
It is especially interesting that the FD in the northeast wall-rock is higher than the FD
in the massive gouge.
We extended the road cut exposure downward by digging a trench on the side of
the road three meters below the location of the exposure described above (Figure
1.5a). We estimate the bottom of the trench to be about five meters below the
original ground surface prior to the excavation of the road. The gouge width shrinks
from one meter in the road cut to only 15 cm at the base of the trench. The structural
division of the gouge in the road cut can in general be applied to the gouge in the
trench, but it is much more compressed and the PSS cannot be identified
macroscopically (Figure 1.5d). The visible cracks in Figure 1.5d were developed
upon dehydration of the exposure face and it is not clear to what extent their pattern
is controlled by the gouge fabric. The relatively cohesive and massive gray gouge
layer on the southwest side of the zone is only 1 cm wide and shows no original
visible fractures. The next gouge layer is more brownish and exhibits dense foliation
30
with fracture spacing in the sub-millimeter scale, and we correlate this layer upward
to the sheared gouge in the road cut. The FD of these gouge layers can be studied
and quantified only in the microscale, which is beyond the scope of this paper.
The Cliff Exposure - The "Cliff" exposure is 100 m southeast of the road cut
exposure on a natural steep slope facing to the southeast. We manually excavated the
fault to a depth of 1.5 m below and away from the slope; this area was not accessible
to a backhoe. The exposure was cleaned following the procedure described above,
and a sequence of digital photos was taken across it upon which the fractures were
mapped and digitized. The 70 cm wide gouge zone at this exposure can be divided
into three distinct structural domains, and each domain can be correlated to the
gouge domains in the road cut. The structural domains also coincide with the
divisions in fracture density domain (Figure 1.5f). This division is sharp and
continuous enough, and the quality of the digital image is sufficiently good, to enable
us to directly calculate the FD of the structural domains without using a continuous
array of digital samples.
The gouge layer on the southwest side of the PSS is massive and highly
cohesive, gray dark and rich in clasts. It is especially stiff and brittle. The layer
contains fractures that are relatively large, showing inconsistent orientations, but no
macroscopic foliation and shear fabric. We measured a FD of 0.71 for this unit.
Many of the fractures are incohesive and they seem to have fresh faces of recently
broken material; therefore, they may have been created when the outcrop was
excavated, possibly due to the material stiffness. We therefore conjecture that the
31
actual in situ FD is less than the measured FD. The gouge layer on the northeast side
of the PSS is highly sheared and characterized by a higher value of FD of 1.31. It is
much less cohesive, sparse with porphyroclasts and contains many anastomosing
shear fractures with curvy faces and stria; the fractures are relatively small, giving to
the gouge a flaky texture. The majority of the fracture population is parallel to the
fault. Farther to the northeast, the next gouge layer is a transition from the sheared
gouge to the northeast wall-rock, with a FD of 0.72. The fractures orientation in this
gouge layer are less uniform with respect to orientation of fractures in the sheared
gouge immediately to the southwest side but more internally consistent with respect
to those of the massive gouge from the southwest side of the PSS. Their surfaces are
also more cohesive and do not have a fresh appearance. Thus, unlike fractures in the
massive gouge, the fractures in this gouge layer are believed to be part of the original
rock fabric prior to excavation, and therefore more related to faulting. We suspect
that the FD contrast between this gouge layer and the massive gouge is higher than
we measured.
The SE exposure - The southeast exposure is an additional 40 m to the southeast
from the ”Cliff” exposure. The outcrop faces to the northwest, opposite the direction
of the two previously described exposures. We manually excavated an exposure to a
depth of about 1.5 m away from the current slope surface (this site was also not
accessible by backhoe) and performed the same analysis as in the other exposures.
Here the fault zone is similar in its structural division, in its FD distribution and in
the sense of asymmetry to those presented for the road cut, trench and Cliff
32
exposures (Figure 1.5g). The FD indexes on the southwest side of the PSS are 0.3 for
the southwest wall-rock and 1.14 for the massive gouge. Northeast of the PSS, the
FD indices are 2.26 for the sheared gouge, 1.33 for the shattered gouge, and 0.46 for
the northeast wall-rock.
Summary - All the three exposures along a 140 m long section of the San Jacinto
Fault near Anza exhibit consistent and distinct structural division and sense of
asymmetry, with the northeast side of the fault always more damaged, regardless of
the slope aspect or depth of the exposure. In each of the three studied exposures, the
gouge on the northeast side of the PSS is highly sheared, followed by a more
shattered gouge layer farther to the northeast. The gouge layer on the southwest side
of the PSS is always cohesive and massive and appears to be inactive. Fracture
density analysis highlights and confirms the apparent structural division and the
sense of asymmetry.
The Mojave Section of the San Andreas Fault
The San Andreas Fault system, in a cross section that includes our study area in
the Mojave, contains in its broader sense the San Gabriel Fault on the south side of
the San Gabriel Mountains and several parallel strands on the north side of the San
Gabriel Mountains. The latter include the currently active SAF and the Little Rock,
Punchbowl, southern and northern Nadeau faults, as well as several other smaller
faults. These faults all exhibit a zone with faulting-related structural features that
measures tens of meters in width (Chester and Logan, 1986; Chester et al. 2004), and
33
they can potentially have damage zones at a scale of ~100 m that correspond to the
observed low velocity trapping structures for several faults (e.g. Ben-Zion et al.,
2003; Peng et al. 2003; Lewis et al., 2005), and to the ~100 m wide zone of
pulverized fault zone rocks observed for the SAF (Wilson et al., 2005; Dor et al.
2006b). A broader damage zone that reflects the entire deformation history of a large
fault can consist of a few km wide region with enhanced anisotropy and reduced
density and elastic moduli (Ben-Zion and Sammis, 2003; Fialko, 2004; Peng and
Ben-Zion, 2004).
The San Andreas fault system in southern California has accommodated
altogether about 240 km of displacement, based on the separation of the correlative
Neenach and Pinnacles Volcanic rocks (Matthews, 1976), but this may not include
50-60 km of slip on the San Gabriel Fault. The displacement may be even larger, in
excess of 300 km, based on the separation of Pelona schist and other basement units
(Dibblee, 1989). The displacement was accommodated on the ancestral faults of the
system, and subsequently on the modern elements of the SAF as they developed
(Powel and Weldon, 1992). Each fault remains a narrow zone of discontinuity along
which local dip-slip adjustments have occurred, sometimes into Holocene times
(Barrows et al. 1985).
The individual displacement values on each of the primary strands of the San
Andreas system in the central and southeastern Mojave are only partially constrained
in most cases. The San Gabriel fault accounts for about 50 to 60 km of the total slip,
whereas the Punchbowl and Nadeau group of faults accounts for 44 and 16 km,
34
respectively. The Little Rock fault has only a minimum value of 21 km of
displacement, and the currently active trace of the SAF appears to have only 21 km
of slip based on the displacement of the late Pliocene Juniper Hills Formation. The
discrepancy between the sum of those displacements (~160 km) to the assumed total
~300 km could be accounted for by additional displacement on the Little Rock Fault,
displacement on the Hitchbrook fault northwest of our study area, and larger earlier
displacement on the current SAF (Barrows et al. 1985) or on the Clemens Well,
Fenner and San Francisquito faults (Powel and Weldon, 1992).
The broad SAF system in the study area is therefore a 7 km wide zone of
extended deformation, and with the San Gabriel Fault the zone is >30 km wide. This
broad zone experienced several generations of deformation after the middle
Miocene, with complex partitioning of displacement between its components; each
deformation phase could have overprinted its signature on rocks within the system.
Individual rock bodies are therefore likely to have recorded a complex displacement
and deformation history. The existence of numerous geometrical perturbations at
many scales, from the big bend to smaller step-overs, and interaction between the
main trace to subsidiary faults all influence the faulting pattern. Even rocks along a
relatively straight portion of the fault may carry in their ‘geological memory’ an
imprint from a zone of structural complexity farther along the fault. In such a
complex system with very large displacement, the velocity structure across the fault
may have changed several times. A single rock body juxtaposed along the fault could
have experienced a contrasting velocity structures over time, and maybe even have
35
experienced an opposite preferred rupture propagation direction, leaving
contradicting long-term signals in its fabric.
The task of finding a coherent geological signal along such a fault that records
the long term behavior is rather challenging. It is particularly complex in the active
environment of the SAF zone with sparse bedrock exposures, even in the arid
portions of the Mojave Desert. As we anticipated the signal to be noisy and possibly
confusing, we searched for evidence of several types and at several scales.
We present here observations from three sites in the Little Rock to Palmdale
area, spanning a 20 km long section of the fault. At two of the sites near Little Rock,
observations are on the fault core scale, and at one site near Palmdale the
observations are at the fault zone scale. We also refer to observations in the damage
zone scale made along the entire 140 km fault section of the Mojave in a parallel
study (Dor et al. 2006b). The sites are not on the geometrically simplest section of
the fault, nor are all of the sites on the currently active strand of the SAF, but the
region has a semi-arid climate that provides opportunities for examination of the
fault in bedrock exposures with minimal excavation.
The Little Rock Paleoseismic Site - The Little Rock site, in the vicinity of E. 96
th
St. south of the town of Little Rock (Figure 1.4), hosted an investigation of the
Holocene slip rate of the SAF (Schwartz and Weldon, 1986; Weldon and Fumal,
2005). They excavated a system of trenches and exposed several parallel strands of
the fault, including the one that accommodated the 1857 rupture. According to their
unpublished mapping, only one strand that separates bedrock units was found at the
36
bottom of the trench system. This strand of the SAF is about 5 to 8 meters north of
the 1857 strand and below an active channel (Figure 1.6a). We re-trenched and
exposed the upper 60 cm of this bed rock fault zone, which is cut across by middle
Holocene (3.5 ka) unconsolidated alluvial deposits (Tom Fumal, pers. comm., 2004).
This fault strand juxtaposes granite on the northeast against the conglomeratic
late Pliocene Juniper Hills Formation. These two units are separated by a 5 cm wide
soft and claylike dark brown gouge zone with distributed shear fabric and a through-
going PSS inclined about 60º to the southwest (Figure 1.6b). This strand probably
represents the post-Pliocene long-term SAF in this area. Individual earthquakes may
still occupy shallow branches close to the ground surface, as indicated by nearby
faults within young sediments.
37
Figure 1.6: Fault core of the SAF at the Paleoseismicity site south of the town of Little Rock. (a)
The active strand of the SAF is marked by the line with the slip sense marks; the amount of channel
deflection is indicated (after Weldon and Fumal, 2005). The red heavy line represents the re-trenched
strip and the yellow box marks the location of the trench exposure shown in (b). The line with
triangles is a thrust fault, footwall to the north. (b). The trench exposure in the yellow box of (a). The
PSS is within a 5 cm wide gouge zone separating the Juniper Hill Fm. from pulverized granite. The
intense damage zone (fault core) related to displacement of this fault outside the gouge is exclusively
within the granite. The damage is asymmetrically located on the NE side of the fault.
The Juniper Hills conglomerate of the southwestern block is essentially intact. A
10 cm long pebble located immediately south of the gouge was reoriented parallel to
the PSS and shows no macroscopic fractures. Damage in the northeastern block is
substantially more intense. The granite is pulverized, with the original grain-scale
fabric preserved. The rock has a powder-like texture similar to the one that
38
characterizes the outcrops of Tejon Lookout granite in Tejon Pass (Wilson et al.
2005) and in other localities along the Mojave section of the SAF (Dor et al. 2006b).
In addition, cm to meter long fractures cut sporadically through the rock mass. This
texture and fabric disappear within 30 to 40 m from the contact of the northeastern
block with the gouge. The rock belt immediately northeast of the gouge is gray, have
powdery-plastic texture, and seems to include a mixture of siliceous material from
the granite on the northeast and claylike material from the gouge on the southwest.
This belt exhibits a diffuse contact with the granite and a sharp contact with the
gouge zone. We interpret this as a proto-gouge layer, representing an intermediate
development stage between the granitic protolith and the clay-rich gouge.
The history of deformation in the northeastern block of this fault is unknown
and the granite could have been pulverized and fractured during a previous episode
of faulting. Some of this deformation could therefore have occurred against other
rock bodies and along other fault strands. Nevertheless, the fault core, which is the
zone that accommodated the recent displacement and experienced mineralogical and
textural changes associated with slip along this fault strand, includes in addition to
the PSS and gouge, a proto-gouge layer of reworked granite within the northeastern
block; no similar changes or damage to the rock can be identified within the Juniper
Hills conglomerate. The fault core here is the result of a limited history of faulting,
as indicated by a narrow gouge, and was not likely developed within the granite
during a previous faulting phase. The smooth transition from pulverized granite to
the proto-gouge layer immediately adjacent to the PSS indicates that this layer and
39
the PSS are genetically related. Hence, the fault core that is associated with slip
along this fault strand, during which the fault juxtaposed the Juniper Hills
conglomerate and the granite, was developed entirely on the northeast side of the
PSS. The asymmetric structure of the fault core with respect to the PSS indicates that
the creation of fault core scale damage during SAF earthquakes favors the
northeastern side of the fault in this location.
Little Rock Creek Site - The active strand of the SAF southwest of the town of
Little Rock and north of the Little Rock reservoir (Figure 1.7) has strong geomorphic
expression with a right-lateral deflection of 600 m in the channel of Little Rock
Creek. Bedrock outcrops with expression of fault zone features are absent from this
active fault environment due to burial by young sediments. However, 300 m south of
the currently active strand, a continuous exposure of a parallel, presumably inactive
strand of the SAF can be traced for several hundred meters. This fault is especially
prominent on the southeast wall of the Little Rock channel, where the gouge zone is
more than 6 meters wide. It continues in a relatively straight manner toward the
southeast, branching locally into two parallel strands. Along most of its exposed
length, the fault expresses a 1-2 m wide gouge zone, separating slightly
metamorphous granodiorite on the northeast from fine sandstones interbedded with
shale on the southwest. The sandstone-shale sequence is plastically deformed with
fold limbs at the meter scale. The time of activity of this ancestral fault strand is
unknown but there are no obvious geomorphic features to indicate significant
Holocene slip.
40
Figure 1.7: Locations of the Little Rock Creek study site and features of the SAF near by. The
location of the trench is marked by a vertical white box. Note the ~600 m deflection of the Little Rock
Creek channel by the active strand of the SAF. The trench was excavated 300 meters to the south,
across an inactive strand of the fault.
We manually excavated a 1.2 m deep and 4 m long trench (Figure 1.8) covering
the span of the fault core, across the gouge zone and the adjacent wall-rocks where
the fault is expressed as a single straight lineament. The fault core here has a
composite structure with several fault rock layers; each of them has a distinct
lithology, texture and color. We applied the box-count methods for each of the layers
within the fault core, using rectangular zones with homogeneous texture and
minimum irregularities for FD analysis.
41
Figure 1.8: Trench log from the Little Rock Creek. Includes digitized fracture maps of framed
zones from the different gouge layers within the gouge zone. The numbers in the rectangles indicate
the fracture density (cm/cm
2
). There is a clear trend of decreasing FD from NE to SW. The fractures
in the NE pulverized granodiorite are in the microscale.
On the northeast, the granodiorite is pulverized, exhibiting textures similar to
those found in the pulverized Tejon Lookout granite at Tejon Pass (Wilson et al.,
2005). However, the width of the pulverized zone is much narrower and extends to
between several tens of cm to several meters from the gouge. The fractures here are
in the sub-micron scale and therefore meso-scale fracture mapping is not effective
for reliable FD analysis. Nevertheless, the FD must be high at the fine scale.
Adjacent to the northeast wall-rock, there is a 90-cm-wide, fine cataclasite that
is macroscopically alike the ultracataclasite described for the Punchbowl fault
(Chester and Chester, 1998) but here it includes sparse large clasts. It contains dense
42
fracture population, dominated by joints (mode I cracks) in the mm scale, with a FD
of 6.8.
The next three gouge layers to the southwest, with a cumulative width of one
meter, differ in their color (from northeast to southwest: gray, blue and brown) but
share the same general texture and overall clayey composition. The gouge layers are
stiff and cohesive but contain a curvy and apparently bended shear fractures. The
fractures also appear to be well-welded (cohesive) and are therefore probably
recently inactive. They are on average parallel to the orientation of the fault. The FD
values of the three layers from northeast to southwest are 5.5, 3.2 and 2.9,
respectively.
Farther southwest, a 75 cm wide gouge layer is probably the most recently
active shear zone based on its similarity to the active gouge zone in other places.
The fractures are small to medium in the cm scale (1-7 cm long); they are curved,
shiny and have slip striations with low rakes. The fracture surfaces are less cohesive
with respect to fracture surfaces in the other gouge layers and the clayey flakes
between them are flexible. Based on this texture and the above similarity, it is
possible that this gouge layer was the main zone to participate in slip events during
earthquakes on the modern SAF until fairly recently. This gouge layer has a through-
going slip surface, which we interpret as the most recent PSS within the fault zone,
dipping 65º and located slightly to the northeast from the center of the layer.
Nevertheless, its FD of 2.4 is somewhat less than those of the gouge layers to the
northeast. The sheared gouge layer is separated from the southwest host rock by a
43
20-cm-wide layer of slightly more massive gouge. Finally on the southwest, there is
a broken sandstone unit with FD of 1.6. Fractures are relatively large and have no
preferred orientation.
Overall, the fracture density within the gouge decreases systematically from the
northeast to the southwest. Although we are not sure whether the entire structure and
damage pattern was developed during the activity period of the current PSS, it
appears that throughout the history of this fault strand, damage accumulated
preferentially on the northeast side of the fault.
West Palmdale Trench Site - Between highway 14 and Tierra Subida road in
Palmdale, the SAF juxtaposes surficial Quaternary alluvial deposits on the southwest
against similar deposits with slivers of the sandstone member of the late Pliocene
Anaverde Formation on the northeast (Barrows et al., 1985). A series of deflected
channels, benches and linear ridges attest to the localized nature of the fault in this
area, with only minor indications for small-scale step-overs and active secondary
parallel strands.
We excavated a 45 m long trench with a general trend of 200º, perpendicular to
the strike of the fault, in order to expose the structure of the fault zone. The trench
crosses the fault where it deflects an active channel by about 15 m. Fortunately, the
sedimentary and crystalline bedrock that host the fault are either exposed on the
surface or just shallowly buried. Due to extensive damage and pulverization, we
found excavation through those rocks to be remarkably easy, except in places where
44
the backhoe had to penetrate old and tremendously cohesive gouge zones associated
with inactive fault strands.
Two fault structures dominate the trench exposure. One is a massive dip-slip
fault zone with a minimum horizontal width of 21 m. The second structure, which
occupies much of the rest of the trench, manifests strike-slip faulting related to the
active SAF. A detailed description of structural features, their relations within the
trench exposure, and their general geological context is presented below and
corresponds to Figure 1.9.
A primary interpreted aspect of these two structures is that the dip-slip structure
is inactive, as indicated by the fabric of its gouge, by the lack of geomorphic
evidence for its activity and by a cluster of relatively young strike-slip faults that
cross-cut its gouge fabric and juxtapose it against young sediments. In contrast, the
strike-slip structure has several active or recently active components, as indicated by
their correlation with surface geomorphic features and by the nature of their fabric at
several scales (consistent with the discussed indicative fabrics in section 2).
Structural relationships indicate that the strike-slip structure is superimposed on the
dip-slip structure.
45
Figure 1.9: A trench log displaying 45 m long cross section of the San Andreas fault zone near
Palmdale. For displaying purposes, the trench profile is divided (bottom inset) into panels I and II
showing the northeastern and southwestern parts, respectively. Features related to the strike-slip
structure are shown with red, light-green and dark-green colors. The active strike-slip structure is
superimposed on an inactive exhumed dip-slip structure. Elements of the composite PSS appear in the
middle of the trench (shown in both panels). Most of the strike-slip activity is concentrated in panel I
northeast of the inferred PSS. See details in the legend and in the text.
The distribution of bedrock and other lithologic units exposed in the trench is
shown in Figure 1.9. Sandstone of the Anaverde Formation (light brown and orange)
appears in the northeastern half of the trench. Several slivers of granite and gabbro
(dark purple) appear within the sandstone, separated from it by strike-slip faults.
Finally, an assemblage of crystalline rocks is present on the southwest side of the
main fault as slivers within the dip-slip fault zone (pink). All the sedimentary and
crystalline rocks yield powdery substance upon light pressure, suggesting that they
46
were pervasively pulverized (Dor et al., 2006b). Pleistocene (and younger?) alluvial
material crops out at the southern end of the trench (yellow). Shallow soil and recent
alluvium were not included in the trench log except in the active channel area (dark
brown). Structural features include faults (red and black lines), shear fabric and its
orientation (red and black small bars), gouge zones (color coded according to type)
and slip striations with a separate symbol for strike-slip, dip-slip and oblique-slip,
marked where kinematic indicators were observed.
Features related to the dip-slip fault appear mainly in the southwest and central
part of the trench. They comprise a wide dip-slip fault zone parallel to the strike of
the SAF and dipping 35º to the southwest. Dark blue zones in Figure 1.9 represent
massive and extremely cohesive, clayey and heavily welded dark brown to black
gouge, having large planar slip surfaces that are parallel to the overall structure. The
more pronounced, dominant and continuous surfaces appear as black heavy lines that
in most cases separate the massive gouge from other units. Smaller shear surfaces
(short black bars) within the gouge and along the major slip surfaces show a
consistent set of pure dip-slip striations. The striations are clearly printed as long,
straight and parallel mini-troughs and ridges into the gouge surfaces (mullions), and
are often associated with linear gypsum crystal growth. Zones in light blue are gouge
layers of the same nature as above, also having a substantial amount of silty material.
In its main part (south of the central part of Figure 1.9), this gouge is shown with
dotted lines and elongated pink patches, representing and generalizing numerous slip
surfaces and slivers of crystalline rock inside the gouge, respectively.
47
Strike-slip faulting dominates in the northeastern and central part of the trench
and is manifested by several tens of strike-slip faults and fault segments (red lines)
with various inclinations from 30º to vertical, dipping both to the north and to the
south. These faults are typically narrow and localized (0.5 to 4 cm wide), defined by
brown-red to brown-dark, incohesive gouge layers with flaky fabric and small (mm
to cm scale) shear surfaces that display mostly horizontal to oblique slip striations.
The material itself is soft, claylike and appears to lack visible porphyroclasts. The
strike-slip faults are surrounded or bounded by additional two types of gouge zones:
the first type (green) is brown to dark-brown and has clayey soft content with some
silty material. In places, it exhibits shear fabric parallel to the orientation of the
nearest fault (dashed red lines). The second type appears to be a “proto-gouge” (light
green). Those gouge element are normally gray or light brown and seem to have
smaller amounts of clayey material with respect to silt; their visible clasts content is
higher than that of the other gouge elements and may bring them marginally under
the definition of breccia (Sibson, 1977). They lack distinct shear fabric and preserve
some of the grain fabric of the host rock (in most places, the arkosic member of the
Anaverde Fm.). The relationship of the ‘proto-gouge’ units to strike-slip faulting is
indicated by their close association with the other darker and sheared gouge units,
with the strike-slip faults themselves, and by their sub-vertical orientation. These
three gouge types seem to reflect different maturity stages of gouge zones, and they
all appear to be associated with recent faulting.
48
We attribute the PSS to one of the following two candidates (Figure 1.9). The
first is the isolated strike-slip fault immediately below the northern end of the active
channel; this fault was the major carrier of displacement during at least the last
several earthquakes as indicated by its strong geomorphic signature. Associated with
this fault is a 15 m large deflection of the channel and additional geomorphic
features along strike, such as aligned notches and other deflected channels. It
reactivates two or more ancient slip surfaces within the dip-slip structure, while
stepping from one surface to the other, creating an overall steeper inclination than
the host surfaces. The other candidate for the PSS is one of the vertical faults that
bound the sliver of fine-grained sandstone (orange) on the top of the hill
approximately at the center of the trench. Those faults juxtapose distinctive
lithological units (the coarse against the fine grained members of the Anaverde
formation and the entire northern sequence of rocks against the major part of the dip-
slip structure). They also separate the same set of units in another trench, located
~100 m to the east, where they also correlate with the dominant geomorphic
expression for recent faulting. Due to this uncertainty with regard to the definition of
the PSS, we refer to the zone that includes both candidates as the composite PSS
zone.
The distribution of strike-slip features across the fault zone in Figure 1.9 shows
that strike-slip activity is drastically stronger and more intense on the northeast side
of the composite PSS with respect to strike-slip activity on its southwest side. Only
three vertical strike-slip faults clustered in a one-meter-wide belt appear southwest of
49
the composite PSS. They most probably slipped during young SAF activity as they
separate the dip-slip complex from young (late Quaternary?) alluvial material.
However, they do not have a clear geomorphic signature that indicates recent
activity. The ground surface outside of the trench southwest of this belt is covered
with old alluvium and soil, and shows no indication of recent fault activity as well.
From the northeastern side of the composite PSS, several tens of strike-slip faults cut
through the exposed rocks. They have most probably been active during recent SAF
events and may have accommodated a fraction of the displacement. This is suggested
by the fresh appearance of their shear fabric, which is alike to the fabric found in the
faults of the composite PSS and in other active gouge zones (such as the gouge of the
road cut exposure along the SJF near Anza, as described above). The association of
this large group of faults with the active faults of the composite PSS is supported not
only by their textural similarity, but also by an overall decrease of their local density
as a function of distance from the major active area of the fault zone. A dense group
of faults is clustered around slivers of crystalline rocks north of the PSS and their
density decreases for the adjacent five meters to the north, after which the density
further decreases over the next few meters. They become more diffuse with larger
volumes of the host rock between them, and the last seven northern meters or so of
the trench exhibit sparse direct evidence for shear, although there are several proto-
gouge zones that may indicate minor strike-slip activity. The overall trend is
therefore of gradual decrease in strike-slip activity from the composite PSS toward
the northeast.
50
In addition, a ‘fault zone valley’ is correlated with the zone of intense strike-slip
activity and can be partially inferred from the profile of the trench in Figure 1.9.
There is no lithological reason for the development of such a valley on this side of
the fault zone, as rocks on the valley side of the fault are sandstones and crystalline
rocks that are likely more resistant to erosion than the alluvial material on the other
side of the fault. Therefore, we attribute this effect to the increase in damage, and in
particular, the partial pulverization of the rock and the weakness of the active gouge
zones on the northeastern side of the fault zone, thereby enhancing the erosion.
To eliminate the possibility that this pattern reflects a local step-over or another
site-related complexity, we opened another trench 100 m to the southeast. We
observed there similar distribution of structural properties and geomorphic features,
confirming the general asymmetric nature of the fault zone structure in the area.
Summary - We observe asymmetry of structural properties in three different
exposures along the Mojave section of the SAF. Two of the observations (Little
Rock Paleoseismicity site and in Little Rock Creek) are at the scale of cm to meters
(fault core scale), whereas the site west of Palmdale is at the scale of meters to 10s of
meters (fault zone scale). Asymmetry has a unique form in each of the exposures, but
they all share the same sense: the northeastern side of the PSS is systematically more
damaged.
51
The Punchbowl Fault
The punchbowl fault is an inactive, exhumed strand of the San Andreas system
in the central Transverse ranges of Southern California (Figure 1.4). The fault has a
long wavelength sinuous trace, dips steeply to the southwest, and was primarily a
right lateral strike-slip fault with a minor reverse component (Chester and Logan,
1986). The Punchbowl fault is parallel to the SAF and it is truncated to the northwest
and to the southeast by the active SAF. Most of the Punchbowl trace is positioned
about 5 km southwest of the SAF. The Punchbowl fault is therefore considered to be
part of the SAF system (Dibblee, 1968, 1987). It juxtaposes along much of its length
Precambrian to Cretaceous rocks of the San Gabriel basement complex against
arkosic sandstones and conglomerates of the Miocene-Pliocene Punchbowl
Formation. The Punchbowl Fault and the SAF form the boundaries of the Punchbowl
basin, which was most likely created as a pull-apart basin in which the Punchbowl
Formation accumulated to a thickness of more than 1 km (Noble, 1954; Woodburne,
1975). Subsidiary faults within the formation suggest that deposition of this unit was
contemporaneous with the movement on the Punchbowl fault (Chester, unpublished
mapping, 1995). According to Weldon et al. (1993), at least half of the displacement
on the fault occurred during the Pliocene and Pleistocene, during and after the
deposition of the Punchbowl formation. The overall right lateral displacement on the
Punchbowl fault complex is assumed to be of the order of 40 to 50 km, with 44 km
proposed by Schulz and Evans (2000) and Chester et al. (2004) based on the
separation of the San Francisquito Formation and the Fenner faults (Dibblee, 1967,
52
1968), and in agreement with the offset of the Punchbowl basin from its inferred
sediment source (Weldon et al., 1993).
The Punchbowl fault zone has a well defined structure with damage intensity
that increases progressively from the outer fault zone toward the fault core and the
ultracataclasite layer (Chester et al., 1993). Several workers have shown that shear is
restricted to the fault core, the zone that includes the ultracataclasite layer and a
zone, up to a few meters wide, of foliated cataclasite on both sides of the
ultracataclasite layer (Chester et al., 1993; Schulz and Evans, 1998; Chester and
Chester, 1998). An argument in favor of localization is the continuous lithology
across each block adjacent to the fault compared to the sharp contrast in lithology on
both sides of the narrow contact zone (Chester et al., 2004). Additional support
comes from fabric analyses that show sharp mechanical and mineralogical
boundaries between the ultracataclasite and the host rock (Chester et al., 1993).
Detailed ultracataclasite studies have shown that slip was further localized at the last
stages of faulting, macroscopically and microscopically, on a prominent slip surface
within the ultracataclasite layer (Chester and Chester, 1998). For this reason the
ultracataclasite layer is our frame of reference for the purpose of the study of
symmetry properties in the Punchbowl fault. Effective study of symmetry properties
within the ultracataclasite layer can be done only in the microscale.
The present day exhumation is the result of a post-Pliocene uplift and erosion of
the San Gabriel Mountains. According to the inferred uplift and erosion rate
(Oakeshott, 1971; Morton and Matti, 1987), the thickness of sedimentary sequence
53
in the Devil’s Punchbowl basin cut by the Punchbowl fault, and the mineral
assemblage and microstructures of fault rocks (Anderson et al., 1983; Chester and
Logan, 1986; Evans and Chester, 1995), 2 to 4 kilometers of rock sequence has been
eroded since the Pliocene, exposing a depth that is comparable to the top of the
seismogenic zone of the modern San Andreas Fault (Schulz and Evans, 2000). Our
study area includes about 2 km of the fault length within the Devil’s Punchbowl
County Park area, and in the Forest Service lands east of the park down to the South
Fork Campground (Figure 1.10). We first discuss observations from the South Fork
area and then from the vicinity of Devil’s Chair. Both areas have been studied in
various contexts in previous studies (Chester, unpublished mapping, 1995; Chester et
al 2004 and references therein).
South Fork Area - With a simple geometry and an accessible 100 m-long
continuous exposure of the fault and the wall-rocks, the South Fork area provides an
excellent investigation site for comparison of structural properties between the two
sides of the fault (Figure 1.10). The basement complex of the southwest block and
the Punchbowl sandstone of the northeast block are separated by an ultracataclasite
layer that is a few tens of centimeters wide and dips 75 degrees to the SSW.
54
Figure 1.10: Geologic map of the Punchbowl fault in the South Fork and Devil’s Chair study
areas (after Dibblee 2002). The amount and direction of dip are indicated in the South Fork area. The
array of white dots marks locations of fracture density measurements shown in Figure 1.11.
We evaluated the fault zone properties across the fault with an 85-m-long fault-
perpendicular traverse (Figures 1.10, 1.11). Nine stations, five of them in the
Punchbowl sandstone and the other four in the basement complex, were chosen to
represent the fracture density at a given distance from the fault. They are designated
as NE55, NE40, NE25, NE04 and NE00.7 for stations on the northeast side of the
fault, and SW30, SW12, SW02 and SW00.1 for stations on the southwest side of the
fault, with the numbers indicating the distance in meters from the ultracataclasite
layer. The choice of each station took into consideration the desired spacing that
enables a reliable representation of FD gradient with the highest possible exposure
quality. For quantification of the FD we used the box-count method with a 25 cm x
55
25 cm frame of exposed rock in each station. The frames were photographed and the
fractures were mapped in the field on the photos and were later digitized. Figure
1.11a shows pictures of these nine frames with the corresponding digitized fractures
maps; the assigned numbers are the Cumulative Fracture Length (CFL) for each
frame in cm.
Several tens of meters northeast of the fault in station NE55, we measured 105
cm of CFL which corresponds to a FD of 0.168. Most (if not all) of the fractures are
tensile. Fractures here are limited to pebbles and are believed to be the reaction of
the tectonic stress field applied to rigid inclusions within a compliant material
(Eidelman and Reches, 1992). In the current state of the rock, the difference in
mechanical properties between the host rock and the pebbles is negligible and
therefore fracturing should not be selective with respect to the different lithological
domains. The fractures could “select” the pebbles only when the pebbles were much
stiffer than the matrix; this could happen only before the burial and litification of the
conglomerate. Therefore, the faulting-related FD is very low and believed close to
the background damage level.
56
Figure 1.11: Fracture density profile across the Punchbowl fault near South Fork. (a) Fracturing
intensity measurements along an 85 m long traverse across the Punchbowl fault in the South Fork area
(Figure 1.10). The upper orange panel includes measurement stations on the NE side of the fault and
the lower blue panel includes stations on the SW side. The pictures frame 25x25 cm of rock
exposures. Titles of the pictures indicate the relative location from the fault (e.g. NE25 is 25 m NE of
the fault). The fracture maps are shown both on the pictures and below them with numbers indicating
cumulative fracture length cm/frame. (b) Close up view of fracture map of station SW00.1 illustrating
the inhomogeneous nature of the fractures orientation and intensity.
57
The decrease in damage level around station NE55 with respect to stations
closer to the fault probably marks the outer margins of the fault zone. This constraint
on the width of the fault zone is in good agreement with findings by Schulz and
Evans (2000) from the Punchbowl fault several kilometers east of our study area.
Closer to the fault at station NE40, fractures are not restricted to only pebbles
and the FD is 0.118 (CFL = 74). Station NE25 shows a FD of 0.194 (CFL = 121),
whereas within the fault core at station NE04 the FD jumps significantly to a value
of 0.885 (CFL = 553). The FD reduces to 0.698 (CFL = 436) immediately adjacent
to the contact with the ultracataclasite at station NE00.7. However, the nature of
fracturing in samples from these two stations is similar: all macroscopic surfaces
show slickensides population with internally consistent rake and the rock is partially
pulverized. Therefore, we expect that a large portion of the fracture surface area is in
the microscale below our mapping resolution. The ~20 cm wide sandstone belt
immediately northeast of the contact with the ultracataclasite appears to be partially
pulverized and fracture mapping in the macro scale can not express the effective
fracture density. These results express the overall qualitative observed damage
pattern on the northeast side of the fault along this section: damage to the rock starts
to increase above the background level several tens of meters from the fault and
increases dramatically within the fault core, causing fragmentation and some
pulverization of the rock.
On the southwest side of the Punchbowl fault, the granodiorite complex is
highly fractured but presents poor correlation between faulting-related structural
58
properties and fault-normal distance. At station SW30, the southwestern distal
station with respect to the fault, the FD is measured as 0.790 (CFL = 494), the
highest among the mapped stations on this side of the fault. The true FD is much
higher as most of the fractures are in the mm scale, below the mapping resolution
limit. Nevertheless, the rock is not pulverized and no macroscopic surfaces with
slicknsides were found. A similar level of fracturing intensity continues farther to the
southwest for a distance of about 100 m along the accessible portion of the wall-rock
(i.e. no qualitative FD gradient is observed). At stations SW12 and SW02, the FD is
0.539 and 0.704 (CFL = 337 and 440), respectively. The fractures at these two
stations have characteristics similar to those described for station SW30. The fracture
density rises slightly to 0.723 (CFL = 452) close to the fault at station SW00.1,
within a zone of shear belts secondary to the ultracataclasite. However, even in this
fine mapping scale, the distribution of fracture density and fracture orientation is not
homogeneous. A careful analysis of the photograph and map at this station (Figure
1.11b) shows that the chloritized foliated rock at the upper left corner presents a
medium FD with sub-horizontal long fractures that might be related to an inherited
metamorphic fabric or to a fault parallel shear. The white rock shows a very high FD
with relatively short sub-vertical fractures at its upper right part, and medium to low
FD at its upper and left parts with essentially no fractures at its lower right part. The
clear inhomogeneity of the damage in such a proximity to the fault, without
significant rheological differences between the different fracture domains, suggests
that most of these fractures are not fault-derived. The general appearance of the
59
crystalline rocks immediately southwest of the ultracataclasite is almost intact in
many places and presents very minor damage, in a sharp contrast to the shattered
appearance of the Punchbowl sandstone (Figure 1.12). Rocks in such a proximity to
the fault are expected to experience pervasive, intense and relatively homogeneous
fracturing during rupture, at least like the Punchbowl sandstone on the other side of
the fault, if no symmetry considerations are involved. From a mechanical point of
view, the crystalline rocks are expected to experience more fracturing, as the
sandstone has higher porosity and can more easily diffuse sharp stress
concentrations. Therefore, the zone of minimum damage in Figure 1.11b may
represent the fault-related FD.
Figure 1.12: Geomorphologic expression of the damage pattern across the Punchbowl fault.
View to the southeast. Slip is localized in this area in the southwest side of the ultracataclasite (bottom
of the trough, mostly covered with debris), separating the slightly fractured (almost intact) basement
rocks on the right from the heavily fragmented Punchbowl conglomerate on the left. The
conglomerate is significantly more degraded than the basement rocks within the fault core. The
asymmetry in the topographic profile of the trough is clearly related to the asymmetry in
fragmentation intensity across the fault.
60
Overall in this area, we did not observe a consistent gradient of damage level on
the southwest side as a function of distance from the fault. We also noticed the
absence of slip striae and fracture sets with preferred orientation, which are structural
features that can be associated with faulting on the Punchbowl fault and were found
on the northeast side. Despite the difficulties in comparing damage level between the
two types of rocks, the details of the fracture density traverse and the overall
intensity pattern of damage show that the damage pattern associated with slip on the
Punchbowl fault is asymmetric with more damage on the northeast side of the fault.
This asymmetry is especially pronounced by the contrast of damage intensity within
the fault core.
An independent measure for the contrast in damage content is expressed by the
morphology of the fault core (Figure 1.12). While the basement metamorphic rocks
on the southwest side are bounded by a high and continuous vertical cliff
immediately along the contact with the ultracataclasite, the sandstone on the
northeast side has developed a meter wide and a meter deep trough along the fault.
The two sides of the fault are clearly subjected to significantly different rates of
degradation. The trough is correlated with the ultracataclasite and the zone of
maximum fragmentation and pulverization on the sandstone side. It is likely that
weathering and fragmentation enhanced each other in a positive feedback
mechanism, causing the development of this trough. The steep topography on the
southwest side of the fault indicates the low rate of degradation that is possible when
the damage level is low. The possibility that rocks on the southwest side gained
61
strength to an almost intact state by healing is unlikely because re-strengthening by
healing should have operated at least partially also on the northeast side, yet no
significant healing occurred within the sandstone. Also, we only observed a few
healed fractures in the cliff on the southwest side of the core zone, supporting the
geomorphic observation.
Devil’s Chair Area - Chester and Chester (1998) studied the fine structure of the
fault in the Devil’s Chair area and determined that slip was localized along the
ultracataclasite layer. Their study further suggests that a single prominent slip
surface accommodated most of the fault’s displacement.
Figure 1.13: Major geographic features in the Devil’s Chair area and locations of the FD
measurement sites marked with white dots (see Figure 1.10 for general location). Station C is in the
vicinity of the study sites of Chester and Chester (1998).
62
We compared the intensity of damage between the igneous rocks on the
southwest and the Punchbowl sandstone on the northeast within the fault core by
measuring linear fracture density along a meter-long scan line perpendicular to the
ultracataclasite on both of its sides. Three stations along 250 m of the fault were
chosen for this measurement (Figure 1.13) and in all of them, the density of fractures
is larger within the Punchbowl sandstone compared to the amount of fractures in the
monzogranite and other rocks southwest of the fault (Table 1.1). The overall
asymmetry in fracture count is about 71%.
Table 1.1: Devil’s Punchbowl Linear fracture density. Number of fractures per meter on each side
of the Punchbowl fault, as measured in the Devil’s Chair area. The overall asymmetry is 71%.
site Punchbowl (fractures/m) Basement (fractures/m)
A 56 33
B 73 19
C 43 17
Mean 57 23
Summary - Fracture density measurements done at two different scales at two
different sites 1.5 km apart along the Punchbowl fault indicate that the Punchbowl
sandstone on the northeast side of the fault is more damaged than the igneous
complex on the southwest side of the fault. The FD traverse near South Fork shows
that the intensity and type of damage in the Punchbowl sandstone is clearly related to
faulting on the Punchbowl fault as damage intensity is proportional to the distance
from the fault, and the fractures are organized with a preferred orientation and
exhibit internally consistent kinematic indicators. In contrast, damage in the
granodiorite shows no clear dependency on distance from the fault and the overall
63
damage style may be related to other phases of deformation or to inherited
metamorphic fabric. The morphology of the fault core at the South Fork site
independently shows that the northeast side is significantly more damaged, leading
to the development of a distinct geomorphic trough within the sandstone, as oppose
to a cliff of nearly intact rocks on the southwest.
Discussion
We have established that at all examined localities there is an asymmetric
pattern of fault zone damage with respect to the current principal slip surface and
that the sense of damage asymmetry is consistent along each studied fault section.
Those patterns are manifested at different scales, in different forms, and we expect to
add to our understanding of the breadth of expression of damage and symmetry
properties as we continue to explore new sites. Nevertheless, our observations
demonstrate that macro-scale fault zone damage from the gouge zone outward to at
least tens of meters expresses distinct asymmetry in form and extent. Those sets of
observations can be interpreted as the result, and hence also as indicators, of
preferred rupture direction of earthquakes. Below we assess the possibility that the
seismogenic depth structure of a fault is controlling the geologic damage pattern by
dictating the preferred direction of ruptures. This can be done by correlating the
sense of observed damage asymmetry to the local velocity structure at depth, since
this correlation is an expected outcome of rupture along a material interface (Ben-
Zion and Shi, 2005).
64
Asymmetry of structural properties in light of velocity structure
The San Jacinto Fault, Anza: The gouge fabric found in the three studied
exposures on the SJF near Anza shows that the northeastern side of the current PSS
is more damaged. This sense of asymmetry is consistent with high resolution seismic
imaging by Lewis et al. (2005), based on fault zone trapped waves recorded across
the three branches of the SJF south of Anza. Their inversion results indicate the
existence of ~100 wide fault zone trapping structures at the different branches that
extend to a depth of about 3.5 km. The seismic imaging shows further that the
trapping structure at each of the branches is not centered on the surface trace of the
fault, but is offset 50-100 m to the northeast. The results imply that more damaged
fault zone material is present on the northeast side of each fault. The damage
asymmetry in our geological mapping and the seismic trapped waves study is
compatible with northwestward rupture propagation direction, with the primary
tensional quadrant of most rupture events on the northeast side of the fault.
The local velocity structure of the Anza seismic gap was imaged by Scott et al.
(1994), who showed that at depths of 3 km, 6 km and to some extent also 9 km, the
northeast side of the fault has faster seismic velocities. This is the side that has more
damaged rock in our geological studies and in the seismic imaging of Lewis et al.
(2005). These results are in agreement with regional imaging of Magistrale and
Sanders (1995) and Shapiro et al. (2005), who show the same sense of velocity
contrast in the Anza area and farther to the southeast. The existence of more rock
damage on the side of the fault that has faster seismic velocity at seismogenic depth
65
is expected if a material interface controls the preferred propagation direction on the
fault.
The San Andreas Fault in the Mojave: The consistency of our gouge and fault
zone scale observations from the Palmdale – Little Rock area of the SAF most
probably reflects the asymmetric damage pattern across the fault in this area with the
northeastern side more damaged. This conclusion probably also applies for the entire
Mojave section of the fault, based on the compatibility of the sense of asymmetry
discussed in this work with results associated with pulverized fault zone rocks. Dor
et al. (2006b) mapped the distribution of pulverized rocks along a 140 km-long
section of the SAF in the Mojave, overlapping with our study area. They found that
the pulverized rock is a systematic structural feature of the fault that can be found,
with just a few exceptions, wherever crystalline rocks crop out within up to 200 m
from the active trace of the fault. The distribution of the outcrops with respect to the
fault was found to be asymmetric, with most of the pulverized rock exposures being
on the northeastern side of the fault. This pattern may reflect, at least partially, the
current distribution of available exposures. Nevertheless, in several places where
comparison is possible between the two sides of the fault, pulverization is more
intense and the zone of pulverization is wider on the NE side of the fault, supporting
the near-field macroscopic observations presented here. This combined multi-scale
set of observations suggests that the northeastern side of the SAF in the Mojave has
accumulated more damage at various scales and forms during repeated ruptures. A
possible explanation for this damage distribution pattern is that most of the ruptures
66
during the recent geological history of the fault propagated from southeast to
northwest in the Mojave area, having their tensional quadrant on the northeastern
side of the fault.
The velocity structure along the Mojave section of the SAF has been studied to
some extent. Shapiro et al. (2005) cross-correlated ambient seismic noise at USArray
stations and constructed tomographic images of principal geologic units in
California. Their 7.5 and 15 s period images based on surface waves indicate that
overall, the Mojave block shows higher seismic velocities with respect to the
southwest side of the fault. This velocity contrast continues farther to the southeast
and is of opposite sense to the velocity contrast across the SAF northwest of the big
bend. As part of the Los Angeles Region Seismic Experiment (LARSE), Fuis et al.
(2001) report for line 1 (crossing the SAF slightly southeast of our study areas) that
in contrast to velocities below the San Gabriel mountains, velocities to 10 km depth
below the Mojave are consistently higher than average laboratory velocities for the
Pelona schist, and therefore the Pelona schist does not appear to be present in the
Mojave desert beneath line 1. This agrees with aeromagnetic data along this transect.
Line 2 of the LARSE experiment crossed the SAF in the Elizabeth Lake area,
northwest of our west Palmdale site. Lutter et al. (2004) report for line 2 that
although poorly resolved, basement velocities at depth are higher north of the SAF.
That inference is also supported by gravity data along the transect. Fuis et al. (2003)
show generalized velocity cross sections based on lines 1 and 2, with the Mojave
side having 8-10% faster seismic velocities at depth. If these two studies are
67
representative of the entire Mojave segment, the faster velocity side of the fault at
seismogenic depth correlates with the side of the fault that expresses more damage,
as expected for rupture along a material interface (Ben-Zion and Shi, 2005).
In both studied sections of the SAF and the SJF, the asymmetry of structural
damage may reflect a preferred propagation direction associated with contrasting
elastic properties at depth. Across both the SAF and the SJF, the velocity contrast
continues farther southeast of the study areas, where ruptures could have nucleated in
order to propagate to the northwest through our investigation sites.
The Punchbowl Fault: The excess in fault–related damage on the northeast side
of the Punchbowl fault compare to its southwest side in the studied area suggests that
more paleoearthquakes propagated to the northwest than to the southeast when the
fault was active. The velocity structure of the Punchbowl Fault at seismogenic depth
is unknown. Based on our observed damage pattern we estimate that the southwest
side of the fault has slower seismic velocity at depth with respect to the Mojave side.
Our observations are compatible with persistent occurrence of earthquakes along
the examined fault sections in the form of wrinkle-like ruptures on material
interfaces with preferred propagation direction. We are aware that the observations
are subjected to various uncertainties, and of the general non-unique nature of any
interpretation (see details in section 4.4). Nevertheless, the observed correlation of
multiple manifestations of rock damage with the velocity contrast across a fault, at
several scales along different sites of geometrically simple segments, supports our
interpretation. We also note that wrinkle-like mode of rupture provides a possible
68
explanation for a variety of outstanding geophysical issues, including the lack of
frictional melting products, short rise time of earthquake slip and suppression of
branching in the structures of large faults (e.g., Ben-Zion, 2001).
Historic earthquake behavior that might be explained by preferred rupture
propagation direction.
While a few earthquakes are statistically insignificant and can not indicate the
long term preferred rupture direction associated with a fault section (which is why
we study the geologic record, as mentioned in the introduction), the timing and
spatial distribution of historical earthquakes on major strike-slip faults may be
explained, in part, by the existence (or lack of) a preferred rupture direction.
The 19
th
century sequence of earthquakes on the SAF: The elapsed time of about
300 years since the penultimate earthquake prior to 1857 (early 1500 A.D., Lindvall
et al., 2002) and the sequence of the 1812 and 1857 earthquakes suggest that in the
early 19
th
century, the entire south-central part of the SAF was ready to fail. It is thus
interesting to ask why the 1812 rupture did not extend over this entire section of the
fault. Moreover, why did the 1857 rupture overlap about 60 to 100 km with the 1812
rupture (Figure 1.14), overshooting several meters of slip into the rupture zone of the
earlier earthquake only 45 years later (Weldon et al., 2002; Fumal et al., 1993)? And
why is the inferred rupture direction for the 1857 Ft. Tejon earthquake (Sieh, 1978b,
Agnew and Sieh, 1978) the opposite of our inferred preferred rupture direction for
the Mojave?
69
Figure 1.14: The south-central San Andreas Fault system in California and the known extent of
the partially overlapping 1812 and 1857 ruptures (heavy and light gray, respectively). The
propagation direction of the1812 rupture is unknown while that of the 1857 rupture was inferred to be
toward the southeast, based on locations of candidate foreshocks. Even if correct, the propagation
direction in the Mojave may have been different if the earthquake was a multi-shock event.
To address these questions we need to examine the available information.
Seismic imaging of the SAF near Parkfield shows 5-20% lower seismic velocity in
the seismogenic zone on the northeast side of the fault (e.g., Ben-Zion et al., 1992;
Eberhart-Phillips and Michael, 1993). Therefore, a southeastward propagation
direction of large (M>7) earthquakes is expected in that area according to the model
of rupture along a material interface. Nucleation in the Carrizo plain and
southeastward propagation was inferred by Sieh (1978b) for the 1857 earthquake. In
the last 40% or so of its final length, the rupture propagated in the Mojave, against
our assumed long term preferred rupture direction inferred from the observed rock
damage asymmetry. High resolution seismic imaging of the SAF south of the
Parkfield area is lacking, but as mentioned earlier the regional imaging by Shapiro et
al. (2005) and traverses of the LARSE experiment (Fuis et al. 2003) suggest that the
sense of velocity contrast observed for the Parkfield area flips somewhere toward the
70
Mojave. It is thus plausible that the last part of the 1857 event propagated into a zone
of opposite sense of velocity contrast in the Mojave area, and was finally arrested
toward Cajon Pass. A support for this comes from the slip function of this earthquake
(Sieh, 1978a). The displacement peaked at ~8-9 m around Wallace Creek, then
subsides to ~6-7 m along the southern Carrizo plain and drops to ~3 m passing Tejon
Pass into the Mojave with values as low as 1-1.5 m in Wrightwood. The 1812
earthquake has an unknown propagation direction, but from the correlation between
the sense of damage asymmetry and the apparent velocity contrast we speculate that
this earthquake propagated to the northwest. The strong velocity contrast in the
vicinity of its southeastern most known part (Shapiro et al. 2005), is compatible with
northwestward propagation direction. If this is correct, the 1812 rupture was
presumably arrested in the northwestern edge of the Mojave due to a change in the
velocity structure, and thus did not leave evidence in the paleoseismic record of
Frazier Mountain (Lindvall et al., 2002). More details about the velocity structure of
the SAF in this area and additional paleoseismic data about the 1812 rupture are
needed in order to clarify more aspects of this problem. We also note that with the
limited available information on the rupture history of the 1857 earthquake, it is
possible that this earthquake was a compound event with two (or more) shocks. The
first and largest one nucleating in the northwestern Carrizo plain, propagating
southeastward and arresting somewhere in the northwestern Mojave (maybe due to
its entrance into a zone of decrease and then a change in the velocity contrast); the
other shock, responding to the large stress transfer of the first one and overriding part
71
of the previously yielded 1812 rupture zone, nucleating in the southeastern Mojave
and propagating toward the northwest. A two shock scenario as above is compatible
with the known velocity structure along the rupture length and with the geological
signal for preferred rupture direction in the Mojave.
The 20
th
century sequence of earthquakes on the North Anatolian Fault (NAF):
During the past century, the NAF experienced a remarkable sequence of ruptures that
began with the great 1939 (M7.9) earthquake near Erzincan and proceeded westward
with earthquakes in 1942, 1943, 1944, 1951, 1957, 1967, 1999a, and 1999b. Stein et
al. (1997) attributed this rupture sequence to successive failures of fault segments
due to stress transfer from failed neighboring segments. However, the 1943
earthquake did not nucleate in the region of stress increase but rather at the opposite
end of the final rupture, far to the west. Similarly, the 1999 August earthquake
nucleated near the head of Izmit Bay and ruptured bilaterally with the main rupture
directed eastward toward the earlier failures. This was followed by the continuation
of rupture to the east in the November 1999 Düzce earthquake.
The general westward migration of the earthquakes is an expected outcome of
rupture along a material interface at the NAF, if the southern block is generally the
side with lower seismic velocity (e.g., Sengor et al., 2005). However, three events of
the 1939-1999 sequence ruptured eastward against the overall un-zipping direction
of the fault. The 1943 and the 1999a earthquakes also nucleated away from the area
of largest inferred stress-transfer loading. The loading, nucleation and rupture
scenarios of these events are similar to the experimental setting and results of
72
Anooshehpoor and Brune (1999) with two different foam rubber blocks. This
suggests that the rupture directions of those events are controlled by a reversed local
velocity structure (i.e., lower seismic velocity north of the fault). If correct, we
expect in those fault sections that more damage is present in the southern block with
the assumed faster seismic velocity, and opposite sense of damage asymmetry at the
other rupture zones along the fault. These expectations can be tested with detailed
field observations of the type discussed in section 3.
Possible Related Observations
In the European Alps, study of the exhumed Gole Larghe fault showed a distinct
asymmetry that was interpreted as the result of preferred direction of rupture
propagation (Di Toro et al. 2005). The Gole Larghe fault branches to the east from
the large right lateral Tonale fault into a tonalitic intrusion. Right-lateral slip along
the Gole Larghe fault took place at a depth of 9-11 km about 30 Ma ago (Di Toro
and Pennacchiioni, 2004) and was at least in part seismic as indicated by numerous
veins of pseudotachylytes. Di Toro et al. (2005) studied the orientation of over 600
injected veins filled with pseudotachylytes, branching from several secondary faults
of the Gole Larghe fault zone and exposed in the glaciated outcrops of upper Val di
Genova. They found that veins are asymmetrically distributed with respect to the
fault, with 67.7% of the veins intruding the southern bounding block. They
concluded that those veins were formed in the tensional quadrant of propagating
mode II ruptures and that the asymmetric fracturing pattern indicates that most of the
73
paleoearthquakes on the Gole Larghe fault propagated from west to east, either
nucleating at the junction of the Gole Larghe fault with the larger Tonale fault or
branching during large Tonale earthquakes. The Tonale fault is a segment of the
major tectonic lineament of the Alps and one of the main lithological boundaries
(Steck, and Hunziker, 1994). It is thus likely that the large Tonale paleoearthquakes
were affected by material contrast which may have dictated the direction of ruptures.
The Nojima fault in Japan, along which the 1995 Kobe earthquake occurred,
displays structural asymmetry at multiple scales and different types of signals. The
Nojima fault is a right-lateral strike-slip fault with a minor reverse component,
striking SW-NE along the northwestern margins of Awaji Island, Japan (Mizuno et
al. 1990). The fault juxtaposes the Miocene Iwaya Fm. and the Plio-Pleistocene
Osaka group on the northwest side of the fault against Cretaceous granitoids on the
south east side of the fault. The geological survey of Japan recovered a drilling core
from the Nojima fault about a year after the Kobe earthquake. The inclination of the
fault near the drilling site was estimated to be 83º and the main fault crossed the drill
hole at depth of 625 m. Tanaka et al. (2001) and Ohtani et al. (2000) identified one
of the seven shear zones found within a 55 m wide fault zone as the primary
seismically-active fault core. Within the fault core, the breccia and ultracataclasite
layers are located almost exclusively on the southeastern side of the principal slip
surface of the fault. Except locally, the northwest side of the fault is weakly
deformed (Ohtani et al., 2000). Tanaka et al. (2001) defined the principal Nojima
fault surface as the northwestern boundary of the ultracataclasite layer at a depth of
74
625.27 m, where it separates the fault core (ultracataclasite + breccia) on the
southeast from damaged fault zone rocks on the northwest.
The structure of the Nojima Fault zone as it crops out southwest of the drill site
(Mizoguchi et al. 2000) can be correlated to the structure of the fault core at depth,
with the same sense of asymmetry and similar contrast between the two sides of the
fault. The PSS, represented here by a 0.1- 0.15 m wide layer of fault gouge, separates
the Plio-Pleistocene Osaka group on the northwest from a two meter wide breccia
layer on the southeast, with gradual transition to a 10 meter wide zone of fractured
granite. Additional manifestation of the asymmetry at the fault zone scale is given by
lab analysis of rock strength of recovered core samples (Lockner et al., 1999). The
results from that study suggest that within a fault-zone width of about 40 m, rocks on
the northwest side have about 40% higher peak shear strength with respect to rocks
on the southeast side (although based on a small population of samples; D. Lockner,
pers. comm.). Thus, both a careful structural study at the fault core scale and initial
results of lab analysis of rock samples at the fault zone scale of the drilling core of
the Nojima fault correlate well with the fault zone structure in a surface exposure.
All those observations indicate that the fault zone structure is highly asymmetric
with the southeast side being more damaged. These observations are consistent with
a statistical preference of rupture direction to the northeast. The velocity structure
associated with the surface geology at the Awaji Island region has rocks with faster
seismic velocity to the southeast. If this velocity structure extends to depth, the
75
observed damage asymmetry is compatible with the theoretical predictions for
ruptures along a material interface.
Possible Interpretation Problems
Many processes can contribute to the fabric and composition of fault-zone rocks.
First, a mature fault zone reflects the cumulative effect of many thousands of
earthquakes, and the mechanical nature of those earthquakes may have changed over
the faults history, printing different and sometimes contrasting signals in the fault-
zone rocks. Where migration of the active strand of the fault occurred, the apparent
observed symmetry properties may not reflect the current distribution of damage
around the currently active fault. Second, a single rupture can alter the fabric in
several ways and have diverse imprint on the rocks (e.g. fracturing due to frictional
instability vs. fracturing due to reduction of normal stress). Third, there are many
interseismic processes acting in a fault zone that can influence not only the
composition but also the internal structure, mainly the microstructure of fault rocks.
Finally, all the above are influenced by both the overall geological setting and local
site effects. In addition, we have to justify our ability to infer properties of dynamic
earthquake ruptures from surface observations. These issues are discussed briefly
below.
Deducing dynamic rupture behavior at seismogenic depth from surface
observations - Earthquake-related observations would be more reliable if done at
seismogenic depth where the bulk of seismic slip occurs. However, the focal depths
76
of earthquakes are generally not accessible, other than in deep mines (e.g. Dor et al.,
2001; Reches and Dewers, 2005; Wilson et al., 2005). Many valuable studies have
been conducted on inactive exhumed faults (Sibson, 1989; Chester et al., 1993;
Schulz and Evans, 2000, and references therein), but these studies are limited in the
sense that they lack knowledge on the seismogenic behavior of the faults (and
alterations during the exhumation process). In this study we use a combined
approach in which we utilize and create exposures on exhumed (Punchbowl) and on
currently active faults (SAF, SJF). While the bulk of seismic slip occurs at
seismogenic depth, simulations of damage generation indicate that conditions in
favor of fault damage creation are limited to the upper 3 km or so of the crust (Ben-
Zion and Shi, 2005). This is similar to the depth extent of imaged low velocity
trapping structures (e.g. Ben-Zion et al., 2003; Peng et al., 2003, Lewis et al., 2005)
and is supported by our related geologic observations of damaged fault zone rocks
that were never buried deep (Dor et al. 2006b). The behavior at the surface is
assumed to be generally dictated by the behavior of dynamic rupture at depth. This is
especially so in the case of symmetry properties that are expected to result from a
wrinkle-like rupture mode, because although the normal stress decreases with
decreasing depth, the material contrast is generally larger in the top few km of the
crust (e.g., Ben-Zion et al., 1992; Eberhart-Phillips and Michael, 1993), thus
promoting strong signals at shallow depths. The rupture is expected to carry its
asymmetric dynamic properties to the surface, where damage generation may be
amplified due to the increase in contrast of the velocity structure. Creation of damage
77
that is concentrated in the top few km will not affect much the elastic properties of
the rocks at depth and therefore will not change significantly the velocity contrast
where it has a control on dynamic effects. Hence, the process is controlled by stable
conditions in the seismogenic depth and has increased manifestations as the depth
becomes shallower.
Contrast in surface lithology and its influence on damage pattern - Faults
commonly juxtapose on the ground surface rocks with contrasts in their strength,
competency, isotropy, flow density and other factors that, to some extent, might
influence the damage distribution across them. Moreover, if this is a dominant factor,
we expect to see some correlation between the strength of the rocks and the intensity
of damage they present. However, observations made in this study and in a parallel
work (Dor et al. 2006b) indicate that damage favors the northeast side of the SAF,
regardless of the rock type. In addition, we show here that in the Punchbowl Fault,
even though pulverization should favor the denser igneous rocks of the San Gabriel
complex, the more porous sandstone on the northeast side of the fault is more
damaged and perhaps even pulverized.
The effect of fault dip on asymmetry - This study focuses on strike-slip faults
because the dip of the fault alone can break the symmetry. Brune et al. (1999)
showed that more damage is expected in the hanging wall of a thrust fault and
especially in the near surface region due to waves trapped in the wedge-shaped
hanging wall of the fault. Not all the fault strands that we studied are perfectly
vertical and their inclination close to the ground surface ranges between 60 to 90
78
degrees, with a common secondary component of dip-slip. If so, how do we
eliminate the inclination factor? If the dip of the fault was a governing factor in the
distribution of damage across the faults, we would expect the damage pattern to be
correlated with the dip. However, the fault core scale damage in the SJF is
concentrated in the hanging wall side, whereas fault core scale damage in exposures
with dipping structures of the SAF and overall damage in the Punchbowl Fault is
more pronounced on the footwall side.
Complex history of the fault, temporal changes in velocity contrast at one site -
Faults with large cumulative displacement may juxtapose different rock bodies
against each other during the faulting history, and any segment may have
experienced changes in the sense of velocity contrast. The geology of one locality
can potentially represent the cumulative effects of several faulting phases in which
the preferred rupture direction has flipped. A migration of the active strand of the
fault over time can also bias the apparent symmetry properties of the fault. In the
clay-rich active or recently active fault gouge, there is less danger of confusing
former faulting phases with the current one, because the material itself evolves with
time. In addition to the accumulated primary (detrital) material, newly-grown
(authigenic), fine-grained phyllosilicates are constantly added (van der Pluijm et al.,
2001). Additional detrital material is also added, as evident in the translocation of
dark, organic-rich clay and the addition of detrital sand and silt downward into open
fractures that are produced by surface ruptures. The entire gouge body is therefore
evolving fast enough so that it is unlikely that ancient fabrics will be preserved for
79
any length of time. This is confirmed by the relatively in-cohesive nature of the
fracture surfaces in an active gouge. If these fractures are inactive long enough, they
become more cohesive and massive (like the gouge on the south west side of the SJF
and in the dip-slip structure in the trench exposure near Palmdale).
The wall-rock within the fault zone may preserve fabric from ancient faulting
phases, but this fabric will most likely be overprinted by damage from the current or
most recent faulting phase. In some cases it is possible to separate different
generations of fractures using cross-cutting relations (e.g. Vermilye and Schulz,
1998, 1999). When uncertainty still exists, it is essential to study multiple signals at
multiple scales as done here, and to correlate the sense of asymmetry with additional
independent observations such as seismic trapping structures. The correlation of
several expressions of the signal from several sites along the fault can reduce the
non-uniqueness that is naturally associated with observations at one site. In this study
we have documented such a correlation of a consistent excess of damage on one side
of the fault expressed in different ways and at several scales. The systematic damage
asymmetry is correlated with the local velocity structure as predicted for rupture
along a material interface (Ben-Zion and Shi, 2005).
Deviation of a single rupture or a temporal cluster of ruptures from the
statistical preferred direction - A fault-zone is a complex system with many possible
influencing factors; the interaction of the fault with other faults can make its
behavior even more complex. Therefore, the preference for rupture direction is only
expected to be statistical over the long-term, and single ruptures might propagate
80
against the preferred direction. Segments of a fault can also be in the spatial or the
temporal transition from one preferred direction to the other, or may not have a
preferred rupture direction at all. Various factors can reduce the statistical preference
related to velocity contrast as a prime factor and hence obscure the intensity of the
geological signals discussed in this work.
The influence of geometrical and compositional complexities - Kinks, bends,
step-overs, branches and secondary faults can locally cause compression or tension,
and possibly bias the signature of the fundamental dynamic effects associated with
large scale ruptures. The scale of geometrical irregularities that may interfere with
the signal is not very clear. For example, it is unknown to what extent the regional
big bend of the SAF around Tejon Pass may affect fault core scale damage pattern in
that area. Significant compositional heterogeneities of the rocks at seismogenic
depths may also influence the faulting pattern by locally changing the sense of the
velocity contrast (i.e., a maffic, dense, regional scale intrusion in a leucocratic lighter
rock). To overcome difficulties associated with geometrical perturbations, we
selected sites with minimum structural complexities in sections of the faults that are
relatively straight, simple and vertical. Nevertheless geometrical irregularities can
occur at all scales. It is therefore important to show consistency in multi-signal
multi-scale of observations along a fault section. Our results from the examined
exposures show an overall consistent pattern of damage asymmetry that is
independent of known or unknown structural and compositional complexities in the
vicinity of a particular exposure.
81
Finally, we emphasize that while single ruptures can produce asymmetric
damage pattern at various scales (e.g. Johnson et al., 1994; Vermilye and Schulz,
1998, 1999; Reches and Lockner, 1994), the observations discussed in this work
reflect clearly the imprint of a large population of earthquakes. This is attested by the
intensity, multi-signal and multi-scale nature of the observations (from fault core to
geomorphic signatures), and by the observed fabrics and field relations which
indicate that certain fault zone components were active at different time periods.
Conclusions
We have made numerous in-situ observations of damage asymmetry at various
scales along three major faults in southern California. We attempted to develop a
methodology to infer systematic structural asymmetry at different scales
(recognizing various potential interpretation problems), and use the observations to
test the hypothesis that large earthquake ruptures on strike-slip faults that separate
different crustal blocks have a preferred propagation direction (e.g., Weertman,
1980; Andrews and Ben-Zion, 1997; Ranjith and Rice, 2001; Ben-Zion and Huang,
2002; Shi and Ben-Zion, 2006). The results indicate that the observed sense of
asymmetry is correlated with the local velocity structure, where known, and suggests
that the asymmetry is the long term result from a preferred dynamic rupture
direction, in agreement with the theoretical predictions. In addition, we pointed out
with examples of historic rupture events from the San Andreas and North Anatolian
faults, that the property of a preferred rupture direction may partially explain the
82
distribution in time and space of large earthquake ruptures and their nucleation
points. The inferences that were presented here could possibly apply to other large
strike-slip faults, where geological studies of symmetry properties may be utilized to
infer on possible preferred propagation directions of earthquake ruptures.
The propagation of earthquakes as wrinkle-like ruptures in a preferred direction
can have fundamental implications for many aspects of earthquake physics and
hazard. The interaction between slip and normal stress along a material interface can
dynamically reduce the frictional strength, potentially to zero, making material
interfaces mechanically favored surfaces for rupture propagation (e.g., Ben-Zion,
2001; Brietzke and Ben-Zion, 2005). This may affect strongly the effective
constitutive laws, short rise-time of earthquake slip, the generated frictional heat,
suppression of branching, the evolution of fault zones, and expected seismic shaking
hazard.
The geological signals for preferred rupture direction are highly diverse. In this
paper, we presented several such signals based on observations in the gouge and
fault zone scales of large strike-slip faults in southern California. In a parallel study
by Dor et al. (2006b), we discuss additional observations associated with pulverized
fault zone rocks. Further progress in theory can provide more detailed predictions on
expected properties of the damage that may be tested further by additional multi-
disciplinary multi-scale observations. A continuing feedback between theoretical
predictions and field observations can lead to the recognition of new types of signals
83
that may be used to deduce fundamental properties of earthquake behavior from in-
situ observations of fault zone structure.
Acknowledgements
We thank Jim Brune for useful discussions that sharpened our description of the
results, and Joe Ibarzabal for allowing us to work in his property near Tierra Subida
Road in Palmdale. The paper benefited from useful comments by Judi Chester,
Diane Moore, Rasool Anooshehpoor and Nathan Benesh. The work was funded by
the National Science Foundation (grant EAR-0409048) and the Southern California
Earthquake Center (based on NSF cooperative agreement EAR-8920136 and United
States Geological Survey cooperative agreement 14-08-0001-A0899).
84
Chapter 2: Pulverized Rocks in the Mojave section of the San Andreas
Fault Zone
Published at Pure and Applied Geophysics, 2006.
Co-authors: Yehuda Ben-Zion
1
, Thomas K. Rockwell
2
and Jim Brune
3
1. Department of Earth Sciences, University of Southern California, Los
Angeles, CA 90089-0740, USA.
2. Department of Geological Sciences, San Diego State University, San Diego,
CA, 92182-1020, USA.
3. Nevada Seismological Laboratory, University of Nevada, Reno, NV 89557,
USA.
Abstract
We present mapping of pulverized fault zone rocks along a 140 km long section
of the San Andreas Fault in the Mojave Desert. The results show that almost every
outcrop of crystalline rock within about 100-m-wide belt along this fault section is
pulverized and lacks significant shear. We find structural similarities between the
San Andreas Fault zone and exhumed faults of the San Andreas system, although
pulverized rocks are not common in all of them. About 70% of the pulverized fault
zone rocks appear on the northeast side of the principal slip zone of the San Andreas
Fault, possibly reflecting an asymmetric structure of the damage zone. Detailed
mapping at selected sites, as well as previous mapping of rock damage at smaller
85
scales, are consistent with the large-scale asymmetric pattern of the pulverized rocks.
A possible pulverization of sedimentary rocks, inferences from regional uplift
indicators, and theoretical considerations imply that pulverization along this portion
of the fault occurred in the top few km of the crust. The width of the pulverized fault
zone rocks and inferred depth extent of pulverization are similar to the dimensions of
imaged low velocity fault zone layers that act as waveguides for seismic trapped
waves. The side of the fault that appears to sustain more damage is the block with
faster seismic velocities at seismogenic depth. This correlation and the inferred
shallow depth for pulverization are compatible with predictions for wrinkle-like
ruptures along a material interface, with a preferred northwest propagation direction
of large earthquakes on the Mojave section of the fault.
Introduction
The term Pulverized Fault Zone Rocks (PFZR) refers mainly to crystalline
plutonic and metamorphic rocks that were mechanically pulverized to the micron or
finer scale, while preserving most of their original fabric and crystal boundaries.
Outcrops of such rocks were observed long ago but little attempt was made to
characterize them systematically until Brune [1] noted the lack of significant shear
parallel to the San Andreas Fault (SAF) within PFZR in several locations. Those
findings were followed by the detailed study of Wilson et al. [2] in a road cut
exposure of Tejon Lookout granite adjacent to the slip zone of the SAF at Tejon
Pass. They mapped the local distribution and internal structure of the PFZR and
86
analyzed its particle size distribution. The main findings from the studies associated
with the Tejon Pass exposure are: 1) The PFZR forms ~70-100 m wide fault zone
layer in the immediate vicinity of the localized slip. 2) There are myriad small scale
fractures within the PFZR with no preference for shear parallel to the SAF. 3) The
PFZR lacks significant amounts of weathering products and is distinctly different
from grus (in-situ accumulation of disaggregated plutonic material). 4) The PFZR
layer is pervasively pulverized to the sub-micron scale.
The observations from Tejon Pass confirm that the initially observed
pulverization is the result of mechanical processes rather than a weathering product.
In addition, the abundance of tension features and the extreme reduction of grain size
without distortion of the rock fabric imply that the protolite of the PFZR was
subjected to a strong tensional stress, apparently due to dynamic reduction of normal
stress associated with slip during SAF earthquakes. Brune et al. [3] observed in
laboratory experiments with foam rubber blocks that the passage of rupture is
correlated with significant reduction of normal stress associated with vibrations
normal to the interface and separation during slip. A summary of several
mechanisms that can produce dynamic reduction of normal stress during earthquake
ruptures is given by Ben-Zion [4]. These include acoustic fluidization [5], collisions
of rough surfaces [6] and a variety of fluid effects (e.g., [7-9]).
Another possible mechanism for strong dynamic reduction of normal stress
across large faults, with testable predictions on the structure of those faults, is rupture
along a material interface that separates different solids (e.g., [10-13, 4]). Numerical
87
simulations show that mode II ruptures on a bimaterial interface tend to evolve for
ranges of conditions with propagation distance to wrinkle-like ruptures with several
characteristic features (e.g., [14-18]). These include: (1) a preferred or more vigorous
direction of rupture propagation that is the same as the direction of slip in the slower
velocity (more compliant) solid, (2) strong dynamic reduction of normal stress at the
tip propagating in the preferred direction, (3) strongly asymmetric motion across the
fault, and (4) tendency of the crack-tip region with significant slip-velocity to
become narrower and higher with propagation distance. Property (1) has implications
for the symmetry properties of the damage pattern generated by many earthquakes
on the fault. Properties (2)-(4) can produce tensile dynamic stress field leading in
some conditions to "opening modes" of rupture. In some ranges of parameters these
characteristics exist only in a weak form (e.g., [19]). See related discussion in [20].
Earlier studies [21-22] suggested that significant off-fault damage is produced
only in the vicinity of fault jogs (and other geometrical complexities). The
consistency of pulverization along the entire relatively-straight Mojave section of the
SAF (see below) suggests that the observed pulverization is the outcome of a
fundamental dynamic property of earthquake ruptures rather than a site-related
effect. Theoretical and numerical studies of dynamic ruptures indicate that off-fault
damage occurs primarily in the tensional quadrants of the radiated fields [23-28]. If
earthquakes on a given fault section propagate predominately as bilateral ruptures, or
as unilateral ruptures without a preferred propagation direction, the cumulative
pattern of rock damage generated by many events will be approximately symmetric
88
across the fault. However, if earthquakes on a given fault section have a preferred
propagation direction, the cumulative damage pattern will be asymmetric, with more
damage in the tensional quadrant associated with the preferred propagation direction.
Ben-Zion and Shi [27] simulated dynamic ruptures along a bimaterial interface in a
model that includes spontaneous generation of damage in the bulk. The results show
that in such cases, damage is generated primarily on the stiffer side of the fault,
which is in the tensional quadrant of the radiated seismic field for the preferred
propagation direction. Damage generation is enhanced by the dynamic reduction of
normal stress at the propagating tip, and the zone of intense damage has an
approximately constant width that is related to the rupture pulse width. The
simulations of Ben-Zion and Shi [27] suggest further that significant generation of
rock damage, under realistic conditions of velocity contrast and material properties,
is limited to the top few km of the crust. See also [28]. Thus the generated damage
will not affect significantly the velocity structure at depth that controls the rupture
dynamics. The above results provide clear predictions that can be tested by geologic
observations of rock damage across large strike slip faults.
Recent observational studies provided information on symmetry properties of
off-fault damage. Dor et al. [29] performed detailed geological mapping of rock
damage over fault core scales of ~0.01-1 m and fault zone scales of ~1-10s of m in
the structures of the SAF, San Jacinto Fault and the Punchbowl Fault in southern
California. They observed consistent asymmetry in the distribution of gouge and
fault zone damage across the principal slip zones of those faults. For cases where the
89
velocity structure of the fault is known they showed that considerably more damage
appears on the crustal blocks that have faster seismic velocities at depth [30-33].
High resolution imaging of the local velocity structure of the San Jacinto fault zone
based on seismic trapped waves [34] indicates the existence of a 100 m wide
damaged fault zone rock in the top 3-5 km of the crust, with a similar sense of
asymmetry as in the geological mapping of [29]. Inversion of fault zone head and
direct P waves in the Bear valley section of the SAF also suggest the existence of a
shallow damage zone that is shifted toward the faster velocity block [35].
In this work we present geologic mapping of PFZR along the Mojave section of
the SAF, with a focus on the overall large-scale properties of the PFZR.
Complementing the mapping of [29] over smaller scales of observations, we provide
mapping results over a damage zone scale of ~10s-100s of m. Our observations show
that PFZR are common along the SAF in the Mojave and probably occupy a ~100 m
wide tabular sub-vertical zone parallel to the fault. In addition, we find that the
distribution of the exposed PFZR is not symmetric with respect to the principal slip
zone, but is shifted on the average to the northeast side of the fault. Based on several
observational arguments (mainly field relations between different rock units and the
apparent damage found in sedimentary rocks), we infer that the observed
pulverization occurred in the top few km of the crust. The large scale properties of
the mapped body of PFZR appear to be similar to those associated with fault zone
seismic waveguide structures. The observations are compatible with predictions for
rupture along a material interface.
90
Observations
Approach
Our study area spans the Mojave section of the SAF between approximately
Tejon pass and Cajon pass (Figure 2.1). This section of the fault was chosen to
extend the smaller scale observations of [29], to provide a regional context for the
detailed studies of [2] at Tejon pass, and for the following additional reasons. (1) The
geometry of the fault along the mapped stretch is relatively simple; the fault section
is near-vertical, relatively-straight, and the big bends of the SAF are outside the
mapping area. (2) Large portions of this fault section lie within an arid climate area
and the fault zone rocks crop out in many places. Further, some fault sections are
situated within or at the margins of locally uplifted terrains which promote exposure.
(3) Detailed geologic mapping of the fault between Quail Lake and Big Pine is
available by [36]. Thus this is an excellent working platform that includes 1:12,000
geologic strip maps and 1:12,000 aerial photos, both with delineated traces of the
small and large faults of the SAF system. These resources assisted us in locating
exposures of crystalline rocks in the vicinity of the fault. As part of their mapping,
[36] described some of the granitic rock bodies near the SAF as “powdery, crumbly,
microbreccia locally well developed but shearing not everywhere intensive enough
to obscure original igneous textures” and “shattered and crushed to white powder”.
Yet, they did not map the extent of pulverization systematically. The section of the
91
SAF between Big Pine and Cajon Pass is covered by a map [37] of recently active
breaks at a scale of 1:24,000.
Figure 2.1: The south central San Andreas Fault (SAF) system with its major strands. Mapping
of pulverized fault zone rocks in this work covers a 140 km stretch of the SAF in the Mojave,
indicated with a thick gray line. Geographic reference points are indicated: LH: Lake Hughes; P:
Palmdale; PC: Pallet Creek.
We concentrate primarily on mapping pulverized and damaged crystalline
plutonic and metamorphic rocks (referred to below as ‘crystalline’). The number of
exposures associated with crystalline rocks is sufficiently large to indicate a
significant pattern, yet can be covered by a reasonable mapping effort. In addition,
we discuss results associated with a few exposures of fault zone rocks of sedimentary
origin. The crystalline rocks can potentially express conditions and processes from
various crustal depths. This is in contrast to Plio-Quaternary sandstones and other
deposits along the SAF that were never deeply buried. Mapping and damage
characterization of these types of rocks provide important constraints for the depth
generation of pulverization and associated dynamic fields.
PFZR are distinct geologic features with unique appearance, texture and
morphology that enable relatively easy identification in the field. We use the term
92
PFZR for rocks with texture similar to that of the pulverized granite in Tejon Pass
[2]. In general, outcrops of PFZR, including the one in Tejon Pass, share similar
morphology, frequently typified by bad-land topography and high drainage density.
Although the ultimate quantification and classification of pulverization (e.g.
chemistry and mineralogy, particle size distribution, crack orientations) require lab
work, PFZR can be classified in-situ for field mapping purposes according to
macroscopic properties that designate a relative class of damage. We label a given
volume of rock as “pulverized” if all the crystals in a hand sample, including the
quartz crystals, yield a powdery rock-flour texture when pressed by hand, and when
the entire rock volume shows such a texture pervasively. Rocks are identified as
“selectively pulverized” when only some of the crystals yield powdery texture or
only certain rock domains within the exposure are pulverized. Both “pulverized” and
“selectively pulverized” rocks show very little or no shear and their original fabrics
are preserved including crystal shapes, crystal boundaries, and magmatic fabrics.
PFZR may still contain small faults that are insignificant as displacement carriers in
the parent structure. Both types of rocks may show fabric similar to that of the
original undamaged rock, but they can be penetrated and pressed to powder by hand
and hence considered here as PFZR. We further differentiate the classes of
pulverization in section 2.3 that describes higher resolution mesoscale mapping.
Figures 2.2 and 2.3 present mapping results of all the accessible exposures of
crystalline plutonic rocks within ~400 m wide strip centered on the main trace of the
SAF in the Mojave between Quail Lake and Cajon Pass. The map of Figure 2.2
93
includes only rocks that are actually exposed with minimum interpolation of the rock
and damage type between exposures (e.g., below stretches of alluvium). We
minimized interpolation because the pattern of pulverization, which is the essence of
our mapping, can change rapidly over a short distance. If PFZR are exposed with no
lateral width (e.g., in vertical road cuts) they are given a minimum width of about 10
meters. Figure 2.3 simplifies the pattern of Figure 2.2 by showing the location of the
mapped classified outcrops. The colored strips show the dominant types of
crystalline rocks that crop out up to a few km from the fault. The width of the strips
is not to scale and was chosen for graphical reasons.
94
Figure 2.2: Map of crystalline plutonic rocks in the damage zone of the SAF, classified
according to their damage pattern (color scale). Also shown are exposures of damaged to
pulverized sedimentary rocks (purple stars). The map covers a 140 km long section of the fault
between Quail Lake and Cajon Pass. The fault-normal dimension was enlarged to be three times the
fault parallel dimension. The pulverized fault zone rocks (red and orange spots) were found mainly on
the northeast side of the fault. Blue dashed frames show sections for which mesoscale mapping are
presented in Figures 2.4 and 2.5.
The mapped outcrops of PFZR were digitized in a GIS format on a spatially-
referenced 1:24,000 map on the basis of the field mapping. The digitized map
enables a more quantitative description of the distribution of PFZR (e.g., evaluation
of symmetry properties) and it provides an accurate spatial reference for later studies.
2.4a
2.4b
2.5a,b
95
The GIS-based map is included as an electronic supplementary material of the paper.
Errors in the shape, location and calculated area of outcrops of the PFZR in the GIS-
based map may result from two possible sources. (1) Errors during mapping in the
field while projecting the boundaries of outcrops on the paper map. This applies also
to the map of Figure 2.2. (2) Errors during digitization of the outcrops on the raster
background spatially-referenced map based on the pattern in the paper map. These
sources of error are assumed to have negligible influence on the evaluation of the
above parameters due to high resolution of the paper and digital maps (1:12,000 and
1:24,000).
This study aims at presenting observed macroscopic properties of PFZR that are
consistent at many sites along the Mojave section of the SAF. The results provide a
spatial framework for future detailed microstructural and laboratory characterization
of the PFZR. The main features of our observations are described in the following
sections.
Spatial distribution of PFZR in the San Andreas damage zone, Mojave
Distribution along the fault: In the map of Figure 2.2, PFZR are shown with red
and orange colors while less damaged rocks are shown in blue. The results indicate
that the vast majority of crystalline rocks cropping out within 50 to 200 m of the
SAF are pulverized to some degree. This is confirmed by GIS analysis, showing that
93% of the total area of mapped outcrops is covered by pulverized or selectively
pulverized rocks. Outcrops of crystalline rocks outside this zone show no
96
pulverization, and the overall damage decreases rapidly to the background level of
the country rock if no geometrical complexities exist. The observations of Figures
2.2 and 2.3 establish the PFZR as a general structural component of the San Andreas
Fault zone in the Mojave. The PFZR appear to occupy ~100 m wide sub-vertical
tabular zone parallel to the fault. They represent a systematic damage product, which
is generated most likely by earthquakes and hence may serve as a diagnostic of the
rupture process.
Distribution across the fault and symmetry properties: Figures 2.2 and 2.3 show
that along the 140 km mapped stretch of the SAF in the Mojave, outcrops of PFZR
are more abundant on the northeast side of the fault. A GIS-based spatial analysis of
the map indicates that 70% of the total area of all outcrops of PFZR (pulverized and
selectively pulverized) is located on the northeast side of the slipping zone. The
detailed distribution pattern is presented in Table 2.1, showing the cumulative area of
each outcrop class in each side of the fault. This regional distribution pattern may
reflect, at least partially, the current distribution of available exposures (66% of the
total area of mapped outcrops is in the northeast side of the fault), but the outcrops
are distributed unevenly with respect to the fault despite lateral variability in the
contrast in surface lithology across the fault, vegetation cover, slope direction and
other factors. Small and isolated outcrops of PFZR on the southwest side of the fault
appear near Quail Lake and east of Elizabeth Lake. In addition, there is a fairly large
concentration of PFZR on the southwest side of the fault between Cheseboro road
and E 106
th
St. near Little Rock Creek. Some outcrops in the Leona Valley area and
97
west of Big Rock Creek appear between two fault strands. All the other outcrops of
PFZR appear to be northeast of the fault. In contrast, outcrops of fractured rather
than pulverized rocks (blue spots in Figure 2.2) immediately near the fault are more
abundant on the southwest side, especially between Pallet Creek and Cajon Pass.
Table 2.1: Cumulative area of outcrops with different damage levels SW and NE of the fault.
The few outcrops that appear between two fault strands are not included.
Cumulative area NE (m
2
) Cumulative Area SW (m
2
) Damage Level
330666 126271 1 (pervasive pulverization)
25543 27566 2 (selective pulverization)
6409 34683 3 (intense fracturing)
An asymmetric pulverization across the fault may be the result of various
reasons not related to rupture behavior (see related discussions below), and the
apparent asymmetric pattern in Figures 2.2, 2.3 and Table 2.1 may be affected by
asymmetric outcrop availability. Yet, a likely explanation for the results is that the
observed distribution reflects an average shift of the PFZR layer to the northeast side
of the principal slip zone. Moreover, an asymmetric damage generation is expected
to produce asymmetric outcrops availability because pulverization and damage
promote erodibility and exposure. A second-order measure of asymmetry is the
fraction of area covered by PFZR out of the total mapped area in each side of the
fault. Based on the mapping results, these fractions are 98% and 82% on the
northeast and southwest sides of the fault, respectively. Similar sense of asymmetry
in the damage structure was found in the smaller scale observations of [29] at
98
selected sites in our study area and is compatible with our mesoscale mapping
described below.
Although the mapped fault section is relatively straight and does not include
major structural complexities, geometrical complexities at smaller scales may exist
and affect the local faulting environment and damage pattern. The rocks surrounding
the slipping zone may have been subjected to changes in the local conditions during
their history. However, rock bodies on the opposite sides of the fault should have had
an equal chance to be exposed to local geometrical complexities. Despite these (and
other) complexities, the distribution of PFZR is apparently asymmetric across the
fault, while maintaining overall consistent width and pulverization gradient with
respect to the slipping zone. These observations suggest that the regional body of
PFZR was created by many earthquake ruptures over a time scale much longer than
that associated with effects of local transient complexities. These different time
scales allow the asymmetry and other large-scale systematic features to exist despite
the overprint of local conditions.
We exclude marble (green spots in Figure 2.2) from the discussion about
symmetry properties because its damage pattern has not yet been properly
characterized. In general, marble rock bodies along the fault are only
macroscopically fractured and not pulverized, but preliminary microscale
observations in thin sections show that they have very high dislocation density and
twining [39] that can be possibly attributed to brittle faulting (e.g. [40]).
99
Classes of pulverization and fault-zone scale mapping results
In several locations along the fault, we were able to map the damage pattern in
the mesoscale and study the width and gradient of damage with respect to the
principal slip zone. Mapping results from three of those locations are presented in
Figures 2.4 and 2.5. For the fault zone scale mapping, we chose locations in which
PFZR have good exposures from the principal slip zone and outward. We found such
locations only in the northwest portion of our mapping area (see locations in Figures
2.2 and 2.3). For the purpose of this mapping, we differentiated the damage intensity
into five classes. As with the regional mapping, the following definitions of damage
intensity are based on in-situ macroscopic distinct properties. Although the damage
gradient is continuous without discrete transitions, we use these classes to
approximate the spatial variations of damage intensity.
The classes of damage, starting from the lowest are as follows: (I) Weak
fracturing - macroscopic large fractures in density that exceeds the background level
of damage in the country rock. (II) Fragmentation - fractures are in the cm scale and
the rock is fragmented, resulting in the creation of rugged surfaces. Some of the
fragments can be easily crushed by hand into smaller, visible pieces. (III) Intense
fracturing - fractures are in the grain size scale. Although similar in texture to grus,
its mechanical origin is attested by its structural context and location in the
deformation gradient. It is typically characterized by rough, rounded surfaces. (IV)
Weak/selective pulverization - some of the crystals survive and remain intact, some
break along sub-crystal fractures, and some yield powdery texture due to
100
microscopic fractures. The typical appearance in the field often can be similar to that
of the next (highest) damage class. (V) Pervasive pulverization - all crystals yield a
powdery texture when crushed by hand. Associated geomorphic landforms tend
towards smooth, rounded outcrops or bad-land topography with extremely high
drainage density, which is typical of very weak, easily eroded impermeable rock.
Damage classes IV and V correspond to “pulverized” (red) and “selectively
pulverized” (orange) in the map of Figure 2.2, respectively. Damage classes I – III
further differentiate the “intense fracturing” class (blue) of Figure 2.2 to allow a
better description of the damage gradient into the country rock.
Overall, the degree of damage of crystalline fault zone rocks at a point appears
to depend mainly on the distance from and side of the fault. In addition, the damage
level depends to a currently unknown extent on rock properties such as macroscopic
uniformity, isotropy and the mineral content of the rock. For instance, melanocratic
rocks tend to be less pulverized compared to leucocratic rocks with corresponding
locations, the opposite of what would be expected from the result of in situ
weathering. Metamorphic rocks seem to be less susceptible to pulverization,
although we do not have at present independent systematic evidence for this or
indication that it affects the distribution pattern of PFZR. We find on the southwest
side of the fault pervasively pulverized rocks within a large body of gneisses (e.g.
near Elizabeth Lake, Figure 2.3) as well as macroscopically fractured rocks within
granites (e.g. near Three Points, Figures 2.3, 2.4). The influence of the rock type on
the distribution pattern of PFZR should be studied in more detail in future works.
101
Figure 2.3: A large scale distribution pattern of crystalline rocks and their damage pattern
between Tejon Pass and Cajon Pass. Each data point represents an outcrop. The pulverized fault
zone rocks (red and green marks) are abundant along the fault, especially on its northeast side
(circles). Annotated frames mark locations of the detailed mapping of Figures 2.4 and 2.5. The
colored strips generalize the distribution pattern of classes of crystalline rocks that crop-out, often
sporadically, within a few km from the fault (the width of the strips is not to scale). This distribution
follows the mapping of [36] between Quail Lake and the dashed line south of Pallet Creek, and the
mapping of [38] from there to Cajon Pass. The Alphabetic codes within the strips correspond to
further classification of the rock types in the original geological maps.
2.
2.
2.
2.
102
Figure 2.4: A detailed distribution of damaged fault zone rocks in Quail Lake – Sawmill Mtn
Ranch (a) and in Horse Trail Campground (b). The color scale corresponds to the degree of damage
discussed in the text. Note the asymmetric damage pattern with more intense pulverization and wider
zone of pulverized rocks on the northeast side in the two sites. The locations of these sites are shown
in Figures 2.2 and 2.3. For comparison of damage pattern between the two sides of the fault, we re-
juxtaposed in (a) exposures that are currently 7 km apart, reflecting the likely geologic state ~200,000
years ago.
The mesoscale mapping results of Figures 2.4 and 2.5 provide additional
information on the symmetry properties of damaged fault zone rocks with respect to
the principal slip zone of the SAF. Our mapping shows that on the northeast side of
the fault, the zone of pulverization is wider and pulverization is more intense at a
given distance from the fault. This is evident in Figure 2.4a, displaying two road cut
exposures that are currently 7 km apart, one from the Quail Lake area (south) and the
103
other from Sawmill Mtn Ranch area (north). The exposures were re-juxtaposed for
this display, reflecting a geologic situation that corresponds to about ~200,000 years
ago based on the current displacement rate of 3.5 cm/year. Similar or stronger
contrast in pulverization intensity is also shown clearly in Figures 2.4b and 2.5a for
two places where crystalline rocks crop out on both sides of the fault.
Figure 2.5: A detailed distribution of damaged fault zone rocks in a site northwest of Lake
Hughes (see location in Figures 2.2 and 2.3). The colors correspond to the scale in Figure 2.4. (a)
Map (b) Schematic composite cross-section of the damage zone obtained by combining features from
(a) into a vertical plane perpendicular to the fault. The principal slip zone is on the southwest side of a
~60 m wide layer of intensely pulverized rocks. Additional strike slip activity is apparent by
subsidiary gouge zones northeast of the principal slip zone. A ‘fault zone valley’ correlates with the
zone of intense damage.
Figure 2.5b gives a schematic composite cross-section of the damage zone in
which the mapped features of Figure 2.5a were combined into a vertical plane
104
perpendicular to the fault. The resulting ~60 m wide fault zone is composed of the
currently active gouge zone, as well as several other secondary gouge zones (purple
lines) embedded within intensely pulverized rock (red). Although slip on the SAF
could alternate between the different gouge zones, both the strong geomorphic
expression of the currently active fault strand and its coincidence with the major
lithologic boundary attest that this is the locus of long term localization of motion.
As such, the currently active fault strand, which is the principal slip zone, is a
suitable center of reference for evaluation of symmetry properties. The other gouge
zones manifest fault zone damage associated with faulting that is secondary to the
main slip zone.
The composite cross-section of Figure 2.5b has an asymmetric structure in
which most of the fault zone structural elements (i.e. secondary gouge zones) and
pulverization are on the northeast side of the principal slip zone. A larger scale
asymmetry appears in the geomorphic expression of the SAF, with the “fault zone
valley” occupying the more damaged northeast side of the fault zone.
Pulverization of sedimentary rocks
During the course of searching for crystalline rocks along the fault, we found
several exposures of sedimentary rocks that exhibit damage or pulverization that
correspond to degrees III to V as described above. This is supported by preliminary
observations in the microscale of particle size distribution and thin sections. Those
exposures are shown in Figure 2.2 as purple stars and include the following
105
sedimentary units: Juniper Hills conglomerate in several sites, Anaverde sandstone in
the Palmdale area, Hungry Valley sandstone near Three Points and undefined unit of
sandstone (possibly Hungry Valley) near Quail Lake. Tentative pulverization of
sedimentary rocks was observed at other sites. Those damaged sedimentary rocks
were found so far mainly on the northeast side of the fault. Pulverization in
sedimentary units is not as abundant as in crystalline rocks, and we found several
examples of nearly-intact sedimentary rocks within the fault zone at several locations
(e.g. Anaverde sandstone east of Leona Valley). In places where sedimentary and
crystalline pulverized rocks are found in contact, they are often separated by small
local faults.
Discussion
We describe persistent large-scale spatial properties of PFZR in the Mojave
section of the SAF, based on geological mapping at several tens of sites,
supplemented by more detailed mapping at selected sites. The results establish the
PFZR as a general structural property of the SAF in the Mojave. In addition, we find
that the distribution of exposed PFZR is asymmetric across the fault, with more than
twice cumulative area of PFZR cropping out on the northeast side of the slip zone
than on the southwest side.
The fault zone structure at any given location can be affected by numerous
variables and local conditions. Dor et al. [29] discuss various factors that can
produce uncertainties in the interpretation of geological mapping of the type done
106
here. The discussed issues include the general problem of connecting dynamic
rupture behavior at seismogenic depth with surface observations, influence of
contrast in surface lithology on the symmetry properties of the damage pattern,
possible effect of fault dip on asymmetry, complex history of the fault, the influence
of geometrical and compositional complexities, migration of the active slipping
zone, and more. A general way to reduce the above interpretation problems, adopted
in our work, is to perform multi-signal multi-scale mapping of rock damage at
multiple sites along the fault. The characteristics of the fault structure discussed in
this study, combined with the work of [29], have multiple manifestations at different
scales and different sites along the fault. This provides a basis for interpreting the
data in terms of persistent rupture behavior and the large-scale structural properties
of the fault at depth.
Geologic observations related to the possible depth of pulverization
There are several indications that pulverization of the mapped PFZR occurred at
a relatively shallow depth. First, the Punchbowl fault only 5 km southwest from the
outcrops of pulverized rock bodies near the SAF, was interpreted to have been
exhumed from about 2-4 km ([41], and references therein). In the lack of evidence
for a differential vertical movement between the two faults, we can assume that this
is the maximum exhumation depth for the SAF in this area as well. The SAF is
located farther away from the locus of uplift in the San Gabriel Mountains when
compared to the Punchbowl area. Displacement along the current trace of the SAF
107
that is associated with our mapped PFZR occurred during or after the time of activity
of the Punchbowl fault [42]. The observed terrains did not arrive from far away,
where exhumation could be potentially larger, because the current strand of the SAF
has only a few tens of km of displacement in this area (e.g. [36]). Therefore, the
inferred maximum uplift of 2-4 km for the area of the Punchbowl fault on the
mountain side of the SAF provides an upper bound for exhumation of the SAF itself.
Second, at various sites along the Mojave section of the SAF, there are several
lines of direct evidence that the Mojave acted as a relatively stable block with only
minor exhumation throughout the late Miocene to present: near Gorman, Miocene
rocks of the Quail Lake Formation are exposed at elevations that are within a
hundred meters or so of Pliocene rocks of the Hungry Valley Formation and younger
late Quaternary formations, with depositional contact between them (i.e., those
formations were not separated from each other by a vertical component of motion on
dip slip faults). Similarly, middle Quaternary fans near Quail Lake are offset 15 km
along the SAF and directly overlie the Miocene rocks. Elsewhere in the western
Mojave, there are Miocene Lake deposits (e.g., Rainbow basin near Barstow) that
show no significant uplift of the western Mojave since at least middle to late
Miocene time, which is essentially the age of the modern San Andreas fault system
[43]. The Victorville fan has been offset tens of kilometers laterally, but shows no
uplift on the Mojave side in the Quaternary. Considering that some of the strands of
the SAF along which we have identified PFZR are relatively young, we again argue
that vertical motion during the timeframe of their activity must be small. These
108
observations suggest that at least since the middle Miocene, deep exhumation of the
fault zone is unlikely to have occurred, although uplift has resulted in the areas
current elevation of more than 1 km. Other Plio-Quaternary formations along the
SAF, such as the Anaverde and the Juniper Hills formations, were deposited in
narrow basins resulting from small structural complexities (step-overs). It is not clear
how much has been eroded but the consolidation is minor, suggesting again only
minor exhumation. In addition, middle to late Quaternary fans are preserved along
the fault, indicating only minor incision in late Quaternary time.
Third, we have found that several sandstone and conglomerate units that were
never deeply buried are damaged or perhaps even pulverized at levels that exceed the
background damage level. This general observation puts the upper bound for damage
generation and maybe pulverization to a possible depth of several hundreds of
meters, although further characterization of the damage to sedimentary units along
the SAF is needed in order to provide more accurate constraints for the pulverization
process.
Forth, additional direct evidence that pulverization occurs at shallow depths is
provided by studies [44, 2] of pulverized quartzite in fresh rupture zones in South
African gold mines. They analyzed rock-powder samples collected from the rupture
zone of M=3.7 earthquake at depth of about 2 km (and near large internal free
surfaces), and found that the grain size distribution and other properties of those
samples is similar to those found in Tejon Pass. Pulverized quartzites are abundant in
109
many fresh rupture zones in the mines [45], sometimes forming under minimal or
even negligible confining stress.
Wilson [46] argued that pulverization at Tejon Pass could have potentially
occurred at a depth of 4-9 km based on exhumation rates in the Transverse Ranges.
Even if the estimated exhumation is correct, the observed pulverization could have
occurred during or after the exhumation. Moreover, the exhumation estimates have
large uncertainties. Ref. [42] have shown that crystalline rock domains like the
Pelona/Orocopia schists and possibly source rocks to our mapped PFZR could have
been exhumed from a great depth during the Clemens-Well, Fenner, San-
Francisquito phase of the SAF (20-17 Ma and 13-12 Ma). Displacement and possibly
some exhumation continued since 4-5 Ma in the same zone along the modern SAF.
However, according to the reconstruction of [42], the modern trace of SAF, which
has well defined and unique structural association with our mapped PFZR, may have
re-occupied the same region that was active during that early phase of faulting, but
not the exact same ancient trace that is delineated now by the (rotated and inactive)
Clemens-Well, Fenner and San-Francisquito faults. The tight structural association
of the PFZR with the modern trace of the SAF implies that the pulverization we
observe occurred during the modern activity of the SAF, after the main phase of the
exhumation.
110
Correlation of mapping results with geophysical observations
A number of recent studies concluded that the low velocity seismic waveguides
that produce trapped waves in the San Andreas, Landers, San Jacinto, and North
Anatolian fault zones are generally limited to the upper ~3-4 km of the crust [47-50,
34]. The seismic trapping structures have a width on the order of 100 m, similar to
the lateral dimensions of the mapped body of PFZR, and are associated with strong
reduction of seismic velocities (e.g., 30-50%) and strong attenuation (e.g., Q values
of shear waves less than 10). The seismically imaged depth extent of the trapping
structures correlates with the maximum inferred pulverization depth of section 3.1,
and the PFZR should have significant reduction of seismic velocities and significant
increase of seismic attenuation due to the strong reduction of grain size and
associated hydrological and chemical effects. In addition, recent imaging studies
based on trapped and head waves indicate that the shallow damaged layers are
asymmetric across the faults [34-35]. We thus suggest that the observed PFZR are
the surface expression of the low velocity fault zone layers that act as seismic
waveguides.
Relations of the mapping results to theoretical predictions
The observed coherent layer of PFZR with relatively shallow inferred
pulverization depth and apparent asymmetric structure with respect to the principal
slip zone requires a generation mechanism. This mechanism should depend on large
scale properties (rather than local site conditions), and should generate strong
111
tensional stress field in the vicinity of the fault that can produce asymmetric long-
term damage pattern in the upper few km of the crust. A curvature along the entire
Mojave section of the SAF can produce asymmetric static stress field [51]. However,
this is unlikely to produce the extreme grain reduction of the pulverized rocks (e.g.,
[2], this study) and the highly localized asymmetric pattern of gouge fabric found by
[29]. As outlined in the introduction, mode II rupture along a material interface (e.g.,
[4, 10-18]) provides a specific set of predictions that are compatible with our
observations.
The parameter-space study of [27] indicates that dynamic generation of rock
damage during such ruptures, under realistic conditions of velocity contrast and
material properties, is limited to the shallow portion of the crust. The events
propagate preferentially in the direction of slip on the side with lower seismic
velocities at seismogenic depth, while damage is generated in the top few km on the
side with faster seismic velocity, which is persistently on the tensional quadrant of
the radiated seismic field for the preferred propagation direction.
The available seismic imaging studies in our area indicate [31-33] that the
northeast side of the Mojave section of the SAF has higher seismic velocities than
the southwest side. The mapped distribution of PFZR in Figures 2.2-2.5 and the
finer-scale observations of [29] show more damage on the higher velocity block.
While the completeness of the mapping is affected by sites availability, the multiple
observed manifestations of damage asymmetry across the SAF are compatible with
the predictions for wrinkle-like ruptures along a material interface. The tensional
112
stress produced by wrinkle-like ruptures may be aided by dynamic reduction of
normal stress associated with the other mechanisms mentioned in the introduction,
although this can not be tested at present in the absence of explicit predictions.
However, none of the other mechanisms explains the observed correlation between
the damage asymmetry and the velocity structure of the fault, and the existence of a
preferred or more vigorous rupture direction which is necessary for the production of
long-term asymmetric damage structure.
Comparison between the structures of the active SAF and exhumed faults of the
SAF system
The general structure that we observe for the SAF in the Mojave includes a very
narrow zone on the order of a few cm to a meter or so in which slip is localized,
surrounded by a zone of strong damage on a scale of several tens of meters without
significant shear parallel to the principal slip zone (in places we find several
secondary or previously active slip zones). The deformation decreases gradually as a
function of distance from the principal slip zone. This structure matches general
characteristics of ancestral exhumed faults of the SAF system such as the Punchbowl
and San Gabriel Faults ([52-54] and references therein).
In addition to the presented observations associated with the SAF, we have also
recently observed a similar, tens-of-meters-wide body of pulverized granite along the
southern side of the Garlock fault on Tejon Ranch, with pulverization decreasing
away from the active trace of the fault. PFZR can be seen in meters to tens of meters
113
wide exposures southwest of the Little Rock Fault, west of Tierra Subida Rd. and
north of the Anaverde Creek near Palmdale. Farther to the east exposures of similar
scale appear between strands of the Little Rock fault and the SAF along Pallet Creek
Rd. We also found PFZR in ~20 cm layer adjacent to the ultracataclasite of the
Punchbowl fault on the sandstone side where it is exposed near the top of the trail
that climbs up from the Southfork Campground. Several outcrops of pulverized
granodiorite, ranging in width from several tens of cm to several meters appear
adjacent to the traces of the Southern and Northern Nadeau faults, where Mt. Emma
Rd. crosses Little Rock Creek, and on Cheseboro Rd. about 1.2 km north of the
junction with Mt. Emma Rd. (some outcrops of PFZR there may be related to the
nearby trace of the Punchbowl fault). A layer of pulverized granodiorite on a similar
scale bounds the northeast side of the fault core of an ancestral strand of the SAF,
300 m south of the active SAF east of where it crosses Little Rock Creek (near
“Little Rock Creek Site” of [29]). The pulverization found in these faults is on a
much smaller scale with respect to other fault zones and may reflect either different
pulverization conditions or different pulverization process; it may also be related to
the amount of displacement under ‘pulverization-generation conditions’.
On the other hand, we found a clear absence of pulverized rocks in the structure
of the San Gabriel fault near the "Earthquake Fault site" of [55], and in several other
sites along Big Tujunga Canyon Road and Angeles Crest Highway. In the San
Gabriel fault and other cases of uplifted structures that have no PFZR, it is possible
that the exhumation was to such a depth that the zone of PFZR has been eroded and
114
is not preserved (if it indeed formed at depths shallower than the exhumation depth).
Outcrops of granitic rocks adjacent to the active trace of the Elsinore fault near the
junction of Hwy 79 and Hwy 76 appear to be almost intact. This is despite the
presence of a near by large step-over that would have promote tensional stresses
(Lake Henshaw). Whether the absence of PFZR from the exposed structure of a fault
implies that no mechanism with sufficiently strong dynamic reduction of normal
stress operated during its activity is a question we cannot answer at this point.
Summary
The observations made in this paper, combined with previous geologic and
geophysical observations, suggest that PFZR along the Mojave section of the SAF
consist of a ~100 m wide layer parallel to the principal slip zone of the fault. We
infer that the observed pulverization occurred at shallow depths based on several
lines of geologic observations and theory. Our results and the smaller-scale
observations of [29] suggest further that the layer of PFZR is not centered on the
principal slip zone, but is shifted on average to the northeast side of the fault. This
apparent damage asymmetry is compatible with a preferred or more vigorous
northwest propagation direction of earthquakes on the Mojave section of the SAF,
which would have a damage-promoting tensional quadrant on the northeast side.
This is also the side of the fault that has faster seismic velocities at seismogenic
depth based on available imaging studies in the area [31-33]. A northwest preferred
115
rupture direction is opposite the currently inferred rupture direction [56] for the 1857
Ft. Tejon earthquake. See a related discussion on this issue at [29].
The correlation between the apparent damage asymmetry and the side of the
fault with faster seismic velocities at seismogenic depth, together with the inferred
shallow pulverization depth, are consistent with theoretical predictions associated
with wrinkle-like ruptures along a material interface [27]. In addition to explaining
our observations, such a mode of rupture can be relevant to various important aspects
of the mechanics and structure of large faults, including the general lack of melting
products; suppression of branching, short rise-time of earthquake slip, effective
constitutive laws, and expected seismic shaking hazard [4, 15]. The SAF in the
Mojave has several structural properties in common with ancestral exhumed faults of
its system, although PFZR do not have an equal presence in all of those faults. The
partial similarity between the structure of the SAF that is exposed at the surface and
those exhumed structures lends support to our inferences on seismogenic processes
along the SAF.
Acknowledgments
We thank Ze’ev Reches, Tom Dewers and Brent Wilson for discussions during
joint field trips. The study was funded by the National Science Foundation (grant
EAR- 0409048) and the Southern California Earthquake Center (based on NSF
cooperative agreement EAR-8920136 and United States Geological Survey
cooperative agreement 14-08-0001-A0899). The manuscript benefited from
116
constructive reviews by Matthew d'Alessio, anonymous referee and Editor Rob van
der Hilst.
117
Chapter 3: Geologic and geomorphologic asymmetry across the rupture
zones of the 1943 and 1944 earthquakes on the North
Anatolian Fault: possible signals for preferred earthquake
propagation direction.
Submitted to Geophysical Journal International
Co-authors: Cengiz Yildirim
1
, Thomas K. Rockwell
2
, Yehuda Ben-Zion
3
, Omer
Emre
1
, Matthew Sisk
2
, Tamer Y. Duman
1
1. General Directorate of Mineral Research and Exploration (MTA),
Ankara, Turkey.
2. Department of Geological Sciences, San Diego State University, San
Diego, CA 92182-1020, USA.
3. Department of Earth Sciences, University of Southern California Los
Angeles, CA, 90089-0740, USA
Abstract
The east and west rupture directions of the 1943 and 1944 earthquakes on the
North Anatolian Fault (NAF) are hypothesized to represent, respectively, long term
preferred propagation directions on the corresponding sections of the NAF. Fault
sections with preferred rupture direction are expected to have an asymmetric damage
structure with respect to the slipping zone. To test the above hypothesis, we study
geologic and geomorphologic manifestations of structural asymmetry with respect to
118
the active trace of the NAF along the 1943 and 1944 sections. The following fault
zone elements are mapped: gouge fabric in the cm scale, fault core structure in a
meter scale, and secondary faults and fault rocks in tens of meters scale. Mapping
results at three sites on the 1943 rupture and one site on the 1944 rupture are
consistent with accumulation of more rock damage on the south side of the 1943
section and on the north side of the 1944 section. Erosion patterns adjacent to the
fault that are not correlated with the distribution of intrinsic and extrinsic erosion-
controlling variables are interpreted as morphologic responses to the damage content
of rocks and its impact on rock erodibility. The valleys of 11 rivers are parallel to the
studied fault sections. About 75% of the total river valleys length along the 1943
rupture is on the south side of the fault, and about 89% of the total length along the
1944 rupture is on the north side of the fault. Morphometric analysis of watersheds in
two correlative terrains displaced along the 1944 rupture section shows that stream
erosion is considerably more intense in the terrain north of the fault with drainage
density values almost double in the north compare to the south. Bad-land topography
at two sites along the 1943 rupture section is substantially more developed at the
~100 m scale on the south side of fault. Our observations along the 1943-1944
rupture sections, including various types of signals that span a large range of scales,
are systematically compatible with an opposite sense of damage asymmetry between
the two fault sections. These observations are consistent with preferred direction of
ruptures that are opposite for each segment, similar to the propagation directions of
these two recent earthquakes. If those rupture directions are dictated by the velocity
119
structure at depth, we infer that the south side of the 1943 rupture has slower seismic
velocity at seismogenic depth than the north side, and that the sense of velocity
contrast is revered along the 1944 rupture zone.
Introduction
During the past century, the North Anatolian Fault (NAF) has experienced a
sequence of large ruptures that un-zipped nearly the entire fault from east to west
(Figure 3.1). Stein et al. (1997) attributed this rupture sequence to successive
Coulomb failures of fault segments due to loading induced by the previous ruptures,
akin to a series of falling dominoes. Surprisingly, the 1943 earthquake did not
nucleate in the region of maximum stress increase but rather at the opposite end of
the final rupture, far to the west, and then ruptured eastward. The following 1944
event nucleated at the far west end of the 1943 earthquake rupture and propagated
farther to the west. Similarly, the 1999 August earthquake nucleated near the head of
Izmit Bay, away from the region of maximum stress increase, and ruptured primarily
eastward toward the earlier failures. This was followed by the continuation of rupture
to the east in the November 1999 Düzce earthquake. The “unexpected” eastward
propagation direction of the three large 20
th
century earthquakes on the NAF, against
the overall un-zipping direction to the west, may reflect a preferred propagation
direction of ruptures on those fault sections.
120
Figure 3.1: Ruptures along the North Anatolian Fault in the 20th Century. Upper panel: the
North Anatolian Fault, Turkey with 20
th
century earthquakes (colored sections) and their rupture
directions (arrows) delineated. Locations of geological mapping sites are indicated with circles.
Location of geomorphological mapping sites is shown in the corresponding figures. Middle panel
shows the stress transfer configuration prior to the 1943 rupture (from Stein et al., 1997). Lower panel
shows setting of experiments by Anooshehpoor and Brune (1999) analogous to the loading and
rupture configuration of the 1943 earthquake; see associated text in the discussion.
Analytical and numerical results indicate that rupture along a bimaterial
interface that separates different elastic solids evolves, for broad ranges of frictional
and material contrast values, to a wrinkle-like rupture with a preferred propagation
direction (e.g., Weertman, 1980; Ben-Zion and Andrews, 1998; Cochard and Rice,
2000; Shi and Ben-Zion, 2006; Brietzke and Ben-Zion, 2006; Dalguer and Day,
2006; Ampuero and Ben-Zion, 2007). Such ruptures produce dynamic dilation at the
tip that propagates in the direction of slip on the more compliant side of the fault,
referred to as the preferred direction, and dynamic compression at the tip propagating
in the opposite direction. A wrinkle-like mode of rupture with a preferred
121
propagation direction can have fundamental consequences for many topics ranging
from long-standing geophysical paradoxes to effective constitutive laws and
expected seismic shaking hazard (e.g., Ben-Zion, 2001).
Figure 3.2: Rupture radiation field and damage generation about a biomaterial interface. a.
Schematic illustration of the seismic radiation field of a propagating rupture (thick bar in the middle).
Note that the radiation is much stronger in the direction of rupture propagation. More damage is
expected on the bottom where the primary tensional quadrant is. b. Plastic strain (black & white scale)
generated by a wrinkle-like rupture pulse propagating to the right on a fault separating a compliant
material (top) from stiffer material (bottom). The stiffer side of the fault is in the tensional quadrant of
the radiation pattern for the preferred propagation direction of the wrinkle-like pulse. Schematic
illustration of Figure 4b of Ben-Zion and Shi (2005).
During earthquake ruptures, rock damage is generated primarily in the tensional
quadrants (Figure 3.2a) of the radiated seismic waves (e.g., Dalguer et al., 2003;
Andrews, 2005; Ben-Zion and Shi, 2005; Rice et al., 2005). If earthquakes on a
given fault section do not have a preferred propagation direction, the cumulative
effect of many ruptures should produce an approximately symmetric damage pattern
across the fault. On the other hand, the cumulative effect of many earthquakes with a
preferred propagation direction should lead to asymmetric fault zone damage. Ben-
Zion and Shi (2005) showed that for subshear wrinkle-like ruptures along a
bimaterial interface having 45º to the maximum compressive stress, significantly
122
more damage is generated on the stiffer side of the fault, which is in the tensional
quadrant of the radiated seismic field for the preferred propagation direction (Figure
3.2b). If the opposite rupture propagation directions of the 1943 and 1944
earthquakes on the NAF reflect persistent preferred propagation directions associated
with the fault zone structures, we would expect strongly asymmetric rock damage
along those fault sections with an opposite sense of asymmetry. This hypothesis can
be tested with detailed in-situ observations of rock damage, as was done by Dor et al.
(2006a, 2006b) for several faults of the San Andreas system in southern California.
To test the above hypothesis, we performed multi-scale, multi-signal
characterization of rock damage along the 1943 and 1944 rupture sections on the
NAF. The studies employ detailed geological mapping at several natural exposures
and trenches across the fault, and analyses of geomorphologic patterns at various
sites along the fault. In section 2 of the paper, we outline the approach and
methodologies used in this study and present results of the various analysis methods.
In section 3 we synthesize the results, evaluate the likelihood of the existence of
structural asymmetry across the studied fault sections and discuss their implications
for the propagation direction and mechanics of ruptures. The combined set of
geologic and geomorphologic mapping results indicate damage asymmetries in the
structure of the 1943-44 rupture sections that are compatible with the opposite
propagation direction of these two earthquakes.
123
Observations
Approach and methodology
We chose the 1943 and 1944 rupture sections (275 and 183 km long,
respectively) for our study because they are long enough to accommodate large
earthquakes, their geometry is relatively simple and because these two recent
ruptures propagated in opposite directions (Figure 3.1), providing a good test case
for our hypothesis.
Geologic mapping: Systematic reconnaissance of the 1943-44 ruptures revealed
several potential sites for detailed geologic mapping. Sites with broad exposure of
bed-rock on both sides of the principal slip zone of the fault are not easily found
because the fault trace is frequently obscured by debris and landslides generated
during recent fault activity, and by thick vegetation cover. Nevertheless, we found
four sites with sufficiently good outcrop quality, where several fault zone elements
are exposed or could be exposed by trenching. For most of the sites, observations can
be made at more than one scale. Along the 1943 rupture we worked in Celtikci,
Bademci and Ladik which are 69, 91 and 244 km east of the 1943 rupture nucleation
zone, respectively. Along the 1944 rupture we found one site suitable for mapping
rock damage across the fault, Hamamli, which is 155 km west of the rupture’s
nucleation zone.
We discuss geologic observations in the centimeter to meter scale, focusing on
the structure of the Principal Slip Surface (PSS) and its immediate surroundings
124
within the fault core, and in the meters to 10s of meters fault zone scale. Here the
fault core denotes the primary zone that accommodates the fault displacement and
hence contains significant concentration of shear fabric, distorted original structures
and translocated material, while the fault zone denotes the structural component that
contains damage in the form of secondary faults, fractures and variety of fault zone
rocks (Chester et al., 1993; Schulz and Evans, 2000). Symmetry properties are
measured and described with respect to the currently or most recently active slip
surface that could be discerned at the fault core scale, and with respect to the
currently active slipping zone in the fault zone scale. In both scales we refer to the
symmetry reference zone as PSS. Dor et al. (2006a) have an extended description of
fault zone elements, scales of observations and frames of references relevant for a
study of this type in their methodology section, and we adopt their definitions when
discussing similar topics. Specifically, the use of the term ‘gouge’ follows the
definition of Scholz (2002) for foliated gouge, modified after Sibson (1977), unless
otherwise stated and special additional properties are described.
The described work involves mapping of excavated trench walls as well as of
road cut and natural exposures. Our methods of cleaning and preparing the
exposures, identifying the PSS, resolving the relative activity time of gouge and fault
zone elements, documentation and data collection and processing follows procedures
used by Dor et al. (2006a).
Geomorphological analysis. The intensity of erosion at one location may be
affected by numerous intrinsic and extrinsic variables including climate, lithology
125
and soil type, original slope structure, aspect, relief, base level, land use, basin
development stage and more. In cases where those variables are very similar between
different terrains but the intensity of erosion (geomorphic work) is significantly
different, additional variables should be considered. We examine various expressions
of erosion in the vicinity of the NAF along the 1943-44 rupture zones while
attempting to identify the influence of rock damage as a factor affecting spatial
variations of erosion intensity. Increased erodibility should correlate with elevated
levels of damage in the underlying rocks because high damage content should make
the rocks more susceptible to erosion. This correlation can be made when carefully
considering systematic patterns that cannot be explained by other erosion-related
variables. Geomorphic patterns that can be interpreted as being affected by fault-
related rock damage can illuminate aspects of the structure of the damage zone of the
fault and related properties of earthquake ruptures.
We observe various expressions for geomorphic work (erosion) in three scales
using different methods: the observations in the largest scale concern the adjustments
of the river-normal profile of major rivers to the location of the current active trace
of the fault along the entire length of the 1943-44 ruptures (i.e. whether fault-parallel
river valleys situated mostly north or mostly south of the fault along the rupture
length); medium scale observations that include quantitative comparison of
morphometric parameters of drainage systems within correlative rock bodies on the
two sides of the fault along the 1944 rupture section; and small scale observations at
126
two sites along the 1943 rupture zone regarding the development of badland
topography within a ~100 m wide zone adjacent to the active trace of the fault.
The geomorphic observations that span the largest spatial extent are based on
mapping results by Herece and Akay (2003), Kondo et al. (2005) and Emre et al.
(2005) who mapped the exact location of the active fault trace using 1:10,000 aerial
photos, 1:25,000 topographic maps, and field surveys. We projected their mapped
fault trace on a Digital Elevation Model (DEM) and found that out of the combined
length of 458 km of the 1943-44 ruptures, 190 km of rupture length are parallel to
rivers and situated within their valleys. In all cases, the rivers deviate from their
original flow direction in their fault-parallel section, often due to tectonic deflection.
For each of the fault sections situated within a river valley, we examined the
consistency of the relative location of the fault with respect to the valley’s deepest
part (where the river flows). Erosion along river valleys is assumed to represent the
largest scale geomorphic response to the fault damage structure (see discussion
section 3.2). To reduce interpretation problems that may be associated with an
individual fault-river section, we checked the entire length of all the river valleys
associated with the two ruptures. If the deepest parts of the river valleys are situated
repeatedly on the same side of the fault along the entire rupture length, we infer that
the rupture has a large scale asymmetric erosion pattern. If the lithological contrast
and other variables along the rupture cannot be correlated with this pattern, an
underlying physical mechanism producing large scale asymmetric damage structure
should be seriously considered as controlling the observed erosion pattern.
127
Another type of geomorphic observation spans a medium size area and includes
GIS-based morphometrical analysis of watersheds. The quantitative linear, relief,
aerial and network properties of drainage basins adjacent to the fault are analyzed
and compared between correlative terrains across the fault. This analysis provides a
basis for comparison of erosion patterns between the terrains, which in turn can
express the distribution pattern of damage. If two terrains near the fault are similar in
their lithology and lithological variations, geological structure, elevation, relief,
slope, aspect, base level, climate, vegetation, land-use and other variables that may
affect erosion, earthquake-related damage can be considered as a reason for
differences in rock resistance and erosion intensity as reflected in the drainage
density and other related parameters. We found that only one pair of terrains is
similar enough to be considered for this analysis: an Eocene andesitic-basaltic cover
unit (Herece and Akay, 2003), offset by the NAF along the 1944 rupture zone in the
vicinity of the village of Ismetpasa. A comparison of various controlling variables
between the two terrains is presented in the results section and summarized in Table
3.
For our analysis, the full drainage structure of watersheds in the two terrains,
including stream order and length of stream segments, must be known. We first used
the 3D analyst extension of ESRI ArcView to generate Digital Elevation Models
(DEM) from 1:25,000 scale digital topographic maps. Eleven basins in the northern
terrain and 16 basins in the southern terrain were identified and their DEMs were
used to derive topographical features and delineate stream networks and watershed
128
characteristics (Moore et al., 1992; Nogami, 1995; Luo & Stepinski 2006). An
automated way of digitizing the drainage pattern using a DEM and generating a
Strahler stream order (Strahler, 1952) is by using the Arc Hydro extension of ESRI
ArcGIS. The stream order generator assigns a numeric order to segments of a grid,
which represents branches of a linear network.
After generating the digitized drainage pattern for each of the basins in the two
studied terrains, we analyzed the following aspects for each basin (Table 3.1):
Linear characteristics:
Bifurcation ratio (R
b
): a dimensionless ratio of the number of streams of any
given order to the number in the next lower order (Horton, 1945; Strahler, 1952), R
b
= (N
u
)/(N
u+1
) where N
u
is the number of streams in order u.
Aerial Characteristics:
Hypsometric integral (H
i
): an integral of the dimensionless area-altitude
distribution hypsometric curve. Its values range between 25 to 75%, with higher
values indicating less eroded surfaces (Strahler, 1952).
Network Characteristics:
Drainage density (D
d
): stream channel length per unit area (Horton, 1945;
Strahler, 1952; Mekel, 1977), D
d
= ∑ L / A, where L is the length of a channel and A
is the total area of the basin. D
d
reflects a balance between erosive forces and
resistance of the surface to erosion, and is therefore closely affected by lithology,
climate, vegetation and rock fracture density; it is probably the most important
129
geometrical property of a stream network, with higher values where erosion is more
intense (Chorley et al., 1984; Hugget and Cheesman, 2002).
Stream frequency: (F): defines the number of stream segments of all orders per
unit area (Horton, 1945, Strahler, 1952), F = N / A, where N is the number of
streams in basin A. Higher values are expected on impermeable bedrock, sparse
vegetation, and in areas of high relief. This variable is strongly related to drainage
density.
Relief Characteristics:
Relief (H): The elevation difference in the basin is a controlling variable that
together with basin area has an influence on slope and stream gradient, and hence on
the intensity of erosion in the basin.
Ruggedness number ( N
r
): this is a dimensionless number related to slope
gradient (steepness) and drainage density, N
r
= H * D
d
where H is the relief (Mekel,
1977; Rengers, 1981). This variables discriminates basins with higher channel
incision from those with lower channel incision. Basins with higher ruggedness
numbers are more susceptible to erosion (Gangalakunta et al., 2004).
These parameters provide quantitative measures of the intensity of erosion in
each basin. We compare the average values of these parameters between the two
terrains and discuss the symmetry properties of the erosion pattern and its possible
relations to the damage structure.
Drainage density is an intrinsic property of a basin, considered as the prime
indicator for the erodibility of the rocks (Ritter et al., 2002). Bifurcation ratio and
130
stream frequency are closely related to drainage density, and ruggedness number is a
direct function of it together with the relief.
Table 3.1: Definitions of morphometric parameters.
No Morphometric
Parameter
Formula Descriptions
1 Bifurcation Ratio (R
b
) R
b
=Nu/(Nu+1) A dimensionless ratio of the number of streams of
any given order to the number in the next lower
order
2 Relief (H) H=h
max
-h
min
Elevation difference between the highest and
lowest points in the basin
3 Ruggednes Number(R
n
) R
n
=HxD
d
The product of basin relief and drainage density
4 Drainage Density (D
d
) D
d
=L/A Total length of stream channels per unit area
5 Stream Frequency (F) F=N/A Number of stream segments per unit area
6 Hypsometric Integral
(H
i
)
H
i
=H
mean
-H
min
H
max
-H
min
The area under the hypsometric curve as a
percentage of total area
The geomorphic observations that span the smallest area describe the
development of ~100s of meters scale gully systems in two exposures of damaged
basement rocks near the villages of Çeltikçi and Tekeler along the 1943 rupture
section. We characterize the gully systems by deliniation of basins and channels on
high resolution digital photos taken normal to the two exposures, and discuss the
properties of the gully systems on the two sides of the fault.Deatiled descreptions of
sites and observational results are presented below.
Geologic observations
Ladik: This site is located about 31 km west of the eastern termination zone of
the 1943 rupture and 3 km east of Ladik lake, a 2 km wide releasing bend of the
NAF. Mapping of the geomorphological expression of the fault along a one km-long
section in this area is presented in Figure 3.3. The location of the active fault is
131
attested by an assortment of markers such as deflected streams and vegetation lines,
small depressions, displaced terraces, linear gullies and scarps. The arrangement of
those features along a remarkably straight line suggests a long term localization of
fault activity along the current active trace, and a near planar fault geometry. The
local fault strike in this location, as expressed by its geomorphic trace is 292°,
parallel to the large scale trend of the NAF in this area.
Figure 3.3: Geomorphologic map of the area around the trenching site near Ladik. The active
trace of the fault (red line) has a remarkably straight expression attested by deflected features, sag
ponds and ridges. Fault rocks and damaged country rocks are exposed in two channels and in a trench
excavated at the bottom of a construction ditch (in the inset).
The expression of the active fault trace correlates with three exposures of fault
gouge, two of which cropping out naturally along the banks of streams and the third
exposed in a trench. The long-term lithological contact of the NAF in this area is
132
between late Cretaceous limestone-sandstone-shale rocks on the northern block and
lower-middle Jurassic sandstone and shale on the southern block (MTA, 1:50,000
Geological map, 2002). This contact must be south of the currently active fault, as
we find rocks of the northern block south of the active fault in the easternmost
channel that cut through our map area (Figure 3.3). Two inactive gouge zones were
found about 60 m south of the currently active fault. It is possible that fault activity
has shifted between different strands before the localization of contemporary activity
in its current position. These gouge zones may also be interpreted as representing
contemporary secondary activity to the one occurring on the main fault, although
they have no clear geomorphic expression.
In addition to the local simplicity of the fault structure, another major advantage
of this site is convenient access to fault zone rocks that until recently were buried.
We dug a trench at the bottom of a 5-12 m deep and ~70 m wide ditch, which was
excavated recently through the hill that contains the fault as part of a highway
construction effort (inset in Figure 3.3). The exposure in the trench is therefore up to
15 meters below the recent original ground surface and the rocks are relatively
undisturbed by potential free surface influence on rock fabric, such as thermal
cycling and weathering. The trench is 60 m long and trends at an angle of 52° with
respect to the fault (its direction was dictated by the construction ditch). Hence, the
trench exposes a fault normal distance of about 47 m. This is also the minimum
width of the fault zone since structural elements related to the fault may exist outside
133
(mainly south) of the trench boundaries. Measurements indicated below are given in
distances along the trench wall.
Figure 3.4 displays a trench log showing bedrock units, fault zone rocks and
fault zone structural elements found within the trench exposure. Although the trench
is centered on the current slipping zone, we found that only the bedrock that belongs
to the northern block and includes late Mesozoic red and yellow beds of shales
appears in the trench (represented in green). Those strata are finely layered and
inclined southward. They appear north and south of the currently active fault core
and display minor bedding-plain slip in a few places, mainly in the northern section
of the trench. Fault zone rocks and related material that were found in the trench
include gouge, colluvium and sag pond deposits (represented in Figure 3.4 by pink,
gray and brown, respectively). Fault zone structural elements include the fault core
with its various layers, and secondary faults (marked with red lines) typically having
a thin layer of gouge. The locations of kinematic indicators such as slickensides are
shown as color-coded dots.
134
Figure 3.4: Photomosaic and corresponding maps of a trench wall from Ladik. Northern part is
shown in the upper panels and southern part is shown in the lower panels. Lower hemisphere equal
area stereographic projections show orientation and rake of striations (green dots) of individual faults.
Fault rocks and secondary faults are more abundant on the south side of the fault core (Figure 3.5).
The colluvium is matrix supported, with a silty gray-green matrix and poorly
sorted angular fragments ranging in size from 0.5 to 10 cm. Several unusual
fragments have an exceptional size of up to 0.5 m. The colluvium appears in slivers
bounded locally by small faults. One such 6 m wide sliver appears 5 m north of the
fault core, and three other slivers, 1-3 m in width, appear south of the fault core. The
width of the southernmost body of colluvium is unknown since its exposure is
truncated by the southern end of the trench.
The sag pond deposits contain soft clayey to silty material. They are finely
layered with beds distinguishable due to variations in colors (rather than variations in
texture, grain size or else). The layers are black, brown, red, green, yellow and gray.
They are mainly horizontal but are often disrupted and mixed. The clasts that are
135
embedded in the sag pond material are similar to those that appear in the colluvium,
but they are sparser. Sag pond deposits appear solely on the south side of the fault
core. They are continuous with a 0.7 m wide sliver of rock that disrupts its continuity
and is bounded by faults on both sides. The sag pond deposits have a stratigraphic
contact on their northern side and a normal fault contact on their southern side; both
contacts are with colluvium.
Structural elements exposed in the trench include the fault core with one major
gouge zone and a few other layers, 30-40 secondary faults that span the entire length
of the trench, and many other smaller fractures. The secondary faults are evident
either as discontinuities separating displaced layers or other markers, due to the
existence of gouge with shear fabric, or both. Almost all faults are inclined to the
south, and most of them are expressed as nearly straight lines on the exposure face,
suggesting that they are relatively planar. The southernmost fault that separates sag
pond deposits from colluvium has an exceptionally shallow dip compare to the other
exposed faults. The gouge layers in the secondary faults are 2 mm to 5 cm wide and
their typical color is reddish-brown to black, probably due to the existence of organic
material and oxides in addition to ground-up and altered country rock. The gouge
material is claylike with small amounts of visible porphyroclasts. In most secondary
faults the gouge is foliated and contains delicate shear indicators such as small,
parallel facoidal surfaces displaying gentle striations. Most secondary faults contain
gouge material that is soft and flexible, flaky, highly incohesive and thus non-
136
lithified, likely reflecting the participation of those faults in recent displacement
events.
Some of the faults and other fractures were only recognized days after the
beginning of the work when cracks gradually opened on the desiccating trench wall
(light blue lines in Figure 3.4). Many of those cracks followed natural parting
surfaces that preexisted in the rock and frequently showed slickensides. They
developed especially in areas with a large content of expansive clay (that swallows
and shrinks as a function of the water content). They were therefore more frequent in
zones that accommodated shear, in particular in the principal gouge zone within the
fault core.
We identified several tens of small surfaces that contain slip striations within the
gouge material of the secondary faults. In almost all cases the striations were
horizontal or sub-horizontal. Oblique and dip slip indicators were sparse, suggesting
that most of the secondary faults within this fault zone are strike slip and hence
related to activity of the NAF. This inference about the faults is supported by their
steep inclination, which is typical of strike slip faults. We found dip slip striations on
the low angle fault at the southern end of the trench, indicating a normal sense of
motion which is compatible with a sag pond environment. Some striations with a dip
slip component of motion were found on branches and parts of secondary strike slip
faults in places were those faults bend, locally forcing a component of dip slip. In a
few places, we were able to measure the rake of the striations. Some of those
measurements appear in Figure 3.4 as stereographic lower hemisphere equal area
137
projections, emphasizing the dominancy of strike slip sense of motion along the fault
zone.
The distribution of the secondary faults within the exposed fault zone is
asymmetric with respect to the center of activity in the fault core (the PSS – see
below): about three quarters of those faults were found south of this reference zone.
The active fault core, shown in Figure 3.5, is 1.8 meters wide and includes the
following layers (from north to south): a 70 cm wide layer of gouge, a 40 cm wide
section of shale, a 50 cm wide layer of colluvium and an additional 20 cm wide
section of shale. The entire fault core dips 51° toward the SSW (200°). The
geomorphic expression of the active fault can be correlated clearly to the location of
the fault core, and in particular to its northern side where the gouge layer is. The
layers in the fault core contain significant foliated shear fabric, considerably
exceeding their abundance in the protolith of the fault core rocks in other parts of the
exposure. We find this and the correlation of the core with the active fault trace as an
indication that this zone has been stable as the principal slip zone for a considerable
amount of time and displacement. In contrast, the sag deposits and associated
structures are inactive, representing a fossil structural environment, and there is no
geomorphic expression of these deposits or structures at the modern surface.
138
Figure 3.5: Fault core in Ladik. The fault core includes four layers rich with shear fabric relatively
to corresponding rocks in other parts of the fault zone. The PSS is on the contact between the fault
core and the northern block. The entire gouge zone is on the south side of the PSS and the other fault
core layers are on the south side of the gouge.
The gouge layer is dark bluish-gray to black. It is relatively stiff and dense. The
gouge material here is clayey and highly foliated, containing many slabs that are mm
to cm wide and overall parallel to the orientation of the fault. Many slabs are further
foliated at the sub-millimeter scale. The surfaces of those slabs show slip striations,
mostly with a sub-horizontal rakes. The gouge slabs behave plastically up to some
extent: they can be bent to a certain limit before they break. Their plasticity most
likely fluctuates with the seasons and the moisture content. Gouge slabs were easier
to part as time passed and the exposure dried out a few days after the excavation, but
they were also stiffer.
Gouge slabs were more defined and easy to part toward the north side of the
gouge layer, with the best expressed and most easy to separate set of surfaces found
in the few cm contact zone between the gouge and the shale of the northern block
(Figure 3.5). This well developed zone of foliated fabric was separated from rocks of
139
the northern block with a 0.5-2 cm wide layer of softer and more delicate gouge with
bluish-gray color that is similar to gouge found within some of the small secondary
faults. We suspect that this was the main zone that accommodated slip during the
1943 earthquake, as well as during previous recent earthquakes. This narrow zone
that combines the highly foliated gouge at the northern end of the gouge zone with
the softer narrow gouge layer is defined here collectively as the PSS within the fault
core. The surfaces within this zone show clear slip striations with a typical rake of
180°.
The other fault core layers are separated from each other by faults but do not
contain internal through-going slip surfaces. These layers are similar in their
lithology to rock bodies in other parts of the fault zone but they are significantly
more foliated and include many facoidal striated fault-parallel small scale shear
surfaces. Original sedimentary fabrics within those layers are distorted.
The composition and the intensity of shear fabric of the gouge layer, together
with its correlation with the surface expression of the fault, suggest that this is the
zone that accommodates most of the fault’s displacement. The slip is further
localized on the north side of the gouge in the PSS zone. This structure implies an
asymmetric pattern on two scales: most of the shear damage and distortion on the
fault core (meters) scale appears on the south side of the gouge, and within the gouge
(cm scale) on the south side of the active slip surface.
140
Thus, the observations of structural asymmetry in Ladik are consistent with each
other in their sense and have a variety of manifestations over scales of cm (gouge),
meters (fault core) and tens of meters (fault zone).
Bademci. Near the village of Bademci, 91 km east of the 1943 earthquake
nucleation zone, the NAF juxtaposes mélange on the north against schist marble and
metabasalt complex on the south. The site we describe here is located within a 400 m
wide and 3.5 km long gentle releasing bend of the fault, and was chosen despite this
geometry due to the opportunity to observe fault zone elements exposed in bedrock.
The local transtensional structure is associated with a series of small step overs with
the distance between individual strands typically about 5 m. The active trace of the
fault follows the bottom and the south side of a local valley. A road with a series of
switchbacks connects the active fault with an exposure of 2 and 3 m-wide gouge
zones that are 10 m apart from each other, located about 130 m to the south of the
PSS. They appear to be inactive as they have no geomorphic expression and one of
them is apparently displaced by a normal fault. Their fabric includes many small
shear surfaces with horizontal striations and with a trend that is oblique to the
general trend of the NAF in this area. These two gouge zones probably formed the
core zone of the NAF before activity shifted to its current location, and they possibly
still accommodate minor displacement during major slip events on the currently
active fault core (there are probably no additional major faults between the active
strand and theses two gouge zones, as we did not observe any along the road cut
between them).
141
We excavated a 10 m long, 2 m wide and ~2 m deep trench across the active
fault core where it crosses one of the switchbacks in a site that was accessible for a
backhoe. The trench (Figure 3.6) exposes the entire fault core and some of the wall
rocks. The fault in the trench separates mélange on the north from colluvium on the
south. Neither wall rocks present clear macro-scale evidence for brittle deformation
other than a cluster of faults in the mélange that are likely inactive as they have no
geomorphic expression.
Figure 3.6: Trench log of the fault core in Bademci. The inset on the left shows a map of the fault
trace in the vicinity of the trench (thick green line). The fault core includes five gouge zones but only
the northern gouge layer has clear continuous slip surfaces. They collectively form the PSS. The
close-up images show that shear fabric within the gouge is well developed in the vicinity of the PSS.
Further to the south the gouge is shattered with many smaller fractures and weaker preferred
orientation compare to the gouge in the north side of the fault core.
The fault core includes five individual gouge zones separated by slivers of
damaged and foliated rock over an 8 m wide zone. It is likely that all or some of the
gouge layers merge at depth below the trench. The two southernmost gouge zones do
not cut the surface and are likely inactive. The four southernmost gouge zones are
bent, and their fabric is distorted and not coherent. The gouge material in these four
zones is dark gray, very dense and relatively stiff. An anastomizing network of mm
142
scale shear fractures dissects the gouge, and the fractures in most places have no
clear preferred orientation (Figure 3.6). This suggests that at least recently, the gouge
was not sheared but was shattered. Although the material is not lithified, it is difficult
to disaggregate the gouge. Toward the northern side of this group of gouge zones,
the fractures become larger and the gouge material is softer. None of these gouge
zones contain a detectable through-going slip surface. These observations suggest
that these gouge zones have been distorted from slip events along the active zone
(see below), although they could still accommodate minor amounts of displacement
that are small compared to the total amount of fault displacement.
The gouge zone at the north end of the trench is very different from the other
four gouge zones. It contains several slip surfaces that can be tracked from the
bottom to the top of the trench (red lines in Figure 3.6). The gouge material is
dissected by numerous shear surfaces that are parallel to the overall orientation of the
fault. Those surfaces divide the gouge into 3-10 mm wide and up to several cm long
slabs of gray-black clayey, non-lithified flexible material (the color and the
flexibility depend on the moisture content. Immediately after the excavation of the
trench the gouge material was dark and flexible). The gouge slabs are further foliated
with sub-millimeter partings. The slabs surfaces are facoidal or flat, shiny in places
and display overall horizontal slip striations. Several of those surfaces connect
together to form the principal slip surface.
The structure of the fault is asymmetric: the active fault core is on the north side
of the fault zone that includes two more significant gouge zones. Within the active
143
fault core, the gouge zone that includes the currently active and dominant slip
surfaces and the most developed shear fabric is the northernmost out of five gouge
zones. If the gouge zones merge below the ground, the resultant gouge zone should
include a PSS on its northern side with intense shear fabric in its vicinity and more
shattered gouge farther to the south. These observations show that more damage in
the form of gouge material, shear-related fabric and intense fracturing was
accumulated on the south side of the zone of localized slip.
We have no well-constrained information about the long term relative time of
activity of the various gouge zones, nor on the temporal distribution of displacement
between the two fault cores. If these observed elements were active
contemporaneously while slip was localized mostly on the current active zone, the
structure can be considered to reflect a systematic accumulation of damage
preferentially on one side of the zone of slip localization. Based on the fabric of the
gouges, we believe that this is the case for the fault core. If, on the other hand, slip
migrated between the various fault zone elements, the asymmetric structure that we
observe may not reflect systematic long term accumulation of damage on one side of
the slipping zone. This ambiguity, combined with the structural irregularities of the
fault in this area, results in uncertainty in the interpretation of the local asymmetric
structure. Nevertheless, the sense of structural asymmetry observed here, and in the
other two sites along the 1943 rupture zone (where more constraints exist regarding
the relative activity time of the various fault zone elements), is similar.
144
Celtikci. The NAF near the village of Celtikci, 69 km east of the 1943
earthquake nucleation zone, has a relatively straight trace and no apparent branches
or step-overs. It juxtaposes Aspian or lower Senonian ophiolitic mélange on the
north against lower-middle Jurassic schist-marble-metabasalt metamorphic rocks on
the south. The mélange rocks are dark, soft, weakly lithified, rich with fragments and
show no noticeable layering or any other form of coherent internal structure. The
metamorphic rocks are of sedimentary origin; they are strongly layered and foliated
and are stained in a variety of colors. The bedrock on both sides of the fault is locally
covered with patches of colluvium and soil. The site is located in a mountainous area
where the fault crosses a topographic saddle.
Figure 3.7 shows a system of 6 trenches that we excavated across the fault zone,
exposing bedrock and colluvium for a distance of up to 60 and 75 m south and north,
respectively, with respect to the currently active fault trace. The exposure in the
trenches is discontinuous for 5 m immediately south of the active fault trace and for
12 m starting 22 m north of the active fault trace due to the presence of roads and a
water line that could not be disrupted by our excavations.
145
Figure 3.7: A system of six trenches excavated near Celtikci. The color patches represent the
lithology found within the trenches and not actual map distribution of rocks. Locations of active fault
traces are marked with red pentagons and dashed red-black lines. The map shows various types
(color-coded) of faults and fault clusters (where spacing between the faults is too small to show
individual faults). Many more faults that cut colluvium, and hence were active recently, appear south
of the current PSS (active fault “C”).
The saddle area is cultivated, and hence geomorphic evidence for recent
faulting, if it exists, has been obscured. Two medium quality indicators for the
location of the active fault appear east of the saddle. They include a bush line
(probably an old field boundary) with a deflection of 10.5 m and a channel with a
deflection of 4 m (Figure 3.7). The location of the recent rupture zone in the saddle
area was corroborated by a villager who witnessed the earthquake and its aftermath
in 1943. After excavating the trench system, we found that the location suggested by
the villager correlates with a fault that has a strong expression in the walls of trench
146
E (Figure 3.7), presenting fresh shear fabric and cutting colluvium and soil up to the
surface. If this recently active fault is projected towards the west along the local
trend of the NAF, it should show up on the walls of trench C. However, its projection
does not coincide with any other active fault there, implying that the active trace of
the fault is discontinuous or steps. This discontinuity, and a projection of similar
trending faults through the deflections mentioned above, suggest that the active fault
is locally segmented, stepping or bent in this area. We refer hereafter to the active
zone of the fault observed in trench E as the PSS.
The long term lithological contact across the NAF was visible in trenches C and
D about 25 m south of the current PSS (Figure7). In trench D, a few tens of cm wide
sub-vertical gouge layer separates the metamorphic rocks of the southern block from
a 1-3 m wide wedge of colluvium. The mélange rocks of the northern block appear
immediately north of this sliver of colluvium. In trench C the structure of the
slipping zone is similar in general to the one observed in trench D, but the colluvial
wedge is wider, replacing the actual gouge layer. We projected the trace of this fault
to its assumed location in trench A, where the trench exposes intact colluvium,
implying that the long term lithological boundary of the NAF in this area is not
active at least since the deposition of this colluvium. The depositional age of the
colluvium is unknown, but since its local source is missing and may have been
displaced or eroded, and since it is relatively consolidated, we conjecture that its
possible age may be late Pleistocene or older. Hence, the ‘long term’ NAF in this
area, which can be referred to as the ‘old PSS’ has not been active for more than a
147
few earthquake cycles. During this last phase of faulting, ruptures must have
occupied the PSS or one of the other faults that cut the colluvium (see below).
In addition to the two faults mentioned above, we observed and mapped several
tens of other faults on the walls of the trench system. We identify those faults as
secondary faults because they cannot be correlated with geomorphic evidence for
active faulting, and hence they must carry only minimal displacement, even if active.
In addition, most of them separate similar or identical rocks.
We differentiate the secondary faults into two groups according to the general
type of rock that they cut: bed-rock faults separate two bodies of bedrock, both of
which belong to either the metamorphic complex or the mélange. Some of those
faults include a layer of gouge while others lack such a layer. The stiffness and
lithification state of those gouge layers tend to correlate positively with the stiffness
and lithification state of the host rock. Most of the bed rock faults that were mapped
are likely related to NAF activity as suggested by their sub-vertical inclination and
their overall east-west trend, with the exception of the faults with the anomalous
trend that were mapped in trench F.
Many of the bedrock faults were associated with evidence indicating that they
are inactive. Some faults in trench C do not cut the layer of soil and colluvium that
covers the mélange rocks at the upper horizon of the trench exposure. They also have
no correlative faults in the colluvium of trench A. These faults ceased being active
before deposition of the colluvium, and they are marked as blue lines in Figure 3.7.
148
Bedrock faults without constraints for time of activity are marked with black lines.
None of the bedrock faults were associated with evidence of recent activity.
The other type of secondary faults we designate as colluvial faults, that is, they
cut and separate two bodies of colluvium. A thin gouge layer was found in one of
them, in the central part of trench A. Other colluvial faults display delicate features
such as rotation and alignment of grains along the slipping plane and obliteration of
shapes and fabrics in the colluvium. They can be recognized as subtle thin lines on
the trench wall, often branching upward into two or more branches. Typically, the
inclination of many of them decreases as they approach the surface. Some of the
colluvial faults do not cut the ground surface and hence were possibly inactive
during the most recent rupture events. Other colluvial faults cut the ground surface,
but this is not necessarily an indication for their activity time because the ground
surface may have been eroded, even after the last earthquake. Nevertheless, because
the bedrock faults in trench C do not show up in the colluvium of trench A, the
activity time of the colluvial faults in trench A (marked with red lines in Figure 3.7)
post-date the activity time of the bedrock faults in trench C. Some of the colluvial
faults in trench A may be correlated with faults marked in black in trench D. These
observations suggest at least two phases of faulting, one represented by the recently
inactive bedrock faults and the latter one represented by the recently active colluvial
faults.
Symmetry properties of the fault zone in Celtikci can be evaluated using the
distribution of secondary faults with respect to the PSS. Our mapping in Figure 3.7
149
shows that faults that were not involved in recent activity, mostly bedrock faults, are
distributed overall symmetrically with respect to the old PSS, and asymmetrically
with respect to the current PSS, with more bedrock faults found on its south side.
The asymmetry in the distribution of colluvial faults that were active in recent
geological history is more substantial, with many more such faults found south of the
current PSS and even south of the old PSS compared to their abundance north of
these two reference surfaces.
Based on these observations we conclude that damage during recent fault
activity (and at least since the deposition of the colluvium), manifested by secondary
faults, accumulated preferentially on the south side of the PSS. During previous
stages of faulting that preceded the deposition of the colluvium, symmetry in the
accumulation of damage is less constrained, although secondary faulting appears to
be overall symmetric with respect to the old PSS.
Hamamli. The task of finding fault zone elements such as the slipping zone,
gouge and damaged fault zone country rocks within river valleys associated with the
fault is usually frustrated because the fault is covered with vegetation, water or
young sediments. Nevertheless, we managed to find several of those elements near
the village of Hamamli, 28 km west of the 1944 rupture nucleation point. The
location of fault zone elements and critical exposures needed for delineation of the
fault zone structure determined the size of the study area (Figure 3.8).
150
Figure 3.8: Geological map of a fault section near Hamamle. The trace of the active fault is few
meters north of an exposure of fractured limestone, and has several tens of meters wide gouge zone on
its northern side. This suggests that more fault core material is present on the north side of the fault.
This section of the fault separates upper Lutetian and upper Eocene basaltic and
andesitic rocks on the north from Jurassic and Cretaceous limestone on the south
(Figure 3.8). The Gerede river flows mostly north of the active trace of the fault (see
section 2.3 below). Due to the river and the presence of Quaternary units that cover
most of the area south of the river, we could not determine whether the location of
the lithological contact between the south and the north blocks coincides with the
active trace of the fault. In the western part of the study area, the active trace of the
fault has several compressional jogs associated with small pressure ridges. In the
eastern part of the study area, the active trace of the fault has a small releasing step-
over associated with a spring and active deposition of travertine. A sheet of
travertine covers the area immediately to the north of the step-over and the spring. A
continuous series of shutter ridges (symbolized in orange in Figure 3.8) creates a
151
topographic ridge along the northern side of the fault between the western margins of
the study area and the travertine deposition zone.
We found in this area two fault zone elements that can provide insight about the
structure of the fault zone and its symmetry properties: an exposure of damaged
limestone south of the fault and an exposure of a gouge zone north of the fault.
A small exposure of limestone (red spot in Figure 3.8) was found on the south
bank of the river where it meanders and creates a cliff. The limestone is fractured,
and the fragment size is on the decimeter scale, showing a moderate damage level.
The northern tip of the exposure is about 5-10 m south of the projected location of
the fault. This exposure is 80 m west of the releasing step-over that is associated with
the spring. If the fault continues its releasing trend farther to the east of the spring
where it is not exposed, additional branches of the fault could be present south of the
current location of the projected fault in Figure 3.8. In this case the distance from the
limestone to the slipping zone could be 5 m or less.
A wide zone of gouge was found north of the fault in the western part of the
study area. The active trace of the fault bounds the gouge zone on its southern side
and the river bounds the gouge on its northern side. The location of the northern
boundary of the gouge zone is therefore uncertain. The minimum width of the gouge
is between 30 and 40 m. We could not eliminate the possibility that some of the
gouge material exposed on the surface arrived there by sliding from the top of the
hill. Despite the uncertainty regarding the exact width of the gouge zone due to the
uncertain location of its northern boundary, and despite the uncertain origin of some
152
of the gouge material found on the surface, the gouge zone layer seems to be robust
and wide. The significant width of the gouge and its structural relations with the
active trace of the fault imply that locally, a 10’s of meters wide fault core separates
the northern country rock from the active trace of the fault.
Although the constraints regarding the exact geometry of the structural elements
that we describe above are partial and the fault is locally geometrically distorted, the
PSS seems to be much closer to the southern side of the fault core and its contact
with the damaged country rocks than to its northern side. This means that more fault
core scale damage products are found north of the slipping zone in this area. This
sense of damage asymmetry is compatible with the one inferred from the large scale
geomorphological observations for this area (see below).
Geomorphologic observations
Large scale adjustments of rivers to the fault trace: Figure 3.9 shows eight
sections of the NAF along the 1943-44 ruptures that deflect the valleys of major
rivers. The deflected sections of the rivers are parallel to the fault and deviate from
their general flow direction (all of them but the Eksik river flow northward). In most
cases, the deflection starts and ends with a clear change in the river’s flow direction.
The rivers, as well as the fault in these sections, are situated within deep erosional
valleys, with the lowest point of the valley (where the water flows) located either
south or north of the fault. We checked how consistent the location of the fault was
153
with respect to the river, i.e. do the rivers flow systematically on one side of the fault
or not, and if yes on which side it is.
The results are summarized in Table 3.2. Along the 275 km long 1943 rupture,
five river valley sections with a total length of 111 km are associated with and flow
along the fault (panels 4-8 in Figure 3.9). Three of these rivers – Eksik, Karovun and
Mavga - flow exclusively on the south side of the fault. The Kizilirmak and the
Destek rivers flow for most of their length south of the fault but have a section that
flows north of the fault as well. Along the 1943 rupture, a total of 83 km of river
length, which is 75% of the river length associated with the fault, flows south of the
location of the rupture.
Along the 183 km long 1944 rupture segment, six river valley sections with a
total length of 64 km are associated with and flow along the fault (panels 1-3 in
Figure 3.9). Five of these rivers – Abant, Goynuk, Mudurnusuyu, Kuzoren and
Karanlik - flow exclusively on the north side of the fault. The Gerede river flows for
most of its length north of the fault but has a section that flows south of the fault as
well. A total of 57 km of river length, which is 89% of the total length of river
associated with the fault, flows north of the location of the rupture. In the section of
the fault that includes the nucleation zone of both earthquakes (Kondo et al., 2004.
Emre et al., 2005), 4 km of river flow north of the rupture and 11 km of river flow
south of the rupture (panel 4 in Figure 3.9).
154
Figure 3.9: Location of fault trace with respect to river valleys. Panels 1-8 show geological maps
of eight fault sections associated with valleys of major rivers. Their locations along the 1943-44
ruptures are shown in the index map below. Lithology if from MAT Geological maps Zonguldak and
Sinop Quadrangles, 1:50,000 (2002). Rivers along the 1944 rupture (panels 1-3) flow mainly north of
the fault and rivers along the 1943 rupture (panels 5-8) flow mainly south of the fault. Rivers in the
nucleation zone of the earthquakes cross the fault several times. This asymmetric systematic erosion
pattern likely controlled and reflects an asymmetric damage structure. See Table 3.2 for details.
155
The fault sections that are associated with river valleys are relatively straight and
free of geometrical irregularities. However, fault jogs and other complexities at a
scale comparable to the typical width of a river valley (that may perturb the flow of a
river) exist in a few places. The three sites in which a river changes its side with
respect to the fault are such geometrical irregularities: segmentation of the fault in a
compressional bend in the western part of the Gerede river section, and extensional
fault jogs in the eastern part of the Kizilirmak and Destek river sections. The Gerede,
Kizilirmak and Destek rivers flow generally from west to east. Given the consistency
in the geometrical relations between the fault and the rivers in the other, simpler fault
sections, it is reasonable to assume that in the absence of the right step-over in the
trace of the fault, the Kizilirmak and Destek rivers would have continue flowing on
the south side of the fault. The left bend in the trace of the fault in the Gerede river
section should have ‘encouraged’ the river to continue flowing on the south side of
the fault. Nevertheless, the river crosses the fault right there to flow along its
northern side. This analysis suggests that if the fault geometry was even simpler than
the current geometry, we would observe larger portion of the river’s length flowing
on the south side of the 1943 rupture and on the north side of the 1944 rupture. In the
short fault section that includes the nucleation zone of both earthquakes, the fault is
geometrically highly complex and the river changes its relative location with respect
to the fault 3 times (Table 3.2).
156
Table 3.2: measured length of river valley sections that flow north and south with respect to the
fault along the 1943 (43) and 1944 (44) rupture sections.
River
Km
North
Km
South
Total
length
% length
North
% length
South
side
change
Abant (44)
3.8 0 3.8 100.0 0.0 0
Goynuk (44)
4.1 0 4.1 100.0 0.0 0
Mudurnusuyu
(44)
8.6 0 8.6 100.0 0.0 0
Kuzoren (44)
7.1 0 7.1 100.0 0.0 0
Karanlık (44)
9.2 0 9.2 100.0 0.0 0
Gerede (44)
23.9 6.9 30.8 77.6 22.4 1
Eksik (43)
0 19.9 19.9 0.0 100.0 0
Karovun (43)
0 3.3 3.3 0.0 100.0 0
Kizilirmak (43)
13 29.2 42.2 30.8 69.2 1
Mavga (43)
4.9 15.1 20 24.5 75.5 0
Destek (43)
10.2 15.3 25.5 40.0 60.0 1
Nucleation zone
4 15.2 19.2 20.8 79.2 3
1943 rupture
28.1 82.8 110.9 21 75
1944 rupture
56.7 6.9 63.6 89 11
Total 43 +44
84.8 89.7 193.7 44 56
These observations indicate that erosion on a large scale is more intense on the
south side of the 1943 rupture and on the north side of the 1944 rupture, suggesting a
systematic asymmetric erosion as indicated by the location of the lowest point in the
river’s profile with respect to the location of the fault. The distribution of rock types
across the fault (Figure 3.9) in these river valley-fault sections cannot explain this
asymmetric pattern: the contrast in rock type is not consistent along either of the
rupture segments, and some rupture sections separate identical or very similar
surface lithologic units (e.g. panel 1 in Figure 3.9). Furthermore, the river valley is
often located on the side of the fault with the assumed more resistant rock types (e.g.
157
Mavga River, panel 7, Figure 3.9). In the absence of other systematic, rupture-scale
controls on the drainage pattern, we conjecture that the observed adjustment of the
river profiles to the fault trace is likely a large scale erosional response to the fault
damage structure.
Morphometric analysis of correlative terrains near Ismetpasa, 1944 rupture: We
studied the morphometry of two correlative terrains along the 1944 rupture that are
offset by the NAF. Specifically, an Eocene andesitic-basalt cover unit that originally
covered the fault (Herece and Akay, 2003) is preserved near Ismetpasa, with about
half of the rock unit on each side of the fault. The terrain south of the fault is 3.5 km
at its widest part, 24 km long and has an area of 56.8 km
2
, whereas the terrain north
of the fault is 4 km at its widest part, 15.5 km long and has an area of 34.5 km
2
. The
two offset terrains are considerably similar in their intrinsic and extrinsic variables
(Figure 3.10, Table 3.3). Figure 3.10 shows results of an aspect analyses. The aspect
is a controlling variable that can have dramatic effect on weathering and erosion due
to its impact on vegetal cover and possibly on other erosion controlling factors. The
two terrains face mostly to the south and have an identical mean aspect direction
(164°). Their topography structure is very similar with a mountainous area in their
northern part been drained overall to the south.
Figure 3.10 also shows slope analyses and the distribution of grades throughout
the terrains. In both terrains, high slope values are clearly associated with channels of
low order, while the hillslopes are quite similar. The apparently higher mean slope
value in the north (12.4°, STDV = 8.4 vs. 9.1°, STDV = 6.3 in the south) seems to be
158
the result of significantly higher grades in the immediate vicinity of active channels
of lower orders in the northern terrain. The difference between the terrains in the
average relief of the basins is 69 m (relief ranges 179 to 379 with mean value of 265
in the north and 120 to 289 with mean value of 196 in the south). The relief
difference between the terrains overall is 75 m. This difference is probably not
significant enough to be the main cause for the difference in the mean slope value. In
turn, it is likely a consequence of deeper incision by active lower order channels in
the northern terrain. In general though, evolving basins with lower slope values are
expected to develop higher drainage density (Parker, 1976).
Figure 3.10: Slope-aspect analysis. GIS generated distribution and statistics of slope (left panels)
and aspect (right panels) of two correlative terrains displaced by the NAF along the 1944 rupture
section. See location of terrains in Figure 3.11. Slope values are higher on the north especially near
active channel, probably reflecting deeper and denser incision. Aspect is very similar between the
terrains with an identical mean.
The two terrains are only a few tens of km apart and experience the same
climate conditions. Agriculture and pasture are the most common landuse
characteristics, and both limit the growth of trees which are sparse in the studied
159
areas. Even in the likely case that the response times of the basins are significantly
larger than the current vegetation regime, they probably experienced similar
vegetation regimes throughout their displacement history due to their geographical
proximity and similarity between their other physical characteristics. The underlying
geological structure of the terrains is not known in detail, although we assume the
two rock bodies should have similar structure, minimizing potential impact on the
differences in morphological patterns. An indirect way of verifying this is to
calculate the bifurcation ratio of the basins in the terrains. The mean R
b
value in the
north is 3.1, and ranges between 3 and 4 with one basin having an R
b
value of 2
(basin 9). The mean R
b
value in the south is 3.2, and ranges between 3 and 3.7, with
one basin having an R
b
value of 4.6 (basin 7). Bifurcation ratio values between 3 and
5 suggest a relatively homogenous lithology and minimum impact of the geological
structure on the morphology (Strahler, 1964, Hugget and Cheesman, 2002).
Table 3.3 summarizes the comparison between the terrains. The similarities in
the various controlling variables suggest that these terrains should have very similar
resistance to erosion unless an additional controlling variable is introduced.
Table 3.3: comparison of variables that may control and affect erosion between the two terrains
(figures 3.10, 3.11). Values correspond to the entire terrain and are not the mean values for all the
basins.
Vriable Southern terrain Northern terrain
Lithology Eocene andesitic - basalts Eocene andesitic - basalts
Geological structure Massif Massif
Relief 330 m 405 m
Slope 0-47°, mean slope 9.1° 0-45°, mean slope 12.4°
Aspect Mostly south faced, mean aspect
164°
Mostly south faced, mean aspect
164°
Climate Semi-arid Semi-arid
Landuse Agricultural – pasture land Agricultural – pasture land
Vegetation Rare woods Rare woods
160
Figure 3.11 shows the two studied terrains, and is color-coded for elevation and
overlaid on a DEM. The automatically delineated stream networks of the basins are
displayed, and the streams are color-coded according to their order. The stream
network on the northern side of the fault appears to be much denser. Since the two
terrains (with similar lithology and structure) are subjected to similar erosion agents,
this difference is probably the result of lower-strength rocks in the northern terrain
that facilitate more erosion and higher drainage density.
Figure 3.11: Automatically delineated drainages in the two correlative terrains color coded
according to their order (GIS ESRI ArcHydro extension). Drainage pattern is much denser in the
northern terrain. See Table 3.4 for details of the morphometric analysis.
161
The results of the linear, aerial, network and relief characteristics analysis are
presented in Table 3.4 and in Figure 3.12. Drainage density (D
d
), which is the most
sensitive parameter to the erodibility state of the rocks, is significantly different
between the terrains, with the mean D
d
value in the north (10) almost double the
mean D
d
value in the south (6). Drainage density should not be influenced strongly
by the relief and Figure 3.12a demonstrates this, showing a weak dependence
between D
d
and relief values. Note that for corresponding relief values, D
d
values are
still higher in the north. Basin area is expected to have a correlation with drainage
density with higher D
d
values in smaller basins. The plot in Figure 3.12b shows that
this correlation applies for the terrain north of the fault and exists in a weak form for
the terrain south of the fault. Figure 3.12b also suggests that the two basin
populations are different from each other because basins with corresponding areas
have higher D
d
values in the north compared to the south.
162
Figure 3.12: Correlations between morphometric parameters. Correlation between drainage
density and relief (a), basin area (b) and ruggedness number (c) showing that drainage density values
are systematically higher in the northern terrain. (d) The higher values of ruggedness number in the
north are affected also by the local relief values.
163
Table 3.4: results of morphometric analysis of two correlative terrains displaced by the 1944
rupture section. Data base is shown in Figure 3.11.
R
b
H Rn Hi D
d
F
Basin
N S N S N S N S N S N S
1 3.3 3 185 120 2.5 0.6 0.43 0.59 14 5 89 22
2 3 3.1 206 230 2.2 1 0.470.65 11 4 62 13
3 3.2 3 179 233 1.8 1 0.62 0.67 10 4 69 12
4 3 3.2 186 150 2.2 0.9 0.640.69 12 6 66 28
5 3 3 267 123 2.2 0.8 0.450.65 8 6 30 32
6 3.1 3.1 365 220 3 0.7 0.47 0.53 8 3 25 12
7 3.3 4.6 379 223 3.2 1.7 0.53 0.56 8 8 23 35
8 4 3 280 200 2.7 1.2 0.550.49 10 6 41 38
9 2 3.1 208 220 2 1.2 0.410.54 10 5 58 17
10 3.5 3.4 319 200 3 1 0.61 0.66 9 5 40 26
11 3.1 3.3 340 180 3.2 1.6 0.49 0.53 9 9 38 68
12 3.1 180 1.5 0.57 8 60
13 3 210 1.5 0.54 7 34
14 3 160 0.8 0.74 5 22
15 3.7 205 0.8 0.65 4 18
16 3.3 289 1.1 0.63 4 8
StDv. 0.48 0.41 76.51 43.50 0.50 0.34 0.08 0.07 1.87 1.71 21.02 16.85
Ave. 3.1 3.2 265 196 2.5 1.1 0.52 0.61 10 6 49 28
N S N S N S N S N S N S
Since smaller basin areas and smaller relief values are expected to correlate with
higher D
d
values, and since the northern terrain includes basins with smaller area and
higher relief, we calculated area weighted and inverse-relief weighted averages of
drainage density for each terrain. The area weighted D
d
values are 5 for the south and
9 for the north, and the inverse-relief weighted D
d
values are 0.023 for the south and
0.039 for the north, in close agreement with the non-weighted values. These results
confirm that the observed drainage densities are intrinsic, independent properties of
the basins.
164
Stream frequency (F), an additional direct expression of rock erodibility by
fluvial processes, shows a strong contrast between basins on the north and basins on
the south, all other things being equal. Mean F values are almost twice as high in the
north. The drainage pattern is dendritic in the south and pinnate dendritic in the
north, i.e. there are more lower-order tributaries in the north. This difference is
confirmed by the hierarchical drainage structure of the basins: 5 out of the 11 basins
in the north have 4 stream orders and the other 6 basins have 3 stream orders, while
in the south only one out of the 16 basins have 4 stream orders and the other 15
basins have 3 stream orders. Extended fluvial elaboration will expand the drainage
network by adding lower order tributaries, hence increasing the drainage density and
the stream frequency (e.g. Chorley et al., 1984).
Higher D
d
and F values are consistent with higher values of Ruggedness number
(R
n
) and with lower values of Hypsometric integral (Hi), and this is what we observe
for the mean values: R
n
mean value in the north is more than twice its mean value in
the south (see also Figure 3.12c), and Hi mean value in the south is 17% larger than
the mean Hi value in the north. The ruggedness number is also useful for comparison
of basins with different relief. Figure 3.12d shows that for corresponding relief
values, R
n
values are still higher in the northern terrain.
In summary, the northern basins show geomorphic parameters that indicate that
more erosion (work) is being done on the north side of the fault. Ritter et al. (2002,
and references therein) suggested that morphometry in mature basins reflects an
adjustments of geomorphic variables that is established under the constraints of the
165
prevailing climate and geology. Since both are similar between the basins with the
only possible difference been the strength of the rock due to damage, these
observations are consistent with a higher degree of rock damage on the north side of
the fault, which in turn is consistent with the near-field observations of fault zone
damage asymmetry. If those basins have not matured yet, the higher relief and the
apparently steeper slopes in the north should have led to a lower D
d
mean value but
the results show the opposite.
Drainage density in badland topography along the 1943 rupture. Drainage
patterns were studied at two sites in the vicinity of the villages of Celtikci and
Tekeler along the 1943 rupture (Figure 3.13). The sites are two km apart and their
drainage patterns have similar characteristics. The fault juxtaposes in this area
Apsian(?)-Lower Senonian ophiolitic mélange on the north side against Paleozoic
(?) - Triassic, Lower and Middle Jurassic schist-marble and metabasites on the south
(Herece and Akay, 2003). The ~100 m wide belt of metamorphic rocks on the south
side of the fault is typified by a dense gully network forming a badland topography.
In contrast, the drainage pattern across the fault in the mélange rocks consists of a
few well-spaced channels. The drainages are delineated in yellow in Figure 3.13,
illustrating the contrast in the density of the gully networks between the two sides of
the fault. This sense of asymmetry is surprising given that the mélange rocks are
weaker and less permeable compare to the metamorphic rocks and are therefore
expected to develope a denser drainage network. At both sites, the density of the
drainages on the southern side seems to be the highest near the fault and subsides as
166
a function of distance from the fault. The damage density is also expected to
decrease as a function of distance from the fault, and we suspect that these two
trends are related, with the damage pattern affecting the erosion pattern. Further
details are given in the following discussion section.
Figure 3.13: Bad-land topography in two sites along the 1943 rupture. The 1943 rupture section
of the NAF near the villages of Tekeler (left) and Celtikci (right) separates highly eroded terrain with
badland topography on the south from mildly eroded terrain on the north. Basins and channels are
mapped on the images at the bottom. Badland pattern is more developed in the basins near the fault,
strengthening the inference that this erosion pattern is likely controlled by the damage structure of the
fault. Width of view in both images is about 1 km.
167
Discussion
Summary and synthesis of mapping results
Our observational studies are associated with various possible manifestations of
damage asymmetry across the 1943 and 1944 rupture zones. The combined set of
geological and geomorphological observations spans scales ranging between mm to
several km.
Along the 1943 rupture zone, we presented geological mapping at three sites,
geomorphological mapping at two sites, and a rupture-long adjustment pattern of
river profiles to the fault. In Celtikci, geological mapping in a system of trenches
shows that secondary faults that are inactive or for which we have no constraints for
their recency of activity are distributed overall symmetrically across the long term
lithological boundary of the fault. In contrast, we found substantially more active or
recently active secondary faults on the south side of the PSS and even south of the
long term lithological fault. This observation suggests that during the recent fault
history, the center of activity has shifted northward and an asymmetric damage
structure has developed, overprinting the older fault structure.
In Bademci, we found the primary active fault core and two other major gouge
zones south of it. The southern gouge zones correspond to faults that have become
inactive at some time in the past; there is no geomorphic expression of continued
activity although we cannot preclude that they still accommodate minor
displacement during large NAF earthquakes. A trench exposure over the active fault
168
core shows that slip is localized along the northern edge of an 8 m-wide zone with
several embedded gouge layers. The most developed shear fabric was found along
the PSS zone, while farther south of the PSS, the gouge is more shattered. If the
various gouge layers merge below the trench, which they probably do, the resultant
gouge structure is similar to that observed along the San Jacinto fault by Dor et al.
(2006a). There, the PSS was found close to the southwest margin of the gouge, with
intense shear fabric immediately adjacent to it, followed by shattered gouge farther
to the northeast. A similar sense of asymmetry was found in other nearby gouge
exposures and in the structure of a ~100 m wide seismically observed low velocity
trapping structure south of the gouge mapping site (Lewis et al., 2005). Seismic
imaging of the local velocity structure along this section of the San Jacinto fault
shows that the more damaged side of the fault has faster seismic velocities at depth
(Scott et al., 1994).
The relative time of activity of the various structural elements in Bademci is
uncertain, although the gouge fabric suggests that displacement activity has been
localized on the northern side of the fault core for quite a while. More relevant
information and constraints could probably be resolved if it was possible to excavate
a longer trench or more trenches. However, in addition to the similarity of the gouge
structure to that of the San Jacinto fault, the sense of asymmetry in the fault core of
Bademci repeats in other sites and in other forms along this rupture. In all the other
sites, the observations are more robust and better constrained. These correlations
169
strengthen the interpretation of the observed structure in Bademci as an expression of
an asymmetric accumulation of damage during many rupture events.
In Ladik we found strong structural asymmetry at three scales: substantially
more secondary faults and other fault zone elements were found south of the fault
core over a scale of several tens of meters; more shear fabric and distorted rocks
were found south of the principal active gouge zone within the fault core over a scale
of meters; and the entire gouge with its well developed shear surfaces was found
south of the PSS over a scale of centimeters. The PSS in Ladik is the most localized
compared to the PSS in other sites, and is well correlated with the remarkably
straight geomorphic expression of the currently active fault. We note that the south
side of the fault in Ladik displays more damage despite the presence of the Ladik
Lake transtensional step-over three km to the west. The northern side of the fault
zone is on the block that should have experienced the influence of the step-over and,
therefore, may potentially have accumulated more damage.
In evaluating the robustness of the observations and interpreting them in terms
of an asymmetric accumulation of damage throughout the recent geological history,
we assign the largest confidence to Ladik. This site has the narrowest geologic and
geomorphic localization of slip, a consistent expression of asymmetry in several
forms and in various scales, and a simple fault structure. In Celtikci there are
constraints on the relative activity time of the secondary faults suggesting the recent
development of an asymmetric fault zone structure, but the distribution of these
faults is the only manifestation of the asymmetry and the fault seems to be stepped in
170
the vicinity of the site. In Bademci we observe a clear asymmetry in the structure of
the fault core, but the relative activity time of the various fault core elements is not
certain and the local geometrical complexity of the fault results in ambiguity in the
observed asymmetry. Ladik is only 31 km from the termination zone of the 1943
rupture while the other two sites are closer to its nucleation zone. We expect a
stronger and more coherent signal toward the termination zone of a fault section that
accommodates ruptures with a preferred propagation direction, based on reasons
discussed below.
The asymmetric distribution of secondary faults across the PSS observed in
Ladik and in Celtikci is similar to the main expression of asymmetry observed by
Dor et al. (2006a) for the structure of the SAF near Palmdale, California. We
consider secondary faulting a manifestation of damage, especially for the evaluation
of symmetry properties, because the seismic radiation during an earthquake is
expected to produce and activate branching surfaces in the primary tensional
quadrant, depending on the pre-stress conditions and the rupture velocity (e.g.,
Poliakov et al., 2002; Rice et al., 2005). Strike slip faults are known to have “flower
structures” (Sylvester and Smith, 1976, Sylvester, 1988) that are generally not
symmetric. However, a generic flower structure is not expected to have a systematic
sense of asymmetry in a fault section that is approximately straight and vertical. In
our study, the consistency of the asymmetry of secondary faulting with other types of
observations, together with the above theoretical considerations, suggests that the
171
asymmetric distribution of secondary faults can be used to characterize the symmetry
properties of the fault zone structure.
Observations of gully networks on a scale of 100-200 m in two locations
between Celtikci and Bademci show that badland topography is significantly more
developed and gully networks are much denser on the south side of the fault.
Badland topography is not developed outside of this zone, and the increase in gully
density with increasing proximity to the fault is consistent with the likely fault-
normal trend in damage intensity. We thus infer that the distribution pattern of
badland topography reflects the damage structure of the fault zone. The mélange
rocks are inherently weaker and less permeable in comparison to the metamorphic
rocks and are therefore expected to develope denser drainage system, but the
opposite is observed. The slope of the hill is similar between the two sides of the
fault and can not explain this difference either. We therefore suggest that the
underlying damage structure of the rocks controls the erosion pattern we observe.
The lateral dimensions of the badland pattern is similar to the width of the layer
of granitic pulverized rocks with reduced grain sizes and badland topography found
along the SAF by Wilson et al. (2005) and Dor et al. (2006b). It is difficult to
identify pulverization in a hand sample of pelitic metamorphic and mélange rocks
because of their original very small grain sizes and the lack of coherent, visible grain
fabric. However, the similarity in the morphology and width between the pulverized
rocks layer from the SAF, where pulverization is clearly associated with the fault
damage zone (Dor et al., 2006b), and the zone of badland topography we discuss
172
here suggests that intense damage may be the cause for the geomorphic pattern we
report. Dor et al. (2006b) found that the layer of pulverized rocks appears to be
shifted to the northeast side of the SAF, and share the same sense of asymmetry with
other, smaller scale geological damage products. We observe similar consistency
along the NAF 1943 rupture zone where the assumed damage-related erosion pattern
and the smaller scale geological damage products have the same sense of asymmetry
with respect to the fault.
The largest scale observations we made involve the relative location of river
valleys with respect to the fault. The total length of the fault sections that ruptured in
1943 and are associated with river valleys is 111 km, which is 40% of the 1943
rupture length (275 km). We found that in 75% of the combined length of these river
valley–fault sections, the river flows along the south side of the mapped fault trace.
Three rivers flow exclusively south of the fault and two rivers cross the fault to its
northern side once, in a place where the fault has a major extensional step-over. The
geometry of the river valleys in these two cases suggests that in the absence of these
two geometrical disturbances the rivers would have continued flowing only on the
south side of the fault. In that case, the fraction of total river length that flows south
of the fault along the rupture would have been larger.
Along the 1944 rupture zone we showed that the active trace of the fault in
Hamamli is probably on the south side of the fault core, closer to the southern block
based on the distribution of fault zone elements. We are uncertain about the exact
173
width of the gouge zone on the north side of the fault and to what extent a local
series of step-overs might affect the pattern we observe at this site.
Quantitative comparison of morphometric parameters of two systems of
drainages developed in equivalent terrains suggests that the intensity of erosion by
means of fluvial processes as expressed by drainage density and other parameters is
considerably higher on the north side of the fault. We did not find any intrinsic or
extrinsic variable that could produce lower rock resistance in the north and explain
the erosion asymmetry that we observe, and we infer that damage asymmetry could
be the reason for this contrast.
The total length of the fault sections that ruptured in 1944 and are associated
with river valleys is 67 km, which is 37% of the 1944 rupture length (183 km). We
found that in 89% of the combined length of these river valley–fault sections, the
river flows north of the mapped fault trace. Five of these rivers flow exclusively
north of the fault and one of them (the Gerede river) crosses the fault from south to
north despite a local bend of the fault to the left.
Geological mapping of strike slip faults has suggested that the zone of high
rock damage extends to a distance of up to hundreds of meters from the fault core
(e.g. Chester et al., 2004). This length scale is compatible with our geological
mapping and the width of seismic trapping structures (e.g., Ben-Zion et al., 2003;
Lewis et al, 2005), but can not explain our larger scale geomorphological
observations in terms of the distribution of damage (the fault-normal dimensions of
the river valleys and the terrains near Ismetpasa is a few km). Our observed large
174
scale geomorphological anomalies may be associated with the 2-3 km zones of lower
damage that are imaged by various geophysical signals (Schulz and Evans, 2000;
Ben-Zion and Sammis, 2003, and references therein). These include gravity
anomalies (e.g., Stierman, 1984), elevated geodetic signals (e.g., Fialko et al., 2002;
Hamiel and Fialko, 2007), seismic anisotropy with fault-parallel cracks (e.g., Peng
and Ben-Zion, 2004; Cochran et al., 2006) and elevated seismic scattering (e.g.,
Ravenaugh, 2000; Peng and Ben-Zion, 2006).Those geophysical observations
provide a physical basis for the presence of a few km wide zone that is weaker than
its surrounding rock due to elevated levels of strain, crack density, hydrological
properties etc., making those zones more susceptible to erosion.
Dor et al. (2006a) provides a detailed discussion on the rationale for inferring
from surface observations properties of earthquake ruptures at depth and on the
associated potential complexities and interpretation problems. To overcome such
difficulties we made multi-signal multi-scale observations at numerous sites along
the 1943 and 1944 rupture sections. The systematic observations at the different sites
along each of the examined fault sections, and self-consistency of signals associated
with the different methods and scales, allow us to infer that the observed structural
and morphological asymmetries reflect underlying asymmetry of rock damage
associated with the NAF. The two sets of observations with an opposite sense of
asymmetry along the 1943 and 1944 rupture sections are compatible with eastward
and westward preferred rupture directions along those fault sections, respectively, as
observed for the large 1943 and 1944 earthquakes.
175
Possible implications of structural asymmetry
The presented evidence of systematic damage asymmetry suggest the existence
of a persistent symmetry-breaking mechanism in the dynamics of earthquake
ruptures along the 1943 and 1944 rupture zones. Since the observations were done
along relatively straight near-vertical fault segments, and the NAF separates different
lithologies (Sengor et al., 2005), the most likely symmetry-breaking mechanism in
this problem (Ben-Zion, 2006) is the tensile vs. compressive changes of normal
stress at the opposite rupture tips on a bimaterial interface.
The loading, nucleation and rupture direction of some of the NAF events (Figure
3.1) are similar to the setting and results of the sliding experiments with two different
foam rubber blocks of Anooshehpoor and Brune (1999). The large ruptures in those
experiments propagated unilaterally toward the loading area, in the direction of
motion of the slower velocity block, with properties similar to those predicted for
wrinkle-like ruptures. Small events in the experiments of Anooshehpoor and Brune
(1999) had different propagation directions and various transient properties. Xia et
al. (2005) performed sliding experiments along a bimaterial interface for several
loading configurations and obtained asymmetric bilateral ruptures. In all cases, the
rupture fronts in the direction of motion of the slower block had stable properties that
did not depend on the employed experimental conditions. In contrast, properties of
ruptures in the other direction varied from case to case and had transient features.
The similarity between the directions of large ruptures in the experiments of
Anooshehpoor and Brune (1999) toward the loading, and the rupture directions and
176
loading configuration of the 1943 and 1999 earthquakes on the NAF, suggests that
the local velocity structures may have dictated the direction of rupture propagation of
these earthquakes.
If the NAF makes a moderate to high angle with the maximum principle stress
as observed for the San Andreas fault (e.g., Hickman, 1991, and references therein),
and if ruptures propagate with subshear velocities as observed for most earthquakes
(e.g., Mai 2004), the cumulative effect of many ruptures on a lithology contrast
between the opposite sides of the fault is expected to produce more rock damage on
the side with faster seismic velocities at depth (Ben-Zion and Shi, 2005). The sense
of damage asymmetry can be reversed if ruptures propagate with super shear
velocities (Weertman, 2002), and damage is expected on both sides of the fault in
situations where the angle between the maximum compressive stress and the fault is
shallower than 15º (Templeton et al., 2006). Assuming that most large earthquakes
on the NAF are associated with standard subshear velocities and moderate to high
angles to the maximum compressive stress, we expect faster seismic velocities at
depth on the sides with higher rock damage. These are the south side of the 1943
rupture sections and the north side of the 1944 rupture section. These expectations
should be tested with future imaging studies along those fault sections.
We note that the width of the fault zone based on the distribution of secondary
faults and related structural elements in Celtikci and Ladik is at the scale of several
tens of meters, but that the strength and coherency of the geological signals are larger
in Ladik, which is closer to the end zone of the rupture. These two features are also
177
compatible with theoretical predictions for rupture along bimaterial interface. Ben-
Zion and Shi (2005) showed that the width of the damage zone, generated by a
feedback between slip velocity on the fault and energy absorption in the bulk, is
expected to be approximately constant (Figure 3.2b). However, the magnitude of
normal stress variations at the propagating tip, and hence expected intensity of rock
damage, increase with propagation distance along the interface due to continual
transfer of energy at the rupture tip to shorter wavelength (Adams, 1995, Ranjith and
Rice, 2001; Ben-Zion and Huang, 2002).
Another example of the possible control the local velocity structure may have on
rupture direction is given by the earthquakes of M=6.0-7.1 1967 (Ambraseys &
Zatopek, 1969; McKenzie, 1972) and M=7.1 (Duzce) 1999b (KOERI, USGS). These
two earthquakes ruptured on parallel branches of the NAF system that separate the
Almacik block from the surrounding crust. The Almacik block was uplifted rapidly
from depth and likely has faster seismic velocities at depth compared to its bounding
blocks. The 1967 earthquake propagated along the southern branch of the fault to the
west and the 1999b earthquake propagated along the northern branch of the fault to
the east (Figure 3.1). The propagation direction of these two earthquakes can be
explained by the local velocity structure in this area: both earthquakes propagated in
the direction of motion of the block with the assumed slower seismic velocities.
The measurements presented in this paper contribute to the growing body of
evidence (Rubin and Gillard, 2000; Rubin 2002; Lewis et al., 2005, 2007, Dor et al.,
2006a,b) that earthquake ruptures along large strike-slip faults have a preferred
178
propagation direction that is related to the velocity structure as predicted for wrinkle-
like ruptures on a bimaterial interface (e.g., Weertman, 1980, Andrews and Ben-
Zion, 1997; Cochard and Rice, 2000; Shi and Ben-Zion, 2006; Dalguer and Day,
2006; Ampuero and Ben-Zion, 2007). A wrinkle-like mode of rupture can have
important implications for various aspects of earthquake and fault mechanics,
including frictional heat, effective constitutive laws, faults interaction and more (e.g.,
Ben-Zion, 2001). An ability to infer the likely propagation direction of earthquakes
on large continental strike-slip faults from observations of structural properties can
contribute significantly to estimates of ground shaking in large metropolitan areas
like Istanbul, San Francisco and Los Angles. The finding discussed in this work, and
the earlier results of Dor et al. (2006a, b) on damage asymmetry across faults of the
San Andreas system in southern California, should be substantiated with additional
geological and geophysical observations.
Acknowledgments
We thank the Turkish villagers for allowing us to work on their property,
providing generous help and support, and Rachel Menashe-Dor for help during field
work. The study was funded by the National Science Foundation (grant EAR-
0409048), the Southern California Earthquake Center (based on NSF cooperative
agreement EAR-8920136 and United States Geological Survey cooperative
agreement 14-08-0001-A0899), and the General Directorate of Mineral Rrsearch and
179
Exploration of Turkey (MTA), Ankara. We thank Kurt Frankel for useful comments
on the morphometrical analysis.
180
Chapter 4: Damage characterization in sandstones along the Mojave
section of the San Andreas Fault with a new method: initial
results and implications for the depth and mechanism of
dynamic rock pulverization
Co-authors: Yehuda Ben-Zion
1
, Judith S. Chester
2
, Jim Brune
3
and
Thomas K. Rockwell
4
1. Department of Earth Sciences, University of Southern California, Los
Angeles, CA 90089-0740, USA.
2. Department of Geology & Geophysics, Texas A&M University, College
Station, TX 77843-3115, USA.
3. Nevada Seismological Laboratory, University of Nevada, Reno, NV
89557, USA.
4. Department of Geological Sciences, San Diego State University, San
Diego, CA, 92182-1020, USA.
Abstract
Following theoretical calculations that suggest shallow dynamic generation of
rock damage (Ben-Zion and Shi, 2005), we evaluate the damage content in
sedimentary rocks that have never been buried deeply while being displaced along
the San Andreas Fault (SAF). Many of the examined sandstones in the vicinity of the
Mojave section of the SAF are immature and display minimal or complete absent of
181
shear. For the analysis of damage in the sandstones we use a new method that
compares the original perimeter length of a grain to the total perimeter length of its
fragments, applied to a statistically representative population of grains from each
sample. We employ this method on samples from the Juniper Hills formation, a
tectono-stratigraphic unit that has been deposited adjacent to active strands of the
SAF system. The results delineate a damage zone of about a 100 m on the southwest
side of the SAF and suggest, together with other considerations, that dynamic
damage generation can occur very close to the Earth surface. When we apply the
method on three mutually perpendicular sections of sample collected 10 m from the
fault we observe an anisotropic damage pattern. In addition we observe preferred
orientation of microfractures and many microscale damage elements associated with
grain contact pressure. Those observations are compatible with failure in a
compressional field. The orientation of microfractures in a sample near the fault is
normal to the SAF and leaning to verticality, in agreement with modeling predictions
for the possible orientation of maximum compressive stress during cyclic loading
associated with slip events on rough frictional fault surface (Chester and Chester,
2000). A change in the preferred orientation of microfractures between this sample
and a more distant sample indicates a variability of the stresses throughout the
damage zone that may be associated with strong dynamic reduction of normal stress
or fault opening.
182
Introduction
Fault zone rocks with their protolite fabric largely preserved but with
significantly reduced grain sizes, yielding a powdery texture, are termed Pulverized
Rocks. Such rocks from the San Andreas Fault (SAF) zone were studied recently by
Wilson et al. (2005) and by Dor et al (2006b). Wilson et al. (2005) have shown that
the pulverized Tejon Lookout Granite in Tejon Pass lacks significant amount of
weathering products, suggesting that the pulverization occurred mechanically. They
also showed that motion on mesoscale faults in this outcrop is compatible with fault-
normal extension and not with SAF-parallel displacement. Dor et al. (2006b) have
mapped the distribution of crystalline pulverized rocks along the Mojave section of
the SAF and found that they occupy a ~100 m wide tabular zone parallel to the
slipping zone of the fault. They presented a line of evidences suggesting that the
observed pulverization occurred at shallow depth (~<3 km). Nevertheless, there are
not yet sufficient observational constraints for the depth range in which dynamic
generation of damage occurs.
In this paper we examine characteristics of damage in sandstones that were
never deeply buried while been displaced along the SAF, with the purpose of
providing constraints on the likely depth of dynamic damage (and pulverization)
generation and the possible damaging mechanisms. We discuss mechanisms within
the framework of dynamic earthquake ruptures because the entire deformation
history of those rocks occurred, with high probability, in association with slip on
strands of the SAF system. We present a new image analysis method for the
183
evaluation of damage content in “dirty” sandstones where particle size measurements
provide only partial results.
We are first concerned with the depth in which damage occurs. Theoretical
considerations based on lab-constrained damage rheology (Lyakhovsky et al., 1997,
2005; Hamiel et al., 2004), analytical considerations (Rice et al., 2005), and
numerical simulations of spontaneous damage generation during dynamic rupture on
a material interface (Ben-Zion and Shi, 2005), indicate that pervasive damage is
unlikely to occur at a depth below the top few km of the crust. The shallow-
generated damage about faults separating different elastic materials at depth is
expected to have an asymmetric pattern correlated with the velocity structure due to
a preferred propagation direction of ruptures (Ben-Zion and Shi, 2005). These
predictions are compatible with properties of the pulverized rocks layer observed by
Dor et al. (2006b). In addition to the shallow inferred pulverization depth, the layer
appears to be shifted to the northeast side of the fault, sharing the same sense of
asymmetry with smaller scale damage products mapped by Dor et al. (2006a) along
the same fault section. The northeast side is the block with faster seismic velocities at
depth (e.g. Fuis et al., 2003). To further assess the shallow damage structure of the
SAF and provide constraints on the likely pulverization depth, we examined the
spatial extent with respect to the SAF and type of damage in sedimentary rocks from
the Plio-Pleistocene Juniper Hills formation. The presence of damage in sandstones
that were never deeply buried similar to the damage found in pulverized crystalline
184
rocks and in association with the trace of the fault requires that dynamic generation
of damage occurs shallow in the crust.
Our second concern, the likely pulverization mechanism, is addressed by
examining properties of microscale damage found in the studied sandstones. Brune
et al. (1993, 2001) suggested that earthquake rupture is associated with fault-normal
loading and unloading due to a dynamic reduction of normal stress. This effect will
create fault-normal component of stress, both compressional and tensional. The tip of
ruptures propagating on a material interface may be associated with an absolute
tension due to fault opening (Ben-Zion, 2001; Ben-Zion and Huang, 2002),
producing strong damage typified by tensional features preferably on one side of the
fault. Failure of rocks in tension may be accommodated by isotropic expansion that
will lead to the production of many dilatational cracks and damage fabric without
preferred orientation (Sammis and Ben-Zion, 2007). The apparent lack of distortion
of magmatic fabrics at all scales observed for pulverized crystalline rocks seems to
be compatible with such a failure mechanism. Tensile stresses and opening may
develop locally due to slip on an irregular fault surface (Chester and Chester, 2000).
Conversely, failure of the rocks in compression should be manifested with preferred
orientation of microfractures that is dependent on the orientation of the off-fault
stresses during rupture passage. Tensile cracks can still be produced in a
compressional field but they are expected to have a preferred orientation.
Mechanisms that may produce microfractures with preferred orientations are
summarized by Wilson et al. (2003) and include the Anderson’s theory of faulting
185
(Anderson, 1942); fault growth model (e.g. Scholz et al., 1993; Vermilye and Scholz,
1998) that may apply also for the propagation of successive ruptures on a healed
fault surface; and wear models emphasizing cyclic loading of the rocks due to
collision of rough surfaces during slip events and development of stress
concentrations (Scholz, 1987; Chester and Chester, 2000). Each of these models
predicts unique orientation of microfractures with respect to the fault surface (e.g.
Figure 1 of Wilson et al., 2003). Here we attempt to resolve whether the micro-scale
damage contained in the Juniper Hills formation has an anisotropic fabric or not, and
if it is anisotropic, which mechanism is compatible with the fabric of the damage.
Our observations show that the Juniper Hills sandstone contains considerable
amount of damage sharing similar features with damage found in crystalline rocks
and over the same length scale, supporting shallow pulverization depth. However,
the fabric of this damage is anisotropic and apparently not compatible with absolute
tensional stresses as a sole failure cause, and we discuss its implications for the
mechanism of dynamic generation of off-fault damage.
Geologic setting and sampling locations
The late Pliocene-early Pleistocene Juniper Hills Formation (hereafter JHF) was
named by Barrows (1980; 1985) who found that it consists of material derived from
the Punchbowl formation and several other clasts types not found in older
sedimentary rocks. The formation consists of tectonostratigraphic elongated units
parallel to the SAF and its subsidiary faults in the area. Distinctive JHF unites with
186
unique clasts content are offset 13 to 16 km by the Northern Nadeau fault and 19 to
21 km by the SAF in and adjacent to our study area. Our samples were collected
from the JHF member TQjh, defined by Barrows et al. (1985) as poorly to
moderately indurated fluvial deposits of coarse arkosic sandstone, lesser
conglomerate and thin-bedded shale that commonly exhibit distinct bedding but are
generally poorly sorted. The clasts are sub-angular to well rounded and varicolored.
In a typical exposure, the rocks are largely poorly cemented and frequently
incohesive, with considerable variations in those properties between exposures and
even between adjacent layers. The soft parts of the rock are sometimes powdery even
far from a fault. Therefore, and in contrast to crystalline rocks, the field texture of the
JHF rocks (as well as of other deposits in the area) does not provide a systematic
indication of their damage content.
Exhumation of the SAF in the area was inferred by Dor et al. (2006b) to be
minimal, and certainly below the maximum uplift of 2-4 km inferred for the
Punchbowl fault. While there is no direct evidence for the maximum burial depth of
the JHF rocks, their consolidation state is very poor in most locations, suggesting
minor exhumation. The mapping of Weber (1999) suggests that uplift that started
0.75 to 0.5 Ma in the southeast portion of the San Gabriel mountains (presumably
due to the complex interaction of the San Andreas, San Jacinto and the Cucamonga
faults in this area, e.g. Kenny, 2000) had raised beds of the middle (?) Pleistocene
Shoemaker Gravel formation in the Big Pine area 1100 m with respect to its beds
near Big Rock Creek. The easternmost zone of our sampling area is 12 km further to
187
the northwest of Big Rock Creek and 25 km northwest of Big Pine area. Although
the JHF is older than the Shoemaker Gravel formation, its distance from the locus of
uplift (further away from the correlative lower body of the Shoemaker Gravel
formation) and the inference that uplift had intensified toward the end of the
Pleistocene suggest again considerably lower exhumation values for JHF rocks.
Our two sampling points furthest from each other are near E. 106
th
St. and near
the intersection of Pearblossom Hwy and Sierra Hwy, with the first being 16 km
southeast and 190 m above the first one. 22 km further to the southeast, the same
JHF member is exposed near Jackson Lake 790 m above the E. 106
th
St. sampling
point. Hence, the elevation gradient changes from 12 m/km to 34 m/km between
these two sections, compatible with the abrupt change in elevation gain of the SAF
trace 12 km to the southeast, immediately east of Big Rock Creek. This suggests that
uplift rates drop substantially in the vicinity of our sampling area with respect to the
locus of uplift further to the southeast, supporting minor exhumation in our sampling
environment. With ~10 km of absolute motion on each side of the fault (to achieve
~20 of relative displacement), bodies of JHF on the southwest side of the fault have
not gone through this zone of abrupt change in topographic gradient.
Figure 4.1 shows the distribution of some of the JHF bodies southwest of the
SAF with our sampling locations. Our qualitative and quantitative description is
based on 13 JHF samples collected at 7 stations along a fault-normal traverse. We
compare this description to observations from 3 samples of the Hungry Valley
188
formation collected at distances of 125 to 3380 m from the SAF in the northwest
portion of the Mojave Desert and in Ridge Basin.
Figure 4.1: Working area and sampling locations. Inset: digital elevation model of the
southwestern Mojave with the SAF clearly seen as a linear scarp separating the San Gabriel
Mountains from the Mojave Desert floor. The sampling area is marked (includes the area in the larger
map + location of samples 6a and 33). The larger map (after Barrows et al., 1985) shows the
distribution of some of the JHF bodies in the area with the location of our sampling points. For
samples 6a and 33 we show only the distance to the fault, not their geographic location. Note that the
JHF here is bounded by the SAF and the N. Nadeau fault.
Research approach
The type of microscale damage we observe is characterized by in-situ shattering
of grains without an apparent distortion of the grain’s shape. Many of the grains are
broken but their fragments still fit together (see detailed description below). Since we
189
work with thin sections under a microscope our work is done in 2D and we adopt the
following approach: for each grain we compare the length of the grain’s perimeter to
the total length of the perimeter of all the fragments that belong to this grain. For
each thin section we evaluate the Factor of Increase in Perimeter Length (FIPL)
based on an analysis of a representative population of grains. We analyze grains with
pre-damage size of about 50 microns or larger, and ignore rock fragments. Visual
characterization and analysis of individual grains is a direct analysis method that
allows good control on the process and reliable interpretation of the results. We
assume negligible brittle failure in the matrix (due to poor consolidation and small
grain size) and our analysis do not account for its damage content.
We resolve between isotropic and anisotropic damage fabric by analyzing three
mutually perpendicular thin sections. If the intensity of damage varies between the
three orientations, damage is anisotropic, implying that the rock failed in a
compressional field. Identical or very similar damage fabric between the three
orientations support failure of the rock in a tensional field (given that the
crystallographic axes of feldspars in sandstones have no preferred orientation). We
also check for the circularity of fragments of grains: A value of 1.0 indicates a
perfect circle. As the value approaches 0, it indicates an increasingly elongated
particle. Circularity approximates the aspect ratio of grains. High circularity values
observed in the three orientations support uniform expansion (however, feldspar
tends to break along crystallographic planes, and therefore circularity of 1 is unlikely
even in a pure tensional field). In addition, we measure in some of the samples the
190
orientation of a large population of microfractures and check for preferred
orientations. These measurements are complemented by a description of the mode of
fractures and other relevant properties.
Non-direct approaches of evaluating damage by means of particle size
distribution include the laser analyzer and the settling tube methods (e.g. Wilson et
al., 2005; Sisk et al., 2006). Sedimentary rocks may include wide or multi-peak
original distributions with potential large variations in space (even within one
exposure). They may also include original micron to sub-micron scale size grains,
including clay minerals, which can not be easily distinguished from clay particles
that were produced by in-situ weathering. Thus, it is difficult to interpret particle size
measurements of sedimentary rocks in terms of damage. We are therefore using
optical and electron microscopy, the most direct techniques for the analysis of the
microscale fabric of those rocks and we describe our methods below.
Methodology
Samples extraction and preparation
Samples in the different outcrops were extracted from layers with similar grain
sizes and colors, and as far as possible from meso scale faults, veins and other
structural disturbances. To further minimize differences between the samples, we
chose rock pieces from massive parts of the layers, avoiding fine layering,
exceptionally large clasts or other strong fabrics (Figure 4.2). The cementation
191
conditions of rocks in some of the sampling locations presented challenges for the
extraction of coherent and oriented rock pieces. Those rocks often tend to
disaggregate upon a slight pressure. After choosing the piece of rock for extraction,
an orientation of an exposed flat surface was measured and the strike and dip
directions were marked with sharpie on that surface. The sample was photographed
in-situ and was then extracted with great care to maintain its integrity and prevent
distortion of its fabric. It was then wrapped in a soft tissue paper, placed in a bag and
carried to the lab. For the impregnation we soaked the samples in a low viscosity
epoxy (Epo-Tek 301©) inside a vacuum chamber. The samples were then sliced into
blanks, and in some cases we sliced three mutually perpendicular blanks from one
sample. Microprobe-quality oriented thin sections were made in an external facility.
192
Figure 4.2: Exposure of the JHF 10 m southwest of the SAF in the Littlerock paleoseismicity
site. The fault has clear geomorphological expression here as a linear trough (above the exposure),
marked by a line of Juniper trees. In the inset: one of the samples taken from this outcrop, still in-situ
with the orientation marks. Note the size and color variability of grains seen on the sample face.
Image analysis
We mapped the original grain boundaries of 170-230 individual grains for each
thin section on transmitted light images while simultaneously verifying the
boundaries of the grains as optically coherent entities using the microscope under
cross polarized light. We chose only grains that are not or only minimally weathered
and avoided large rock fragments and grains divided into many sub-grains. The
digitized grain map was analyzed in the image analysis software “Image J” to yield
193
an image with the grains numbered, and an associated data set that includes the grain
sizes. It is apparent that the majority of the grains are small while the samples area is
dominated by large grains that are frequently more fractured than the smaller grains.
A random selection of grains from the entire grain population would result in a non-
representative subset of mainly small grains that are relatively less damaged, which
may lead to a considerable underestimation of the damage content. Therefore, we
divided the population into 4 size bins and randomly chose 12 grains from each of
the bins (Figure 4.3a).
Figure 4.3: Intermediate and final products of the analysis process: a. 40 grains in four size bins
(color coded) are chosen from a population of mapped grains (light transparent gray). b. An image of
a single grain taken under reflected light, optical microscope. c. grayscale version of the same grain,
masked from its environment. d. a map of the grain’s outline, filled, representing the pre-damaged
state of the grain. e. binary (bitmap) image of the fragments belong to the same grain. Their total
perimeter length is compared to the perimeter length of the “intact” grain in (d).
Each of the selected grains was photographed under reflected light in an X100
magnification (Figure 4.3b). We used this image to generate two products: the first
was a grayscale image of the grain itself masked from the background (Figure 4.3c),
and the second was an image of the exact trace of the grain boundary, which was
194
used as a reference for the original “intact” grain (Figure 4.3d). The image of the
masked grain was transformed into a bitmap (binary) mode using 50% threshold
value and then inverted so that the grain (or its fragments) appears in black and the
background in white (Figure 4.3e). In some of those images, epoxy-filled gaps
between fragments were bright enough to pass the threshold value and appeared as
black spots, adding considerably to the measured total perimeter length. In such
cases we cleaned the images manually from those spots in Photoshop by visually
comparing the bitmap and the original images.
The two images were then analyzed in Image J for their total perimeter length.
The “intact” grain image gave the length of the original perimeter of the grain before
it was fractured (we assume the expansion/dilation of the grain to have negligible
effect on the perimeter length), and the “damaged” grain image gave the cumulative
perimeter length of all the fragments that belong to this grain. The Factor of Increase
in Perimeter Length (FIPL) for the entire sample is an area-weighted average of the
FIPL of all the grains that were analyzed in that sample (in calculating the average,
grains were given weight according to their original area).
Fracture orientation measurements
The fractures observed in our samples are mostly mode I, although some of
them are associated with minimal (negligible) offset of fragments. Transganular
fractures are not common and most of the fractures are contained within one grain.
The following are types of microfractures observed in our samples, distinguished
195
according to their relations to grain size and to their hierarchical internal
arrangement within and with respect to the grain (Figure 4.4):
Type I: Transganular fractures, extending beyond grain boundaries into the
matrix and neighboring grains (not common).
Type II: Fractures that cut an entire grain. They are frequently parallel to each
other, dissecting the grain or part of it to elongated, columnar fragments.
Type III: Fractures that cut at least half of the grain’s width in the direction of
the fracture, and terminate within a grain. Some of them taper into the grain while
others connect fractures to other fractures or to grain boundaries.
Type IV: Considerably shorter than the grains average axis length. Those are
usually connecting fractures that terminate against type I or II fractures. They often
cut elongated columnar fragments into rectangular or otherwise angular fragments,
creating webs in crisscrossing relationship (e.g. category II fractures of Laubach,
1997,).
Type V: Sub-grain boundary and grain boundary fractures.
Type VI: Fluid inclusion plains: those are shown as linear traces of bubbles,
marking the location of healed (or sealed) fractures (e.g. Tuttle, 1949).
The vast majority of all fractures observed optically are joints, associated with
dilation normal to fracture wall. Almost all of them appeared to be open (not healed
or sealed), except fractures of type VI. For our analysis we measure the orientation
of fracture types II-IV because these are less likely inherited from the source rock.
While this is true also for type I fractures, we ignored them in our orientation
196
measurements because they are not common and would not stand out as a group
statistically. We also measure the orientation of fractures type VI. Type V fractures
are ambiguous for interpretation and are curved in most cases, making their strike
determination problematic. In the samples selected for fracture orientation analysis,
we measured the orientation of fractures in all the grains chosen for the perimeter
length analysis.
Figure 4.4: Fracture types (see text for complete definition): a. Type I – transgranular fractures. Not
common in our samples and therefore were not considered for the fracture analysis. b. Types II (red) –
fractures that cut the entire grain; Type III (blue) – cut at least half of the grain’s length in its
direction; IV (green) – short, connecting fractures. c. Type VI – fluid inclusion plains, typically cross-
cut by other (open) types of fractures. d. Type V – sub-grain and grain boundary fractures. Those
were not included in our analysis because they are hard to interpret and to measure.
197
Observations
Description of the Juniper Hills host rock (Based on a sample taken 670 m from the
fault)
At the outcrop scale the rock is light brown to yellow arkosic sandstone. The
layers differ from each other and hence defined by grain sizes. In this location as
well as in other sampling locations the layers are tilted. The rock contains calcite
veins and occasional meso scale faults. The rock is highly friable and can be
disaggregated fairly easily using a rock hammer. A hand sample can be pressed by
hand to yield powdery texture, suggesting poor cementation, significant amount of
silty (or finer) matrix and fractured macroscopic grains. The visible grains are in the
mm-scale, supporting 1-10 cm wide rounded pebbles in various colors. This outcrop
is representative of the Juniper Hills formation in most sampling locations.
Description of thin section: in most of its area the sample is grain supported with
some matrix supported area in places. The sample is composed of angular to sub-
rounded quartz and feldspar grains with heterogeneous damage: most of them are
intact while some include few noticeable fractures. A minority of the grains is
heavily fragmented (see below a description of damage features). Many of the quartz
grains contain fluid inclusion plains. The population of quartz-feldspar grains spans a
wide distribution of sizes between about 50 (and sometimes less) microns to 700
microns with maximums at 50-100 microns and at 400-700 microns, and with some
very large grains exceeding 1000 microns in diameter. The smaller grains tend to be
198
more angular. The sample contains also 5-10% of rock fragments. Those tend to be
larger than the single-crystal grains and are very heterogeneous in their crystal
content (size and type) and internal fabric. Their typical size is 1.5-3 mm but some
are larger. The matrix is partially opaque to transmitted light, but we could identify
an average visible grain size of 5-20 microns including, in addition to quartz and
feldspar fragments, grains of micas and oxides.
The rock fabric and composition may vary between exposures and even between
layers within one exposure.
Common damage features present in all/most samples (Figure 4.5):
1. Heavily fragmented grains: the grain is fractured into many fragments
significantly smaller than the grain itself with large increase in the total
perimeter length. Frequently the fragments have high aspect ratio (high
circularity values)
2. Partially fractured grains: part of the grain is fragmented and part of it is
intact. Some partially fractured grains have one or just few long
fractures.
3. Deformed mica crystals: mica crystals kinked and squeezed due to a
displacement of another, more rigid grain. Suggestive of operation of
compressional forces.
4. Impingement (Hertzian) fractures: fractures that appear to emanate from
grain boundaries at a high angle. If more than one fracture is present, the
199
fractures tend to be arranged in a parallel, radiating or cone geometries.
The apex of the radiating pattern (or the fracture-grain boundary
intersection) is at a grain-grain contact, and fracturing can be clearly
interpreted as the result of a contact pressure between the grains. Those
fractures tend to form elongated columnar fragments that are often
divided into shorter fragments by intermediate and short connecting
fractures (possibly due to bending of the elongated fractures). Those
long, intermediate and short fractures correspond to fracture types II, III
and IV described above, respectively. Since the observations are done in
2D, most of the contact points from which fractures are radiating are not
observed and therefore many more of the observed fractures probably
result from impingement. These types of fractures were produced in the
short term compaction creep experiments of Chester et al. (2004).
In general, damage is highly heterogeneous in all samples, with some grains
fragmented down to the micron scale while others remaining intact. Visually, fault-
parallel and other shear components are apparently absent at all scales: sedimentary
fabrics are intact and even the most fractured grains, with few exceptions, preserve
their original outline, with the fragments appearing to fit together in a hierarchical
fashion.
200
Figure 4.5: Microscale damage features: a. Partially fractured grain. b. Heavily fractured grain.
Note the three columns of fragments: may be caused by impingement, like in (e, f). c, d. Cross
polarized (c) and transmitted (d) photos of bended and sheared mice grain, in this case clearly due to
an impingement between the two grains in contact: note the fractures radiating from the grain on the
left. e, f. Cross polarized (e) and reflected (f) photos of fractures radiating in a Hertzian pattern from a
contact zone between two grains.
201
FIPL measurements
Accuracy of the method: The statistical significance of the method was tested on
a large population of analyzed grains from sample 8E-b. This sample was taken 10 m
from the fault and is highly variable in its grain sizes and their fracture content. We
mapped 186 grains on the thin section and analyzed 104 of them. The area-weighted
average FIPL for all the 104 grains is 8.14 and the area-weighted standard deviation
is 5.34 (Table 4.1). A histogram for the FIPL values of all the 104 grains is presented
in Figure 4.6. FIPL values are plotted against grain sizes in Figure 4.7, showing that
the grain size limits the FIPL values although there is no strong correlation between
them (R
2
= 0.34).
Figure 4.6: Distribution of FIPL values for 104 grains in sample 8E-b.
202
Figure 4.7: FIPL as a function of original grain area for 104 grains in sample 8E-b. Although
FIPL values have no strong correlation with the grain sizes (note small R
2
value), grain size seems to
control the upper limit on the FIPL values.
10 sets of 40 grains were chosen randomly from the 104 grains population. Their
area-weighted average FIPL and area-weighted standard deviations are presented in
Table 4.1. We performed t-test in order to verify that the sets chosen are indeed
representative of the larger population. None of the sets is statistically different from
the parent population (the P values of all of them is larger than 0.05, and for some of
them it is closer to 1; all the sets pass Shapiro-Wilk and Shapiro-Francia W
normality tests with over 95% probability, see Appendix 1), suggesting that the
sampling method captures the properties of the entire grain population (assuming
that the 104 grains are representative of the grains in the sample).
203
Table 4.1: Area-weighted average FIPL and standard deviation for the 104 grains population
analyzed from sample 8E-b and for each of the 40-grains sets chosen from this population. T-test
results show no statistical difference between each of the sets and the entire population.
Results of FIPL measurements along the traverse: qualitative examination of the
thin sections before the measurements suggested that fracturing and fragmentation of
grains is considerably intense in the samples taken 10 m from the fault, high in the
sample taken 65 m from the fault, and subsides significantly in the samples further
away from the fault. It was practically hard to distinguish between the samples taken
115 m from the fault and farther in terms of the damage content. This already
suggested, qualitatively, that the width of the damage zone on this side of the fault is
of the order of 100 m.
Our measurements are summarized in Table 4.2. Confirming the qualitatively
observed pattern, the FIPL is highest in the sample taken 10 m from the fault and is
still high in the sample taken 65 m from the fault. Farther away from the fault the
FIPL decreases but there is apparently no strong drop in its values. A t-test shows
that the samples taken at 10, 65 and 115 m from the fault are statistically different
from each other, while the samples taken at 115, 335 and 1930 m from the fault are
statistically indistinguishable from each other in their FIPL values distribution. All
sample results passed Shapiro-Wilk and Shapiro-Francia W normality tests with over
95% probability (see Appendix 1). These results suggest that the damage gradient is
40 grains
Set 1 2 3 4 5 6 7 8 9 10
Av.
All
104
FIPL
7.40 8.06 7.09 6.10 8.72 8.40 7.99 7.15 8.92 7.29
7.71
8.14
A. weighted
STDV
4.45
5.84
4.35
3.48
6.01
5.25
5.03
4.37
5.02
4.18
4.8
5.34
t-test
P
value
0.8
0.92
0.86
0.86
0.82
0.95
0.72
0.28
0.42
0.54
204
significant in the ~100 m distance from the fault, after which the damage gradient
becomes insignificant. Fragments in the 10 m sample appear to have a slightly higher
circularity values. The entire visual data set appears in Appendix 2.
Table 4.2: summary of measurements done on the JHF samples.
Sample 8E-b 31 28 27 6a
distance from fault 10 65 115 335 1930
FIPL 8.14 5.21 2.42 3.81 1.53
FIPL (not weighted) 5.61 3.85 2.16 2.43 1.41
Circularity
1
0.57, 0.74
2
0.58
3
0.54 0.54 0.56 no data
Thin section
inclination
28 70 80 46 83
t-test P value
4
0.00 0.00 0.4 0.38
1. Average circularity of all fragments from all measured grains.
2. For thin section 8E-a, taken from the same sample.
3. For thin section 8E-c, taken from the same sample.
4. For each sample with respect to the next sample closer to the fault.
Isotropy of damage fabric in sample 8E: Table 4.3 shows measurements done on
three mutually perpendicular sections cut from sample 8E, taken 10 m from the fault.
Sections 8E-a have over two-thirds the FIPL value of section 8E-b, and a t-test
indicates that they are statistically identical. Examination of the data sets shows that
in the 104 grain population of sample 8E-b there are few grains with exceptionally
large FIPL, but the majority of the grains have FIPL values similar to those in the 43
grains population of sample 8E-a. The exceptionally large FIPL values at the tail of
the distribution affect the area-weighted FIPL value of the entire population but do
not affect its other statistical properties significantly. Therefore, the two sections can
be considered one population although the damage content in 8E-b is larger.
Conversely, both 8E-a and 8E-b are distinctly different from section 8E-c (different
205
with >95% confidence) in their damage content. This is likely related to the relations
between the orientation of the thin section and the orientation of microfractures in
this sample (see Table 4.4 below and Figure 4.11 in the discussion).
Table 4.3: summary of measurements done on samples 8E.
1. Average FIPL for all
10 sets.
2. With respect to the
other two sections.
Comparison with observations from the Hungry Valley formation. The bulk of
the Hungry Valley formation (HVF) is widespread in Ridge Basin and is truncated
on the north by the SAF. Tectonic slivers of the formation appear north of the SAF
west of Three Points. The HVF is highly variable in texture and composition
(Barrows et al., 1985). Both age and displacement of the formation are poorly
constrained and highly debatable in the literature. Its age may range between late
Miocene (Miller and Downs, 1974) to Pleistocene (Kahle, 1979) and its
displacement is of the order of 12 to 27 km (Barrows et al., 1985), or according to
Ramirez (1983) its values may be at the order of 220 km. There are no direct
constraints on the maximum burial depth of the HVF but it is probably not
significant as there are indications for only moderate exhumation of the area since
the late Miocene. This is suggested by depositional contact between rock formations
of successive ages and only minor incision (Dor et al., 2006b). In addition,
consolidation of the formation in most places is poor.
Sample 8E-a 8E-b 8E-c
Orientation 62/060 28/240 90/150-330
FIPL 5.79 8.14 (7.71)
1
4.79
FIPL (not
weighted)
4.53 5.61 3.35
Area Weighted
STDV
3.14 5.34 2.42
t-test P value
2
8E-b: 0.23
8E-c: 0.03
8E-a: 0.23
8E-c: 0.00
8E-a: 0.03
8E-b: 0.00
206
We compare three samples of the HVF: 22a collected at 125 m northeast of the
fault west of Three Points; 24 collected at 1055 m southwest of the fault in a road cut
of Hwy 138 near its intersection with the I5; and 25a collected in the heart of the
Ridge Basin 3380 m southwest of the SAF and in the immediate vicinity of no other
known faults. The HVF west of Three Points is described by Barrows et al. (1985) as
extremely heterogeneous fine to course grained arkosic sandstone, poorly to
moderately cemented. In places it is well indurated and bedded and in places it is
massive, poorly to moderately sorted. Other lateral variations occur. It is composed
typically of subrounded quartz and feldspar grains. Large clasts are varicolored sub-
to well-rounded granitic, volcanic and some metamorphic pebbles and cobbles. We
find this description appropriate for our 22a sampling location.
The other sampling localities (which are not described in such details in the
literature) are very similar to each other and differ from the above in the following:
the rock is well-bedded and fairly sorted sandstone with less lateral variability
compare to the Three Points location. The samples were collected from very poorly
cemented layers of uniform material with mm scale size visible bright-colored clasts.
Figure 4.8 shows cross polarized and reflected light photomicrographs of
samples 25a and 22a. In sample 25a the grains are sub-rounded to sub-angular, with
quartz and feldspar as main constituents. The rock appears to be quite porous with
significant volume of calcite cement although the sandstone is grain supported. The
grains are intact to weakly fractured as it is apparent from both images of the sample.
Nevertheless, some post-cementation compaction has occurred as evident by the
207
appearance of open Hertzian fractures and calcite twinning (marked by green
arrows). The origin of these compression features is unknown but most likely is
related to the extensive folding occurred in Ridge Basin rather than to faulting on the
SAF. It thus gives some sense about the background damage those rocks
accumulated regardless of SAF or other fault activity.
Figure 4.8: Photomicrographs of Hungry Valley formation. Cross polarized (upper panels) and
reflected light (lower panels). The samples were taken 3380 m from the fault (left) and 125 m from
the fault (right). The far-field sample is only mildly damaged but still contains some post-depositional
compression features such as open Hertzian cracks and calcite twinning of the cement (green arrows).
The near-field sample is intensely damaged although grain boundaries are not distorted. The level of
damage at 125 m from the fault in this sample appears to be much higher than the damage level in the
10 m JHF sample.
208
The mineral composition of sample 22a as it appears in the upper-right image of
Figure 4.8 is similar to that of sample 25a, but its other characteristics are quite
different. The most striking difference between the samples is intensity of fracturing.
Compared to the relatively intact sandstone in sample 25a, the sandstone in sample
22a is heavily fragmented (as clearly seen in the reflected light image), although the
original grain shapes are not distorted. In addition, the amount of cement in sample
22a is considerably lower than that of sample 25a. Original variations in the porosity
between the various parts of the formation are possible, but such a significant
difference may be attributed, at least partially, to a fault related volumetric strain.
Sample 24 shows intermediate characteristics between the two other samples in
terms of its fragmentation intensity and cement content. This description is
representative of those locations based on comparison with several other samples
taken from the same or near-by locations.
In comparison with the JHF, the intensity of damage in the HVF is substantially
higher. At 115 m from the fault, the damage in the JHF sample drops closer to the
far-field damage content than to the near-field damage content (Table 4.2), while at
125 m the HVF sample is dramatically more damaged compare to the far-field
background damage level, and apparently more fragmented than the JHF samples 10
m from the fault. The type of the damage, however, is apparently rather similar
between the formations. The possible reasons for those differences are addressed in
the discussion.
209
Orientation of microfractures
The results of the fracture orientation measurements are displayed in Table 4.4.
Fracture types II, III and IV that were measured in samples 8E-b and 65
(corresponding to 10 and 65 m from fault, respectively) show preferred orientations.
In sample 8E-b those fracture types trend preferentially to the northeast with the 95%
confidence intervals spanning +/- 16 to 26 degrees. In sample 31 fracture type II
trend preferentially to the east-northeast with a minor set trending to the northwest,
and fracture types III and IV trend preferentially to the northwest with the 95%
confidence intervals spanning +/- 44-54 degrees. Despite the scattering in strike
direction, the two samples show preferred orientation of fracture types II, III and IV
that is distinctly different between the samples. Fracture type VI (fluid inclusion
plains) show 95% confidence interval of 90 degrees, i.e. no preferred orientation in
both samples. Those fractures are likely inherited from the source rock and were not
formed in the process that created the other fracture types.
210
Table 4.4: Rose diagrams showing the distribution of orientations for the various fracture types
from samples 8E-b and 31 with corresponding statistics. Red line marks the orientation of the SAF
in the area.
Discussion
Precision of measurements
To evaluate the precision of the semi-automated image analysis technique, we
compared the total perimeter length of a damaged grain obtained by the method
described above and by a manual digitization of the perimeter of all the fragments
belonging to that grain. The automated and manual digitized grain images are shown
211
in Figure 4.9. Figures 4.9b and 4.9d include noise (epoxy spots that passed the
threshold filter) while Figures 4.9c and 4.9e show the same images cleaned from that
noise. The total perimeter length of all the fragments in the manually digitized image
(the one cleaned from the noise) as well as the resultant FIPL value are 8% higher
than the corresponding values obtained for the image digitized automatically by
Image J. An underestimation of the total perimeter length is expected since during
the conversion of the image to a binary mode (bitmap) some details are lost due to
the segmentation (thresholding) process, especially along the edges of particles. This
will affect mainly the smaller particles observed optically and in most cases will not
have a dramatic effect on the result. The insets at the bottom of Figure 4.9a
demonstrate the different amount of details captured by the manual vs. automatic
digitization. In the automatically digitized image (center) some particles are lost and
others are shrunken with respect to the real image (left) and to the manually digitized
image (right). Note that the particles in the real image include dark edges that were
not captured in the automated digitization but were mapped as part of the particle’s
area in the manual digitization. By visually comparing grayscale and binary images
of the same grains, an 8% of underestimation of the total perimeter length seems to
be a reasonable value.
212
Figure 4.9: Possible artifacts shown on a single grain. a. A gray scale image of a grain from sample
8E-a. The insets at the bottom show small section of that image (left) with the corresponding
automatic (middle) and manual (right) digitized images. The automatic digitization tends to miss
some details due to the thresholding process. b, c. An automatically digitized fragment map of the
image in a, used for analysis in Image J. (b) includes noise and (c) is manually cleaned from the noise.
Holes are not measured by the program. d, e. Same as in b and c but manually digitized. FIPL values
appear in parenthesis.
Figure 4.10 shows scanning electron microscope (SEM) images, exposing small
particles that were not observed in the 100X magnification optical microscope
images. The diagonal insets in the upper panels show the location of the SEM images
in the grain map of Figure 4.9e, allowing a comparison of the amount of details
captured between the SEM and the optical microscope images. The fragments
observed under reflected light do not appear to be further fragmented. They are
smooth (due to the probe-quality polishing), have sharp edges, and the fractures
between them are filled with epoxy. Particles that are not observed optically, with the
213
optical limit (of X100) at about 5 microns, appear to be restricted mostly to those
fractures, floating in the epoxy matrix. Some zones in the epoxy are densely
populated with those small particles and some zones appear to be completely free
from particles, corresponding to dark and bright zones inside the fractures in Figure
4.9a, respectively. We interpret those small particles to be mainly small rock
fragments but also possibly particles from the sandstone matrix. During handling of
the sample, when it is extracted in the field and when it is placed in the container for
epoxy impregnation, it yields upon a slight touch silty or clayey size whitish powder.
This powder inevitably mix and flow with the epoxy while it is filtered through the
sample in the vacuum chamber. The inhomogeneous spatial distribution of particles
in the epoxy is probably due to a non-uniform mixing and a complex flow pattern.
214
Figure 4.10: SEM images of frames from the grain in Figure 4.9, showing many fragments
floating in an epoxy matrix that were not detected by the optical microscope analysis. They are
restricted to certain zones within open fractures between larger fragments that were mapped. Small
black and white diagonal frames show the same frame from the manually digitized fragment map
(Figure 4.9e) and correspond to the locations marked on the grain in Figure 4.9a.
It is very difficult, if not impossible, to evaluate the origin of each of those small
particles. They may be fragments from the grain in which they are currently
contained, they may be fragments from another grain, or they may be part of the
original matrix of the sample. It is also likely that the amount of grain fragments
within the fractures was originally larger and was reduced due to the filtration of the
epoxy through the sample. Assuming all the particles are rock fragments, and hence
their perimeter length should be measured for our analysis, we manually digitized the
small particles in Figure 4.10b, lower panel (white lines). We estimate the area of the
215
particles-epoxy mix inside the fracture in the lower panel of Figure 4.10b to be
roughly 1/20 of the total area of the dark fracture zones in this grain, where the
fractures are filled with particles-epoxy mix. We calculated the perimeter length of
the particles in this image, multiplied it by 20 and added it to the total perimeter
length measured for the entire grain. This brings the grain’s FIPL from 6.94 to 9.41.
Uncertainties regarding this result are due to the unknown original distribution and
composition of fine material inside the fractures, that was altered by the flow of
epoxy and possibly also by sample handling. If we first apply the 8% correction and
then add the perimeter of the small particles, the FIPL goes up to 9.97.
The adjustments suggested above increase the FIPL of this particular grain by
44%. Calculating similar adjustments to other grains may result in different values of
increase (or even decrease) of the FIPL calculated based solely on the automated
process, since the grains and their damage patterns are very different from each
other. In a continuing study we hope to explore those adjustments and to find out if
such adjustments can be applied systematically to a large population of grains,
thereby increasing the accuracy of the measurements. At present, we conjecture that
the automatic procedure underestimates the actual FIPL, but that the true and
measured FIPL are different in less than an order of magnitude.
Structure of the damage zone (on the southwest side of the SAF)
The reduction in FIPL values between the samples in the first 115 m southwest of the
fault, with the samples been statistically different from each other, suggests a true
216
gradient in the intensity of damage over that distance. The disappearance of damage
gradient between the samples at 115, 335 and 1930 m from the fault, indicated by
their relatively low and similar FIPL values and by the lack of statistical difference
between them, suggests that the width of the zone that accommodates SAF-related
damage on the southwest side of the fault is of the order of 100 m. This length scale
for the (half) width of the damage zone was observed (including in the distribution of
microscale damage) for many other fault zones, including faults of the SAF system
e.g. Punchbowl, San Gabriel by Wilson et al. (2003) and Chester et al. (2004).
Furthermore, the length scale of ~100 m is also the average width of the crystalline
pulverized rocks layer (Dor et al., 2006b). Although the later includes both sides of
the fault, and despite the apparent shift of the layer to the northeast side of the fault,
the zone of pulverized rocks was found by Dor et al. (2006b) to be wider in our
working area. Therefore the width of the damage zone inferred in our study likely
overlaps the layer of crystalline pulverized rocks.
Mechanical interpretation of damage fabric
The observation of grains shattered in-situ, preserving their original outlines, is
overwhelmingly systematic and common to both crystalline and sedimentary rocks
spatially associated with the SAF. Such damage pattern seems to be consistent with
the apparent lack of shear and with failure of the grains under tension. However,
failure under an absolute tension and failure of individual grains in tension under
compressional field should result in distinctly different properties of the damage. The
217
most important criterion is the lack of preferred orientation of damage fabric
generated under an absolute tension. Our observations are consistent with failure of
grains in a compressional field. First, we observe numerous damage elements typical
of grain contact pressure, in particular impingement (Hertzian) fractures; second,
mapping of several hundreds of fractures in two samples show that open fractures,
which are likely associated with current or recent fault activity, have a preferred
orientation that is especially strong near the fault; and third, circularity values in the
range of 0.54-0.58 (with the exception of section 8E-a with circularity of 0.74)
suggest that the average fragment is elongated, in a shape not expected from uniform
expansion of the material.
Anisotropy of damage fabric finds an expression in sample 8E. Figure 4.11
shows schematic illustration of sample 8E configured to represent its orientation in
space with letters indicating the faces sliced as thin sections 8E-a, 8E-b and 8E-c and
numbers indicating FIPL values. The blue spots represent schematically the 2D
Figure of grains on the thin sections sliced from these faces. Face a is dipping 68°
toward 60°, face b is dipping 28° toward 240° and face c is vertical, striking 60°-
240°. Faces a and c contain 71% and 59%, respectively, the damage content of face
b, although face a is statistically indistinguishable from face b in the distribution of
the FIPL results. This observation is compatible with the damage fabric been
anisotropic in this sample. While such anisotropy may simply be the result of a
statistical inhomogeneous distribution of fractures in space, the sense of the
anisotropy appears to be related to the preferred orientation of microfractures: the
218
mean direction of fracture types II, III and IV is between 27° to 42°, with 95%
confidence interval of +/- 16° to 26°. These directions are oblique with an acute
angle to face c. The relatively low damage content on face c is compatible with the
fractures leaning toward verticality, which is expected in a strike slip environment.
In this case, face c does not expose many of the fractures because their planes are
quasi-parallel to it. Face b, which is inclined at a low angle, should expose most of
the fractures unless they are horizontal or shallowly inclined to the west-southwest.
Face a has an angle of about 57° with the mean direction of the largest fracture
population (mean strike 27°, type IV fractures, 428 data), and since it has a steep
inclination it may not expose many fractures that have smaller-than-the-mean angle
with it and are inclined eastward. The correlation between section angles and FIPL
values suggest that the fractures are overall leaning toward verticality.
219
Figure 4.11: A schematic illustration of sample 8E positioned in its in-situ configuration in space
with the three faces used for the preparation of the thin sections (marked as a, b, and c). Blue spots
illustrate grains and red lines show fractures. The mean directions for fracture types II, III and IV are
42°, 54° and 27°, respectively, and we infer that they are largely vertical because vertical face c have
a FIPL value that is distinctly lower than that of the other faces.
Implications of results for dynamic rock failure
The JHF was deposited in between and displaced by strands of the SAF system
that are currently or recently active, and therefore its entire damage history is most
likely related to the current faulting regime. We therefore conclude that damage in
elevated levels within ~100 m from the active trace of the SAF is the product of fault
activity, i.e. large SAF earthquake ruptures. Although some grains can survive
transportation while they are partially fractured, the intense fracturing we see with
open fractures that frequently cut the entire width of the grain suggest that fracturing
occurred mostly in-situ and was not inherited from the source rock. The clear
220
preferred orientation of the fractures is an independent evidence that fracturing is
post-depositional.
The JHF was never deeply buried and therefore this damage was generated
shallow in the crust, probably at depths of several hundreds of meters and possibly
within meters from the surface of the Earth. This inference is supported by the lack
sealing or healing of the fractures. For the fractures to remain open, ground water
filtration and temperatures mast have been minimal. Fluid inclusion planes (healed
fractures), that form a small minority in the fracture population, are cut by open
fractures, have no preferred orientation, and are therefore likely inherited from the
source rock.
The nature of damage in pulverized crystalline rocks spatially associated with
the trace of the SAF (Dor et al., 2006b), although not studied systematically so far in
the microscale, appears to be qualitatively similar to that observed in our study. This
may suggest that pulverization can occur very close to the surface of the earth, in
agreement with previous inferences about minimal exhumation in the area and
theoretical considerations discussed in the introduction.
The anisotropy of the damage fabric and the abundance of damage elements
associated with pressure show that absolute tension was probably not the primary
mechanism responsible for the failure of the JHF rocks. The microfractures in
sample 8E-b are preferably orientated normal to the SAF. Chester and Chester
(2000) presented a mechanical modeling of stress and deformation in the vicinity of
a wavy frictional surface and showed that stress is heterogeneous near the fault due
221
to the juxtaposition of geometric irregularities. This is associated with inelastic
deformation within a region close to the fault surface. Within this region and under
loading conditions relevant for the SAF (i.e. with relatively high angle between the
maximum far-field compressive stress and the fault), the maximum near-field
principal compressive stress during cyclic loading associated with slip events is
oriented nearly normal to the fault surface. In such conditions, fractures normal to
the fault surface may form throughout the damage zone of the fault.
Fault opening and significant reduction of normal stress are expected to produce
a change in the orientation of stresses as a function of distance from the fault. This
may lead to a variability in the preferred orientation of microfractures within the
damage zone that reflect changes from the background stress to near-zero transient
shear stress on the fault. The change in preferred orientation of microfractures
between the samples at 10 and 65 m from the fault is therefore compatible with
strong dynamic reduction of normal stress or fault opening.
Summary
The distribution of damage we observe delineates a ~100 m wide damage zone
on the southwest side of the SAF near Littlerock, California. Due to its clear spatial
association with the trace of the fault, damage was likely generated during SAF
earthquakes. The exhumation history of the area, the poor consolidation state of the
rock and the significant abundance of open fractures with preferred orientations
222
within the damage zone are all compatible with generation of damage close to the
surface of the Earth during seismic events.
A future study will be focused on similar types of measurements done on a
denser array of samples taken along this and other traverses, with the purposes of
improving the precision of the FIPL measurements (see item 1 of the discussion) and
delineating more precisely the damage gradient and the variations in fracture
orientations with respect to the fault. Future results will be used to infer on the actual
increase in surface area due to fracturing at shallow depth, and to constrain rupture
mechanism and strong ground motion in the immediate vicinity of the San Andreas
and other large faults.
223
Chapter 5: The geometry of slip surfaces in the hanging-wall of the
Sierra Madre fault, La-Canada, California: evidence for
Mohr-Coulomb failure induced by a dynamic off-fault
stress field.
Co-authors: Charles G. Sammis, Yehuda Ben-Zion
Department of Earth Sciences, University of Southern California, Los Angeles,
CA 90089-0740, USA.
Abstract
An analysis of fault-slip data measured on 115 cm-scale slip surfaces in the
granitic hanging-wall of the Sierra Madre fault indicates that their geometry and slip
vectors match predictions of off-fault Mohr-Coulomb slip in the stress field of a
propagating mode II earthquake rupture (Rice et al., 2005). At the microscale, we
identified two dominant fracture orientations. The first appears both near and far
from the fault and is compatible with Andersonian failure on the main fault. The
second appears only within meters from the fault and may be associated with the
formation of the cm-scale slip surfaces. Characterization of damage fabric in the
microscale suggests that in-situ shattering of crystals with minimal strain
immediately above the fault plane may be associated with an opening mode of
rupture. We conclude that the architecture of the slip surfaces was developed over a
finite displacement history with fairly stable faulting conditions, and that with
224
continuing displacement, as the rock mass approached the surface, a dynamic
opening mode could have led to the shattering of crystals in the immediate vicinity
of the slip zone.
Introduction
The exposure of the Sierra Madre fault on the banks of the Arroyo Seco Canyon
in La Canada, California, provides a rare window into the internal structure of a
major active thrust fault capable of generating large earthquakes (Rubin et al. 1998).
The Cretaceous granitic hanging-wall of the fault is separated from the Pleistocenic
conglomeratic footwall by an mm to a few cm wide gouge layer. The hanging wall
contains a myriad of cm scale slip surfaces having a conjugate geometry. These
surfaces have inclinations larger than that of the fault surface, and are observed up to
a distance of several meters from it. A layer of granite containing those slip surfaces
about a meter wide immediately adjacent to the slipping zone is pulverized. These
damage elements, described in detail below, may be interpreted in terms of
damaging mechanisms associated with the propagation of earthquake ruptures. We
have identified two classes of models that discuss the dynamic generation of damage
elements of the type we observe. Each makes distinct predictions about the geometry
of damage produced in the vicinity of a propagating rupture front which we test here
by detailed mapping, structural analysis and microscale damage fabric
characterization.
225
The first class of models is relevant specifically to thrust faults and emphasizes
an opening mode with a complete reduction of normal stress during the passage of
the rupture. Following experimental work (e.g. Brune, 1996) in which fault-opening
waves were observed, especially in the fault toe, Shi et al. (1998) constructed a 2D
solid lattice model of a thrust fault, showing that the rupture process is associated
with a pulse of normal opening displacement. Separation of the bounding blocks
occurs (Figure 5.1a) with a fault-normal component of opening larger in the hanging
wall and complete reduction of the shear stress. An asymmetric response between the
footwall and the hanging with a drop of the normal stress potentially to zero was
observed in the 2D finite element modeling of Oglesby et al. (1998; 2000). These
results require that rocks adjacent to the fault surface, mainly in the hanging-wall,
will experience absolute tensile stresses during the passage of ruptures. Dynamic
tensile regime should generate isotropic expansion of the rock resulting in a
randomly oriented population of joints and minimal distortion of the rock fabric.
Figure 5.1: Model predictions. a. Figure 6a of Shi et al. (1998) showing a snapshot of typical
particle motion pattern in a 2D lattice model. The rupture propagates up-dip with a pulse-like
behavior and fault normal displacement in the hanging-wall larger than in the footwall, leading to
fault separation. b. Figure 10b of Poliakov et al. (2002). A generic illustration of normalized shear
stress isolines and expected geometry of slip surfaces used by Rice et al. (2005) to display their model
results.
226
The second class of models discusses the stress field near the tip of a
propagating rupture and its explicit role in the generation of off-fault damage. Rice et
al. (2005) showed that a slip-weakening rupture front of a self-healing slip pulse can
cause Mohr-Coulomb failure on cohesionless slip surfaces in the wall rock. For
mode II ruptures, the area in which the failure criterion is exceeded on the hanging
wall of a thrust fault during an up-dip rupture propagation, increases with rupture
velocity, strength drop, poroelastic Skempton coefficient and with a smaller angle
between the maximum pre-stress direction and the fault. Slip occurs by the activation
of a favorably oriented conjugate set of slip surfaces with strike parallel to the main
fault and with various inclination angles with respect to the fault (Figure 5.1b).
We performed detailed structural mapping of 115 small slip surfaces, measured
the preferred orientation of microfractures and characterized microscale damage
fabric in order to test which of the two sets of predictions compares most favorably
with the type of damage observed in the exposed fault zone. We find that the
geometry and kinematics of the slip surfaces is compatible with Mohr-Coulomb
failure within the stress field induced by ruptures propagating on this fault (Rice et
al., 2005), and that the pulverization immediately above the fault plane could
possibly be related to dynamic fault opening. The two mechanisms require very
different dynamic off-fault stresses and are not likely to operate simultaneously.
Therefore, if dynamic fault opening did occur in addition to the activation of the
observed slip surfaces, it represents a separate stage in the mechanical development
of the fault or it corresponds to a different depth.
227
Regional geological setting
The Sierra Madre fault zone comprises a series of reverse faults extending more
than 75 km along the southern margins of the San Gabriel Mountains (Figure 5.2).
These are convex-southward faults which separate the crystalline pre-Tertiary rocks
on the north from the Tertiary and Quaternary sedimentary formations on the south.
The faults are discontinuous, with dips ranging from 30º to sub-vertical. All the
segments dip northward with the crystalline rocks thrust upward toward the south
over sediments as young as the mid-Pleistocene Pacoima Formation (Oakeshott,
1971). The reverse slip along the branches of the Sierra Madre fault zone is
responsible for uplift of the San Gabriel Mountains to elevations of 2-3 km and is
thought to began about 7 million years ago (Blythe et al., 2000). Paleoseismological
investigations suggest
that the Sierra Madre fault ruptures during very infrequent,
large-magnitude
(M
W
=>7) earthquakes (Rubin et al. 1998; Tucker and Dolan, 2001).
Figure 5.2: Location of the working site near La-Canada, California, at the foothills of the San
Gabriel Mountains (SGM) along the Sierra Madre fault (thin line bounding the SGM on their
southwest side. SBM – San Bernardino Mountains).
228
Local geological setting
The upper most tip of the Sierra Madre fault appears close to the mouth of the
Arroyo Seco Canyon immediately east of the Jet Propulsion Laboratory (Figure 5.3).
Trench studies in the vicinity of the bridge exposure found no evidence of slip during
the Holocene (Crook et al., 1987), but the recurrence interval of earthquakes on the
fault was found to be at the order of several thousands of years so the lack of
evidence for Holocene activity does not imply that the fault is inactive. The location
of the fault at the slope break at base of the mountain front indicates that this is one
of the active strands responsible for the current uplift. The minimum vertical throw
on this fault strand was estimated using surface and borehole data to be 244 m
(Crook et al. 1981). At this location the fault juxtaposes Cretaceous granite over a
Pleistocene alluvial fan and crops out only about a meter above the bottom of the
canyon channel on its east and west banks (see Figure 5.5).
229
Figure 5.3: Geology of the Arroyo Seco canyon at its mouth. Aerial photo shows the same general
area in an oblique view. The Sierra Madre Fault is marked by black continuous line where surface
trace is known and with dotted line where trace is projected. Arrow connects the location of the study
site in the geological map and in the picture. Note contact between granite and Pleistocene unit. The
photo was taken in 1934, when the area was not populated yet. Picture and map from Crook et al.
(1987).
Methodology
Field mapping: The granite in the hanging-wall is dissected by a web of small
slip surfaces 2-15 cm wide. These slip surfaces are the dominant and also the
smallest visible structural elements in the damage zone. In order to characterize their
geometry and kinematics we measured the orientation of 115 such slip surfaces.
Blocks bounded by those surfaces were removed progressively to expose more
230
surfaces. Where possible, we recorded the plunge (inclination) and trend (azimuth)
of their slip striations.
A more challenging task was the determination of the sense of slip on the
surfaces based on the morphology and other properties of the striations. We used an
updated classification of kinematic indicators by Doblas (1998) to determine the
sense of motion on a fault if the striations on the surface contained the required
information. Figure 5.4 shows three kinematic indicators that were used as criteria
for determination of the sense of slip. These include small steps or “pull-aparts” on
the fault plane (4a); “V” or crescent-shape markers (4b); and smeared dark grains
(4c). We were able to determine the sense slip on 49 of the 115 surfaces.
Figure 5.4: kinematic indicators used in this study. Yellow bars are cm long. Arrows indicate the
inferred sense of motion a. Step-overs - small pull-aparts that indicate movement in the direction of
the step. The cartoon shows a cross section perpendicular to the surface along the thick black line. b.
Crescentic marker, indicating motion of the missing block to the direction opposite to the tip of the
marker. c. Smeared grain, shows that the missing block moved upward in this case. It also shows that
motion did not change sense over time as the grain is smeared only in one direction.
Microstructural analysis: we collected 6 oriented samples at horizontal
distances of 0, 1, 2, 5.5, 13.3 and 250 m from the fault, corresponding to fault-
normal distances of 0, 0.7, 1.5, 4, 9.7 and 183 m (assuming a sub-surface fault
inclination of 47°, the dip measured in the outcrop). The fault normal distances were
231
used to name the samples. Sample 0 contains the 2-4 mm wide gouge layer in
contact with damaged granite at this location, and was cut into three mutually
perpendicular oriented thin sections. One oriented thin section was prepared from
each of the other samples. The microfracture and damage fabric analyses discussed
below are based on three polished thin sections from samples 0 (only one of the
orientations), 4 and 183.
We measured the orientation of transgranular microfractures in samples 4 and
183. Fractures were mostly straight or slightly curved, and for the purpose of
mapping could be approximated as straight lines. We compare properties of the
damage fabric between these two samples and sample 0, and discuss the fault-normal
variations in damage properties based on observations from all the samples.
Observations
Mesoscale observations
Figure 5.5 shows the local geometry of the outcrop. The fault is exposed on the
banks of the south-flowing channel of the Arroyo Seco canyon immediately north of
a bridge above the canyon. The dip is 47° to the northwest (334°). The gouge that
separates the granitic hanging-wall from the conglomeratic foot-wall is one mm to 5
cm wide (Figure 5.6). It ranges from brown to almost white in places, foliated, with
parallel flexible clayey slabs. Slip striations within the gouge layer have an average
rake of -52E (hade = 218), suggesting oblique thrust fault motion.
232
Figure 5.5: Local geological settings in the mapping site: dark gray – granite; light gray – alluvium.
Line with triangles mark the Sierra Madre fault, triangles point in the direction of the hanging-wall.
The channel of the Arroyo Seco canyon is marked with doted lines and the bridge above the canyon
connecting JPL and a parking lot is marked with a thick gray bar across the map.
A 0.5-1 m wide layer of granite immediately above the fault plane appears to be
pulverized (e.g. Dor et al., 2006b). Grain boundaries and other delicate magmatic
fabrics are largely preserved although the rock within this layer yields a powdery
texture when squeezed between two fingers. Above this layer the rock hardens over a
distance of about a meter or less. Starting immediately above the gouge layer, a web
of small scale slip surfaces and veins dissect the rock. Their density is the largest in
the 1-2 m of rock above the gouge layer, and it then subsides significantly (the
estimation of the intensely damaged zone may be biased because it is much harder to
identify those small scale slip surfaces in the stiffer, non-pulverized rock). The
exposure extends for about 15-20 m beyond the pulverized zone, depending on the
annual water level in the stream and vegetation cover, with a progressive decreasing
233
number of slip surfaces. The exposure is not continuous further north, although one
can get an impression about the subsidence in fracture density and the width of the
damage zone by walking northward along a stream-parallel road starting about 15 m
above the exposure.
Figure 5.6: Meso-scale structure of the Sierra Madre fault near JPL. a. Cross fault view:
Pleistocene alluvial deposits in the footwall overlaid by ~5 cm thick gouge zone. The granite above
the gouge is pulverized. Slip surfaces within the granite are marked with red lines; they are quasi-
parallel to the strike of the fault but have a steeper inclination. b. Close up map view of the slip
surfaces (red lines).
Figure 5.6a shows a vertical cross-fault view of the eastern bank of the stream
with some of the small slip surfaces delineated. Most strikes are quasi-parallel to the
main fault plane but the dips are mostly steeper than that of the main fault. Some dip
to the northwest (as does the main fault) while others dip to the southeast. The
spacing between these slip surfaces nearest to the main fault is on the order of one
cm (Figure 5.6b), where they exhibit low-angle brunching and coalescence. Many of
the surfaces contain a film of gouge that is typically shiny with a silky touch, and
almost all of them show slip striations. Some of the slip surfaces are planar and some
234
are curved. Figure 5.7 shows two relatively small slip surfaces in a typical geometry
with their slip striations at a high angle to each other.
Figure 5.7: Small-scale slip surfaces. Note the different sense of curvature, different attitude and
different trend and plunge of slip striations (black lines). A very thin gouge layer covers the surfaces.
The original granitic fabric, including individual crystals, can be identified in the rock between the
surfaces. The rock appears to be intact but it can be easily powdered by hand.
We measured the dip angle and direction of 115 slip surfaces at distances up to
about 3 m from the main fault plane, and the trend and plunge of slip striations on
them. The sense of motion was inferred for 49 surfaces. The average strike for the
slip surfaces (eastern hemisphere) is 88° with a standard deviation of 43°. The strike
of 64% of the surfaces is within one standard deviation (Figure 5.8a). Most of the
slip surfaces have high inclination angles (Figure 5.8b). 70 slip surfaces are dipping
northward with an average inclination of 69°, standard deviation of 12° and with
62% within one standard deviation. 45 slip surfaces are dipping southward with an
235
average inclination of 73°, standard deviation of 15° and with 80% within one
standard deviation.
Figure 5.8: Distribution of strike and dip of slip surfaces, 115 data.
The normals to all 115 slip surfaces are plotted in Figure 5.9a. The sense of
motion inferred for 49 slip surfaces is color coded: 36 slip surfaces have a normal
sense (red squares) and 13 slip surfaces have a reverse sense (green squares). Despite
236
potential mistakes in the interpretation of kinematic indicators on the surfaces, the
preference for a normal sense of motion within this population of slip surfaces is
vigorous. The orientation in space of all the slip striations is plotted in Figure 5.9b,
with their plunge ranging 30°-90° and without a preference for trend (i.e. there is no
preference for right or left lateral horizontal component of motion in the data).
In summary, the population of slip surfaces is characterized by a preference for
steep inclinations, trend that is oblique, in an acute angle, to that of the main fault,
preference for normal faulting and an intermediate to high angle slip vectors.
Figure 5.9: Lower hemisphere equal area projections of normals and slip vectors. a. Normals to
all 115 slip surfaces. Slip surfaces with identified sense of motion are color coded with red for normal
and green for reverse. b. Plunge and trend of all slip striations.
Microscale observations
Sample 183 was extracted from an outcrop of a relatively intact rock with only
large, sparse fractures. On the microscale the rock appears to be slightly
metamorphosed with over 50% feldspar crystals, >5% micas and the rest quartz.
237
Crystal size is on the mm scale. Although relatively far from the fault, this sample
contains many intracrystalline and intercrystalline fractures that are mostly open.
Alteration and re-crystallization is minor and shear at this scale appears to be
completely absent. All other samples, as suggested by their composition, are
damaged products of this rock type.
Samples 0.7 to 9.7 share similar features although the distortion of original rock
fabric becomes more intense closer to the fault. Shear bands with opaque fine
material or re-crystallized material appear in all of them. Despite strong alteration
and shear associated with the shear bands, the original crystals and crystal
boundaries are mostly preserved. All the samples show intense intracrystalline and
intercrystalline fracturing.
Sample 0, collected immediately above the fault plane, display the following
features (Figure 5.10): the fault gouge is about 2-4 mm thick, containing sparse white
visible sub-rounded to angular fragments up to ~100 μm wide. The rest of the gouge
material is dark brown to reddish-brown in color and is opaque in an optical
microscope. Variations in colors express clear Riedel shears consistent with the
sense of motion on the fault. The gouge has a very sharp textural and mechanical
boundary with a highly damaged layer immediately above it. This layer is matrix
supported with matrix color ranging from black to light brown. Numerous shear
bands cross this layer, usually containing ultra-fine dark material. Most of the shears
are parallel to the fault. Particle sizes range from the minimum optically visible size
to a rare size of about 500 μm. Most of the visible particles are part of aggregates in
238
which their shapes fit together with the shapes of other visible particles like pieces of
a puzzle. Those puzzles appear to be grains that were apparently shattered in-situ
without distortion of the grain’s shape (observed best under reflected light; see
Figure 5.13a below). This is somewhat surprising given the intense shearing that the
material appears to accommodate.
Figure 5.10: A transmitted light photomicrograph of sample 0. The inset shows the sample, with
the gouge layer and slip striations facing toward the viewer, scale in cm. The thin section plane is
parallel to the down-dip direction of the fault and it intersects the face of the sample along the black
dashed line in the inset. A 2-4 mm wide gouge layer at the bottom of the figure has a sharp
mechanical and compositional boundary with the damaged rock above it. Clear Riedel shears
compatible with the sense of motion (red arrow) appear inside the gouge. The damaged rock contains
many shattered grains supported by a brown opaque highly sheared matrix with some apparent shear
bands parallel to the fault.
Figure 5.11 shows transmitted light photomicrographs of samples 4 and 183
overlaid by a transgranular fractures map. In sample 183 (Figure 5.11a) all the
239
fractures are marked with black lines and in sample 4 (Figure 5.11b) fractures are
marked with green and blue lines, corresponding to two preferred orientations we
identified. Rose diagrams of all the fractures mapped in samples 183 and 4 appear in
Figure 5.12 and data is summarized in Table 5.1. Fractures in sample 183, assuming
to represent the far-field (or regional) damage fabric, have major and minor preferred
orientations (Figure 5.12a): the dominant fracture set trends to the NW, and a minor
fracture set trends to the NNE. Fractures in sample 4 have two preferred orientations
as well (Figure 5.12 b-c): fracture set A trends to the NW, like the dominant fracture
set in sample 183, and fracture set B trends to the ENE. The ENE trending set and
the population of mesoscale slip surfaces (Figure 5.12d) have a similar preferred
orientation.
240
Figure 5.11: Transmitted light photomicrographs of samples 183 and 4. a. Sample 183, showing
an undisturbed crystal-scale and larger magmatic fabric of the granite. Transgranular fractures,
marked as black lines, trend mostly to the northwest. b. Sample 4. Original crystal boundaries are
largely preserved but the rock fabric is intensely disturbed by several generations of veins, alterations
and a dense fracture network. Two sets of transgranular fractures are marked: the green set trends to
the northwest, like the dominant fracture set in sample 183; and the blue set trends roughly east-west,
similar to the dominant orientation of the mesoscale slip surfaces.
241
Table 5.1: Properties of measured fracture sets in samples 183 and 4.
Figure 5.12: Rose diagrams. a. Sample 183 (Figure 5.11a), b. Sample 4, set A (Figure 5.11b, green
set), c. Sample 4, set B (Figure 5.11b, blue set), d. All mesoscale slip surfaces (figures 5.8, 5.9).
The intensity of fragmentation (shattering) of the crystals intensifies and
fragments become smaller closer to the fault. The original crystal outlines are
completely preserved in sample 183, and are mostly preserved in sample 4. In
sample 0 crystals can still be identified despite the heavy fragmentation, but we
No. of data Mean direction 95% confidence
interval (+/-degrees)
Sample 183 346 332° 6
Sample 4 – set A 229 322° 5
Sample 4 – set B 192 70° 6
242
observe that their outer shape is more rounded compare to crystals in the other two
samples. The average crystal size appears to be much smaller near the fault compare
to the average crystal size in sample 183 (the average size of crystals that we
measured in samples 4 and 0 is 49% and 22%, respectively, of the average crystal
size in sample 183. Note that this is not the average grain size). The more rounded
shapes of crystals in sample 0, compare to the irregular shapes of crystals in the other
samples, can probably be attributed to wear, rotation and shearing. In addition,
fragments in sample 0 seem to have a higher aspect ratio compare to fragments in the
other two samples (Figure 5.13). They still have sharp corners and they fit very well
with neighboring fragments, indicating that fragmentation of the crystals occurred in-
situ. The more intense fragmentation near the fault is expected, but the different
fragments shape near the fault is intriguing, suggesting that the grain size reduction
mechanism operating in sample 0 is unique to the vicinity of the fault plane.
Figure 5.13: Photomicrographs of single grains. Samples names / distances from the fault appear in
the upper left corner of each image. a. Sample 0: the crystal, embedded in a cataclastic matrix, is
strongly shattered but its fragments still fit together. Aspect ratio of fragments is high. b. Sample 4:
quartz crystal (lighter color) cut by few transgranular and intragranular fractures. c. Sample 183:
quartz crystal, cut by a single transgranular fracture.
243
Discussion
Activation of cm-scale slip surfaces during earthquake ruptures
A strike-slip fault along with two subsidiary slip surfaces in the general
geometry predicted by Rice et al. (2005) for dynamic failure outside the fault plane
(Figure 5.1b) is projected on a lower hemisphere net in Figure 5.14a. These elements
are projected twice: as planes and as normals to these planes. The fault plane and its
normal are projected with thick line and thick open circle, respectively, and the slip
surfaces and their normals are projected with thin line and thin open circles,
respectively. The slip vector on the main fault is shown with a filled circle. The
surface that contains the fault normal and slip vector (its projection is the perimeter
of the circle in this case because the fault is vertical) must also contain the normals to
the slip surfaces because their strike is parallel to the fault strike and perpendicular to
the fault slip vector.
244
Figure 5.14: Projection of model prediction on the Sierra Madre fault geometry. a. Lower
hemisphere projections of a strike slip fault and the associated generic damage geometry predicted by
Rice et al. (2005). See Figure 5.1b here. The main fault surface is projected as a thick line and its
normal is projected as a thick circle. The slip vector is projected as a filled circle. Slip surfaces are
projected as thin lines and their normals are projected as thin circles. Since those elements are
vertical, their normals could be projected also as the gray circles at the top. The normals to the slip
surfaces must be contained in the plane that includes the normal to the main fault and its slip vector. b.
A fault and the exact geometry for preferably oriented slip surfaces activated due to Coulomb yielding
associated with off-fault dynamic stresses predicted by Rice et al. (2005) for a case of 45° between the
fault and the maximum pre-stress direction, shown here in a configuration analogue to a vertical cross
section of a thrust fault. Only slip surfaces on the side of the fault that corresponds to the hanging-wall
are shown. c. Lower hemisphere equal area stereographic projection of the Sierra Madre fault and the
plane containing the fault slip vector (filled circle) and the normal to the fault (thick circle). The
predicted favorably oriented slip surfaces from (b) are projected as thin circles.
245
Assuming Andersonian faulting conditions, the fault dip (47°) is also its angle
with the maximum pre-stress direction. On the basis of detailed boreholes data,
trenching and surface mapping in and around our study area, Crooks et al. (1987)
indicate that “the fault zone consistently dips about 45° N”. A case with 45° between
the fault and the maximum pre-stress direction is simulated by Rice et al. (2005). For
mode II rupture on a fault with such an angle with the pre-stress direction, and under
some conditions (high rupture speed, large residual stress ratio, short slip weakening
distance and non-zero poroelastic effects), the Mohr-Coulomb failure criterion is
exceeded on the side of the fault that corresponds to the hanging wall of a thrust fault
with up-dip rupture propagation, and slip surfaces in a favorable orientation are
activated. The preferably oriented right lateral an left lateral slip surfaces that will be
activated have an angle with the fault surface of 13° and 77°, respectively. In reality,
it is expected that surfaces in a range of orientations centered on the favorable
orientations will be activated. The resultant geometry between the fault and the
favorably-oriented slip surfaces as it appears in Figure 12 of Rice et al. (2005) is
presented here in Figure 5.14b, rotated to match the inclination of the Sierra Madre
fault. While the figures in Rice et al. (2005) show a map view of a strike-slip fault,
Figure 5.14b is a vertical cross section of a thrust fault.
Figure 5.14c shows the prediction of the Rice et al. (2005) model for the
conditions above superposed on the geometry of the Sierra Madre fault for ruptures
propagating in the direction of the slip vector found in our working location. Since
the fault is dipping 47° to 334° with a slip vector plunging 35° to 023°, the plane that
246
contains the slip vector and the normal to the fault is dipping 63° to 092°. The
normals to the slip surfaces predicted by the model for this geometry (Figure 5.14b)
must therefore lie on the line representing this plane in the projection, and they are
shown in Figure 5.14c as open circles.
We overlaid the predicted geometry of slip surfaces of Figure 5.14c on our
contoured data set of normals to the observed slip surfaces (Figure 5.15a). Like the
normals to the predicted slip surfaces, the peaks in the distribution of normals to the
observed slip surfaces (blue counters) lie on the plane containing the normal to the
main fault plane and the slip vector. If the observed slip surfaces were formed in the
predicted geometry, they must have been rotated to their current inclination.
Figure 5.15: The contoured data set of 115 normals to slip surfaces projected over the
prediction of Figure 5.14c. The peaks in the distribution of the normals (blue contours) lie on the
plane containing the normal to the fault, its slip vector, and the normals to the predicted favorably
oriented slip surfaces. A rotation of 75° around as axis normal to that plane of those predicted surfaces
in the directions allowed by their sense of slip (Figure 5.14b) brings them to the orientation of the
observed slip surfaces. The rotation in the figure does not appear to be identical for the two predicted
surfaces due to a visual effect.
247
Due to the sense of motion on the predicted slip surfaces, the predicted south-
facing slip surfaces can only be rotated clockwise and the predicted north-facing slip
surface can only be rotated counterclockwise. Trying various rotation angles for
those slip surfaces around an axis perpendicular to the plane containing the normal to
the main fault surface and the slip vector, we found that rotating both predicted slip
surfaces 75° in the allowed directions (Figure 5.15a), shift their projected normals to
the peak (blue contour) in the distribution of the observed slip surfaces. The result of
this rotation is shown in Figure 5.16: both predicted-rotated slip surfaces are now
dipping in a steep inclination, about 30° steeper than the fault surface itself, similar
to the geometrical relationships observed in the field (compare to Figure 5.6 and data
in Figures 5.8, 5.9, 5.12 and 5.15). In addition, both predicted slip surfaces, which
had a reverse sense of motion in their original inclination, have a normal sense of
motion in their rotated inclination, compatible with our observation of preference for
a normal sense of motion in the data (see Figures 5.8 and 5.9). Scattering in the
inclinations, as shown schematically in Figure 5.16 by the thin dashed lines, will
allow some of the slip surfaces to maintain their reverse sense of motion.
248
Figure 5.16: Rotated predicted data compared with observed data. The predicted slip surfaces of
Figure 5.14b (thin continuous lines) rotated 75° in the direction allowed by their sense of motion
(dashed thicker lines). The resultant geometry is similar to that observed in the field (Figure 5.6a),
with a preference for normal faults, as shown in the data (Figure 5.9a). A scattering in the original dip
angles of the slip surfaces population will allow some of the rotated slip surfaces (thin dashed lines) to
maintain their reverse sense.
Large rotation of closely spaced conjugate sets of slip surfaces, at the order of
the rotation inferred for our case, should be associated with displacement of surfaces
by each other and fragmentation of the rock between the slip surfaces into smaller
blocks with rounded faces (especially if the rock is soft or pulverized), and this is
what we observe in the 1-2 m of rock above the fault plane. As we get closer to the
fault surface, especially in the 0.5 m immediately above it, the slip surfaces create an
anastomosing web of planes that bound rock fragments with concoidal shapes (as in
Figure 5.7). The size of the slip surfaces becomes smaller as they are more closely
spaced. This geometry is compatible with an internal rotation of the rock mass. As
249
the rotation-fragmentation progress, the original slip surfaces architecture is
obliterated, and the small fragments may experience different kinematics than the
one described above for the slip surfaces. In our measurements we therefore avoided
particularly small surfaces with concoidal shapes (at the order of ~2 cm, e.g. Figure
5.7) and chose larger and more planar surfaces that ‘survived’ the rotation.
Under ideal conditions where successive rupture events propagate with the same
slip vector on a perfectly planar fault surface we would expect minimal scattering in
the orientation of the slip surfaces and their slip vectors. A common observation in
natural faults is crosscutting relations between sets of slip striations (e.g. Sagy et al.,
2007), sometimes during a single slip event (Nobuaki and Kenshiro, 2003). Given
that the slip vector may change between rupture events and even during the
propagation of one rupture, leading to interseismic and intraseismic changes in the
off-fault stress field and the favorable orientation of slip surfaces, it is even
surprising to find sets of slip surfaces with preferred orientation, but this is what we
observe. The existence of such identifiable sets implies that the slip vector on this
fault was relatively uniform throughout its displacement history. The large scattering
in the orientation of the slip vectors (Figure 5.9b) may be associated not only with
changes in the dynamic stress field, but also with exhumation and un-roofing of the
fault. When the a rock mass approaches the surface, it becomes more susceptible for
strain due to thermal cycling, which may activate some slip on existing slip surfaces,
an artifact that may affect the distribution of the slip striations data. Nevertheless, the
slip vector for the slip surfaces predicted by the model is expected to be on a plane
250
perpendicular to the fault strike and parallel to its slip vector, i.e. having a high
plunge angle for the Sierra Madre fault, and the observed slip vectors are plunging in
intermediate to high angles (Figure 5.9b).
Microfractures and pulverization
The two distinct sets of fractures mapped in sample 4 can be interpreted in a
mechanical framework: fracture set A, trending to the northwest, is similar in its
orientation to the dominant fracture set found in sample 183 (Figure 5.12a, b). Those
sets are perpendicular or nearly perpendicular to the fault strike. Their density seems
to be similar between the two samples (although not considered here as a fracture
density measurement, the number of fractures belong to those sets is very similar
between the samples, given that the area of the thin section of sample 4 is 2/3 the
area of the thin section of sample 183. As of graphic reasons some of the image area
of the thin section from sample 183 is cropped in Figure 5.11). Since this fracture
orientation appears both near and far from the fault without an apparent drop in the
fracture density, this orientation is probably a representative of the regional fracture
fabric, which is compatible with northwest-southeast compression. Such a
compression direction is expected if the Sierra Madre fault is operating under
Andersonian faulting conditions, with the maximum compressive stress horizontal
and normal to fault strike. This supports our assumption of Andersonian conditions
when applying the Rice et al. (2005) model for the Sierra Madre fault. Fracture set B
(Figure 5.12c) shares similar orientation with the average trend of the mesoscale slip
251
surfaces (Figure 5.12d), and is most likely associated with their formation (e.g.
Moore and Lockner, 1995).
Another aspect of the damage observed in the microscale is that transgranular
and intragranular fractures reduce the grain size without distortion of the original
crystals shape, while larger scale shear bands and veins dissect the rock and disturb
magmatic fabrics. These damage elements are observed in all samples between 9.7
and 0.7 m from the fault with increasing intensity toward the fault. The damage
pattern in sample 0 immediately above the fault plane appears to be different with
much denser intragranular fracture network, breaking crystals to much smaller
fragments with higher aspect ratio (Figure 5.13). The crystals are smaller and more
rounded compare to crystals in samples further away from the fault. They are
embedded in a highly sheared fine matrix with cataclastic nature. Those observations
suggest that the damage process in the immediate vicinity of the fault plane is
different than the damage process that occurs further away from the fault. Crystals in
the cataclasite layer are subjected to two types of grain size reduction processes that
operate in a weak form or do not operate outside this layer: one of them is associated
with wear and shear, resulting in a reduction in the original size of the crystals and in
smoothing of the crystal outlines; the other one is responsible for the in-situ
fragmentation of the crystals without an apparent strain, producing fragments with
high aspect ratio.
The apparent lack of grain scale strain associated with the in-situ fragmentation
is compatible with failure in a tensile mode. While intense shear and cataclastic flow
252
is expected immediately near the fault surface, tensile regime is less intuitive in such
an environment. One mechanism that can lead to failure due to a grain-scale tensile
stress in a compressive environment is failure under bi-polar loading by same-sized
neighbors in the load-bearing stress chains that transmit force in a granular media
(e.g. constrained comminution, Sammis et al., 1987). This process is not unique to
dynamic rupture and should result in grain size distribution with fractal dimension of
2.6 or 3, depending on the matureness stage of the fault (Sammis and King, 2007).
Another possible source for tensile stresses associated with earthquake ruptures is
fault normal unloading and loading i.e., an opening mode (Brune et al, 1993) that
may lead to a complete reduction of the normal stress. Such an opening mode is
expected to be especially pronounced in the hanging-wall of a thrust fault, and in
particular as the rupture approaches the surface near the fault toe (Brune, 1996; Shi
et al., 1998; Oglesby et al., 1998, 2000). If as the result of the opening mode tensile
stresses are developed, an isotropic expansion of the rock is expected. Under these
conditions the expected damage fabric should have no preferred orientation. The
‘exploded’ crystals in sample 0 and the overall pulverization (with minimal
distortion of fine magmatic fabrics) immediately above the fault surface are
qualitatively compatible with this description. Nevertheless, further quantification of
the fragmentation in three mutually perpendicular thin sections is required to test
whether damage is isotropic, compatible with an absolute tension, or anisotropic,
compatible with tensile failure due to directional compressive grain-grain contacts.
In addition, the grain size distribution and the fractal dimension (if fractal) should be
253
measured. We currently can not determine which of the two mechanisms is
responsible for this type of failure, although we note that failure is likely associated
with dynamic rupture.
Damage accumulation over many seismic cycles
As mentioned above, the existence of identifiable preferred orientation in the
population of slip surfaces and their interpretation as plastic response to off-fault
dynamic stresses suggests that the fault slip vector did not changed significantly
during and between rupture events while those surfaces were active in
accommodating the off-fault response. This is surprising because during the faulting
history, the roughness of the fault and its interactions with other faults and structures
should cause dynamic stress perturbations and static geometrical perturbations that
are expected to influence the fault’s kinematics, e.g. lead to changes in the slip
vector. Another intriguing observation is that only one double-set of slip surfaces
was developed (although the scattering in the data may suggest the immature
development of an additional set). After significant rotation, like the one we inferred
for the slip surfaces, they should have stopped been favorably oriented for activation.
Upon further faulting, the development/activation of an additional set of surfaces is
expected. This did not happen or happened in a very minor form. If this form of
damage was developed in the dynamic stress field of a propagating rupture and not
quasistatically during interseismic periods, the observed architecture of the slip
surfaces must have been developed over a finite (short) displacement history in
254
which faulting conditions remained stable. This fault been part of a seismically
active thrust fault zone and the existence of pulverized rocks that are likely
associated with seismic faulting (e.g. Reches and Dewers, 2004; Dor et al. 2006b) in
its structure, land support to the inference that the damage we observe was generated
dynamically.
Based on this conclusion, and because the activation of the slip surfaces can
occur under particular conditions (Rice et al., 2005), there are two possibilities to
consider: the first possibility is that the geometry we observe was developed over a
rather short displacement history during the fault life time, in which ruptures met
those required conditions. Then, and after this phase in the fault development ended,
activity of the slip surfaces ceased, and their fossilized architecture was exhumed.
The second possibility is that the conditions required for the activation of the
surfaces are met only at some interval along the fault surface at depth (for example,
high rupture velocity, specific poroelastic conditions etc.). When the rock mass
containing those surfaces propagate up-dip, the conditions then become insufficient
for their activation – and their fossilized architecture is exhumed while at depth the
process continues. At the moment we are unable to answer which of the possibilities
is more realistic.
The stress field required for the activation of the favorably oriented slip surfaces
and an opening mode with complete reduction of the normal stress are unlikely to
operate simultaneously. Nonetheless, if a rock mass containing the fossilized slip
surfaces architecture approached the surface and was then subjected to an opening
255
mode, since the later is predicted to operate very strongly at shallow depths and in
particular in the fault toe, the signature of the two mechanisms could be found
together. Hence, we conclude that the observed slip surfaces in the few meters above
the fault surface were probably activated when the Mohr-Coulomb yielding criterion
was exceeded during ruptures propagation (Rice et al., 2005), while pulverization
immediately above the fault surface could possibly be related to a dynamic opening
mode.
Summary
We identified that the following processes may have operated in the hanging-
wall of the Sierra Madre fault:
1. Mohr-Coulomb Slip on- and rotation of small scale favorably oriented
cm-scale slip surfaces in a few meters zone above the fault surface during
the passage of earthquake ruptures.
2. Development of fracture fabric in the few meters above the fault surface
associated with the generation of the mesoscale slip surfaces.
3. Development of fracture fabric on a regional scale associated with
compressive stresses in an Andersonian geometry with respect to the
fault surface.
4. Cataclastic process in a few cm wide layer immediately above the fault
plane, leading to strong grain size reduction and shearing of the rock.
5. In-situ fragmentation of crystals without significant shape distortion in
256
the cataclastic layer immediately above the fault surface. This could
possibly be the result of absolute tensile stresses associated with dynamic
opening mode.
6. Riedel shears and extreme grain size reduction within mm scale wide
gouge layer.
257
Bibliography
The references in Chapter 2 are indexed. Their index is shown here as [**].
[11] Adams, G.G. 1995. Self-excited oscillations of two elastic half-spaces sliding
with constant coefficient of friction. J. Appl. Mech., 62, 867-872.
Agar, S.M., Prior, D.J. and Behrmann, J.H. 1989. Back-scattered electron imagery of
the tectonic fabrics of some fine-grained sediments: implications for fabric
nomenclature and deformation processes. Geology, 17, 901-904.
[56] Agnew, D. C., and Sieh, K.E. 1978. A documentary study of the felt effects of
the great California earthquake of 1857. Bull. Seism. Soc. Am., 68, 6, 1717-
1729.
Allen, C.R., Amand, P.St., Richter, C.F., and Nordquist, J.M. 1965. Relationship
between seismicity and geological structure in the Southern California
region. Bull. Seism. Soc. Am., 55, 753-797.
Ambraseys, N. N. and Zatopek, A. 1969. The Mudurnu valley, west Anatolia,
Turkey, earthquake of 22 July 1967. Bull. Seism. Soc. of Am., 59(2):521-
589.
Ampuero, J.-P. and Ben-Zion, Y. 2007. Cracks, pulses and macroscopic asymmetry
of dynamic rupture on a bimaterial interface with velocity-weakening
friction, ms. in preparation.
[40] Anders, M.H., N. Christie-Blick and S. Wills. 2001. Rock deformation studies
in the Mineral Mountains and Sevier Desert of west-central Utah:
Implications for upper crustal low-angle normal faulting. Geol. Soc. Am.
Bull. 113 (7): 895-907.
Anderson, E.M. 1942. The Dynamics of Faulting and Dyke Formation with
Applications to Britain, Oliver and Boyd, Edinburgh.
Anderson, J.L., Osbourne, R.H. and Palmer, D.F. 1983. Cataclastic rocks of the San
Gabriel fault: An expression of deformation at deeper level in the San
Andreas fault zone. Tectonophysics., 98, 209-251.
258
[26] Andrews, D. J. 2005. Rupture Dynamics with Energy Loss outside the Slip
Zone, J. Geophys. Res., 110, B01307, doi:10.1029/2004JB003191.
[12] Andrews, D.J. and Ben-Zion, Y. 1997. Wrinkle-like Slip Pulse on a Fault
between Different Materials. J. Geophys. Res., 102, 553-571.
[19] Andrews, D.J. and R.A. Harris, The wrinkle-like slip pulse is not important in
earthquake dynamics, Geophys. Res. Lett., 32, L23303, (2005),
doi:10.1029/2005GL02399.
Anooshehpoor, A. and Brune, J.N. 1999. Wrinkle-like Weertman pulse at the
interface between two blocks of foam rubber with different velocities.
Geophys. Res. Lett., 23, 2025-2028.
Barrows, A.G. (1980). Geologic map of the San Andreas Fault zone and adjoining
terrains, Juniper Hills and vicinity, Los Angeles County, California.
California Division of Mines and Geology open-file report 80-2 LA, map
scale 1:9000.
Barrows, A.G. (1980). Geology of the San Andreas Fault zone and adjoining
terrains, Juniper Hills and vicinity, Los Angeles County, California.
California Division of Mines and Geology open-file report 85- LA.
[36] Barrows, A.G., Kahle, J.E. and Beeby, D.J. 1985. Earthquake Hazard and
tectonic history of the San Andreas Fault zone, Los Angeles County,
California. Open file report 85-10 LA., California Department of
Conservation, Division of Mines and Geology.
[4] Ben-Zion, Y. 2001. Dynamic Rupture in Recent Models of Earthquake Faults. J.
Mech. Phys. Solids., 49, 2209-2244.
[20] Ben-Zion, Y., 2006. Comment on “The wrinkle-like slip pulse is not important
in earthquake dynamics” by D. J. Andrews and R. A. Harris, Geophys. Res.
Lett., 33, L06310, doi:10.1029/2005GL025372.
[15] Ben-Zion, Y. and Andrews, D.J. 1998. Properties and Implications of Dynamic
Rupture Along a Material Interface. Bull. Seism, Soc. Am., 88, 1085-1094.
259
[14] Ben-Zion, Y. and Huang, Y. 2002. Dynamic Rupture on an Interface Between a
Compliant Fault Zone Layer and a Stiffer Surrounding Solid, J. Geophys.
Res., 107: Art. No. 2042, doi: 10.1029/2001JB000254.
Ben-Zion, Y., Katz, S. and Leary, P. 1992. Joint inversion of fault zone head waves
and direct P arrivals for crustal structure near major faults. J. Geophys. Res.,
97, 1943-1951.
Ben-Zion, Y and Malin, P. 1991. San Andreas Fault Zone Head Waves near
Parkfield, California, Science, 251, 1592–1594.
[47] Ben-Zion, Y., Peng, Z., Okaya, D., Seeber, L., Armbruster, J.G., Ozer, N.,
Michael, A.J., Baris, S., and Aktar, M. 2003. A shallow fault zone structure
illuminated by trapped waves in the Karadere-Duzce branch of the north
Anatolian Fault, Western Turkey, Geophys. J. Int., 152, 699-717.
[54] Ben-Zion, Y. and Sammis, C. G. 2003. Characterization of Fault Zones, Pure
Appl. Geophys., 160, 677-715.
[27] Ben-Zion, Y. and Shi, Z. 2005. Dynamic rupture on a material interface with
spontaneous generation of plastic strain in the bulk. Earth. Plant. Sci. Lett.,
236, 486-496, DOI: 10.1016/j.epsl.2005.03.025.
Biegel, R. L., Sammis, C. G., and Dieterich, J. H. 1989. The frictional properties of a
simulated gouge having a fractal particle distribution, J. Structural Geol.,
11, 827-846.
Blythe, A.E., Burbank, D.W., Farley, K.A. and Fielding, E. 2000. Structural and
topographic evolution of the central Transverse ranges, California, from
apatite fission track, (U/Th)/He and digital elevation model analyses. Basin
Research., 12, 97-114.
[18] Brietzke, G.B. and Ben-Zion, Y. 2006. Examining tendencies of in-plane
rupture to migrate to material. Geophys. J. Int. 167 (2): 807-819.
Brune, J.N. 1968. Seismic moment, seismicity, and rate of slip along major fault
zones. J. Geophys. Res., 73, 777-784.
260
[1] Brune, J.N. 2001. Fault normal dynamic loading and unloading: an explanation
for “non-gouge” rock powder and lack of fault-parallel shear bands along
the San Andreas fault. EOS Trans. Am. Geophys. Union. 82. Abstract.
Brune, J. N. 1996. Particle motion in a physical model of shallow angle thrust
faulting, Proc. lndian. Acad. Sci. 105, 197-206.
[3] Brune, J. N., Brown, S. and Johnson P.A. 1993. Rupture mechanism and
interface separation in foam rubber models of earthquakes; a possible
solution to the heat flow paradox and the paradox of large
overthrusts. Tectonophysics., 218, 1-3.
Brune, J.N., Anooshehpoor, R. and Zeng, Y. 1999. Ground motions for shallow
angle thrust faults. Final technical report, University of Nevada at Reno,
eismological Laboratory, United States.
[8] Byerlee, J. 1993. Model for episodic flow of high-pressure water in fault zones
before earthquakes, Geology, 21, 303-306.
Chester, F.M. 1995. unpublished mapping.
Chester, F.M. and Chester, J.S. 1998. Ultracataclasite structure and friction processes
of the Punchbowl fault, San Andreas system, California. Tectonophysics.,
295, 199-221.
Chester, F.M., Chester, J.S. 2000. Stress and deformation along wavy frictional
faults. J. Geophys. Res. 105, 23,421–23,430.
[41] Chester, F. M., Chester, J. S ., Kirschner, D. L., Schulz, S. E. and Evans, J. P.
2004. Structure of large-displacement strike-slip fault zones in the brittle
continental crust. In: Karner, Gary D., Taylor, Brian, Driscoll, Neal W.,
Kohlstedt, David L. (eds.), Rheology and Deformation in the Lithosphere at
Continental Margins. Columbia University Press, New York, MARGINS
Theoretical and Experimental Earth Science Series 1, 223-260.
Chester, F.M., Evans J.P., and Biegel, R.L. 1993. Internal structure and weakening
mechanisms of the San Andreas fault. J. Geophys. Res., 98, 771-786.
261
Chester, J.S., Lenz, S.C., Chester, F.M. and Lang, R.A. 2004. Mechanisms of
compaction of quartz sand at diagenetic conditions. Earth Planet. Sci. Lett.
220, 435-451.
[52] Chester, F.M. and Logan, J.M. 1986. Implications for mechanical properties of
brittle faults from observation of the Punchbowl fault zone. California.
Pure. Appl. Geophys., 124, 79-106.
Chorley, R.J., Schumm, S.A., Sugden, D. 1984. Geomorphology. University Press.
Cambrigde.
[16] Cochard, A. and Rice, J.R. 2000. Fault rupture between dissimilar materials: Ill-
posedness, regularization, and slip-pulse response. J. Geophys. Res., 101,
25,321-25,336.
Cochran, E. S., Y.-G. Li, and J. E. Vidale. 2006. Anisotropy in the Shallow Crust
Observed around the San Andreas Fault Before and After the 2004 M 6.0
Parkfield Earthquake, Bull. Seism. Soc. Am., 96, S364–S375, doi:
10.1785/0120050804.
Crook R. Jr, C.R. Allen, B. Kamb, C.M. Payne, and Proctor R.J. 1987. Quaternary
geology and seismic hazard of the Sierra Madre and associated faults,
western San Gabriel Mountains - US Geological Survey Professional Paper.
Crowell, J.C. 1962. Displacement along the San Andreas fault, California. Spec. Pap.
Geol. Soc. Am., 71, 61 pp.
[51] d'Alessio, M.A., written communication, 2006.
Dalguer, L.A. and Day, S.M. 2006. Toward the Reliability of Fault Representation
Methods in Finite Difference Schemes for Simulation of Shear Rupture
Propagation. Eos Trans. AGU, 87(52), Fall Meet. Suppl., Abstract S52B-04.
[25] Dalguer, L A; Irikura, K; Riera, J D. 2003. Simulation of tensile crack
generation by three-dimensional dynamic shear rupture propagation during
an earthquake. J. Geophys. Res.. 108, B3, 2144, doi:10.1029/2001JB001738
Dershowitz, W.S. and Herda, H.H. 1992. Interpretation of fracture spacing and
intensity. Proc. U.S. Symp. Rock Mech., 33
rd
, 757-766.
262
Di Toro, G. and Pennacchioni, G. 2004. Superheated friction-induced melts in zoned
pseudotachylytes within the Adamello tonalites (Italian Southern Alps). J.
Struct. Geol.26, 1783-1801.
Dibblee, T.W., Jr. 1967. Displacements on the San Andreas fault system in the San
Gabriel, San Bernardino, and San Jacinto Mountains, southern California.
(in Conference on geologic problems of San Andreas fault system,
Stanford, Calif., Proc. STANFORD UNIV. PUBS. GEOL. SCI. v. 11, p.
260-276, illus).
Dibblee, T.W. Jr. 1968. Displacement on the San Andreas fault system in the San
Gabriel, San Bernardino and San Jacinto Mountains, Southern California.
In: Dickinson, W.R. and A. Grantz, (Eds.), Proceedings of the Conference
on Geologic Problems of San Andreas Fault system, Geological Science,
11. Stanford University Publications, pp. 260-278.
Dibblee, T.W., Jr. 1987. Geology of the Devil’s Punchbowl, Los Angeles County,
California. Geological Society of America Centennial Field Guide
Cordilleran Section, 207-210.
Dibblee, T.W., Jr. 1989. The San Andreas Fault and major rock terranes of
California displaced by it and its tectonics. in Baldwin, E. Joan, et al.
ANNUAL FIELD TRIP GUIDEBOOK. SOUTH COAST GEOLOGICAL
SOCIETY. Vol. 17, Vol. 1, p. 223-275.
Dibblee, T.W. and Minch, J.A. 2002. Geologic map of the Valyermo quadrangle,
Los Angeles County, California: Dibblee Geological Foundation, Map DF-
80, scale 1:24000.
Di Toro, G., Nielsen, S. and Pennacchioni1, G. 2005. Earthquake rupture dynamics
frozen in exhumed ancient fault. Submitted to Nature.
Doblas, M. 1998. Slickenside kinematic indicators. Tectonophysics., 295, 187-197.
[29] Dor, O., T.K. Rockwell, and Y. Ben-Zion. 2006a. Geologic observations of
damage asymmetry in the structure of the San Jacinto, San Andreas and
Punchbowl faults in southern California: A possible indicator for preferred
rupture propagation direction. Pure Appl. Geophys., 163, DOI
10.1007/s00024-005-0023-9.
263
Dor, O., Ben-Zion, Y., Rockwell, T.K., and Brune J.N. 2006b. Pulverized Rocks in
the Mojave section of the San Andreas FZ, Earth Planet. Sci. Lett. 245,
642-654. doi:10.1016/j.epsl.2006.03.034.
[45] Dor, O., Z. Reches and G. van Aswagen. 2001a. Fault zones associated with the
Matjhabeng earthquake, 1999, South Africa, Rockburst and Seismicity in
Mines, RaSiM5 (Proceedings), South African Inst. Of Mining and
Metallurgy, 109-112.
Dor O., Reches Z., van Aswegen G. and Bosman, K. 2001b. Earthquake Rupture at
the Focal Depth of M=5.1 and M=3.7 Earthquakes in Gold Mines, South
Africa. EOS Trans. Amer. Geophys. Union., 81. Abstract.
Eberhart-Phillips, D. and Michael, A.J. 1993. Three dimensional Velocity structure,
seismicity and fault structure in Parkfield region, central California. J.
Geophys. Res., 98, 9, 15,737-15,758.
Eidelman A. and Reches Z. 1992. Fractured pebbles: a new stress indicator.
Geology., 20, 4, 307-310.
Emre, O., Kondo, H., Yildirim, C., Ozaksoy, V. 2005. Fault geometry and slip
distribution of the 1943 Tosya Earthquake rupture, North Anatolian Fault,
Turkey. Eos Trans. AGU, 86(52), Fall Meet. Suppl., abstract T41B-1291.
Engelder, J.T. 1974. Cataclasis and the Generation of Fault Gouge. Geol. Soc. Am.
Bull. 85, 10, 1515–1522.
[55] Evans, J.P. and Chester, F.M. 1995. Fluid-rock interaction in faults of the San
Andreas system – inferences from San-Gabriel fault rock geochemistry and
microstructures. J. Geophys. Res., 100, 13007-13020
Fialko, Y., D. Sandwell, D. Agnew, M. Simons, P. Shearer, and B. Minster. 2002.
Deformation on nearby faults induced by the 1999 Hector Mine earthquake,
Science, 297, 1858-1862.
264
Fialko, Y. 2004. Probing the mechanical properties of seismically active crust with
space geodesy: Study of the co-seismic deformation due to the 1992 Mw7.3
Landers (southern California) earthquake. J. Geophys Res., 109,
doi:10.1029/2003JB002756.
[31] Fuis, G.S., Clayton, R.W., Davis, P.M., Ryberg, T., Lutter, W.J., Okaya, D.A.,
Hauksson, E., Prodehl., C., Murphy, J.M., Benthien, M.L., Baher, S.A.,
Kohler, M.D., Thygesen, K., Simila, G., and Keller, G.R. 2003. Fault
systems of the 1971 San Fernando and 1994 Northridge earthquakes,
southern California: Relocated aftershocks and seismic images from
LARSE II. Geology, 31, 171-174.
Fuis, G.S., Ryberg, T., Godfrey, N., Okaya, D.A., and Murphy, J.M. 2001. Crustal
structure and tectonics from the Los Angeles basin to the Mojave Desert,
southern CA. Geology, 29, 15-18.
Fumal, T.E., Pezzopane, S.K., Weldon, R.J.II, and Schwartz, D.P., 1993. A 100-year
average recurrence interval for the San Andreas fault at Wrightwood,
California. Science., 259, 199-203.
Gangalakunta P. Obi Reddy, Amal K. Maji, Kothiram S. Gajbhiye. 2004. Drainage
morphometry and its influence on landform characteristics in a basaltic
terrain, Central India – a remotesensing and GIS approach. Int. J. Appl
Earth Observation and Geoinformation, 6 ,1–16.
Goodwin, L.B. and Tikoff, b. 2002. Competency contrast, kinematics, and the
development of foliations and liniations in the crust. J. Struc. Geol. 24,
1065-1085.
Hack, J.T. 1957. Studies of longitudinal stream profiles in Virginia and Maryland.
U.S. Geological Survey Professional Paper 294-B, 45–94.
Hamiel, Y., Liu, Y., Lyakhovsky, V., Ben-Zion, Y. & Lockner, D. 2004. A Visco-
elastic damage model with applications to sTable and unsTable fracturing,
Geophys. J. Int., 159, 1155–1165, doi: 10.1111/j.1365-246X.2004.02452.x.
265
Hamiel, Y., and Y. Fialko. 2007. Structure and mechanical properties of faults in the
North Anatolian Fault system from InSAR observations of coseismic
deformation due to the 1999 Izmit (Turkey) earthquake, J. Geophys Res. (in
press).
Henry, C. and Das, S. 2001. Aftershock zones of large shallow earthquakes: Fault
dimensions, aftershock area expansion, and scaling relations. Geophys. J.
Int.,147, 272-293.
Herece and Akay. 2003. Atlas of North Anatolian Fault. Special Publication series-2.
MTA, Ankara, Turkey.
Hickman, H. S. 1991. Stress in the lithosphere and the strength of faults, Rev. of
Geophys. suppl., 759-775.
Horton, R. E. 1945. Erosional development of streams and their drainage basins:
hydrophysical approach to quantitative morphology. Geol. Soc. Am. Bull.
56:275-370
Hugget, R. & Cheesman, J. 2002. Topography and environment. Prentice Hall.,
London
[38] Jennings, C.W., Strand, R.G. and T.H. Rogers (compilers). 2001. Geological
map of California, Calif. Div. Mines and Geol., American Geological
Institute.
Johnson., A.V., Fleming, R.W., Cruikshank, K.M., Martosudarmo, S.Y., Johnson,
N.A., Johnson, K.M. and Wei, W. 1997. Analecta of structures formed
during the 28 June 1992 Landers=Big Bear, California earthquake
sequence. USGS Open File Report 97-94.
Kahle, J.E. 1979. Geology and fault activity of the San Andreas fault zone between
Quail Lake and Three Points, Los Angeles County, California. California
Division of Mines and Geology open-file report 79-3 LA, 42 p., 5 plates,
map scale 1:12,000.
266
Kenny, M. 2000. Quaternary uplift of the eastern San Gabriel Mountains; a case for
crystalline basement rock folding. 2000 AAPG Pacific Section and Western
Region Society of Petroleum Engineers meeting; abstracts AAPG Bulletin
84, 6, 878. DOI: 10.1306/A9673770-1738-11D7-8645000102C1865D.
Kirchner JW. 1993. Statistical inevitability of Horton’s laws and the apparent
randomness of stream channel networks. Geology, 21: 591–594.
Kondo, H., Awata, Y., Emre, Ö., Do ğan, A., Özalp, S., Tokay, F., Yıldırım, C.,
Yoshioka, T. ve Okumura, K. 2005. Slip distribution, fault geometry and
fault segmentation of the 1944 Bolu-Gerede Earthquake rupture, North
Anatolian Fault, Turkey. Bull. Seism. Soc. of Am. 95, 4,1234-1249.
[48] Korneev, V.A., R. M. Nadeau and T.V. McEvilly. 2003. Seismological Studies
at Parkfield IX: Fault-zone Imaging using Guided Wave Attenuation, Bull.
Seism. Soc. Am., 93, 1415-1426,.
Laubach, S.E., 1997. A method to detect natural fracture strike in sandstones. Am.
Assoc. Petroleum Geologist Bull. 81, 604–623.
Le Pichon X., Kreemer C. and Chamot-Rooke, N. 2005. Asymmetry in elastic
properties and the evolution of large continental strike-slip faults. J.
Geophys Res.,110, Art. No. B03405 MAR 19.
[34] Lewis, M., Peng, Z., Ben-Zion, Y. and Vernon, F. 2005. Shallow Seismic
Trapping Structure in the San Jacinto Fault Zone near Anza, California,
Geophys. J. Int., 162, 867–881, doi:10.1111/j.1365-246X.2005.02684.x.
[35] Lewis, M.A, Y. Ben-Zion and J. McGuire. 2007. Imaging the deep structure of
the San Andreas Fault south of Hollister with joint analysis of fault-zone
head and direct P arrivals, Geophys. J. Int., doi: 10.1111/j.1365-
246X.2006.03319.x.
Lindvall, S.C., Rockwell, T.K., Dawson, T.E., Helms, J.G., and Bowman, K.W.
2002. Evidence for two surface ruptures in the past 500 years on the San
Andreas Fault at Frazier Mountain, California. Bull. Seism. Soc. Am., 92, 7,
2689-2703.
267
Lockner, D., Naka, H., Tanaka, H., Ito, H., and Ikeda. R. 1999. Permeability and
Strength of Core Samples from the Nojima Fault of the 1995 Kobe
Earthquake, in Proc. of Internat. Wrkshp on the Nojima Fault Core and
Borehole Data Analysis, Tsukuba, Japan, Nov 22-23, USGS Open File
Report 00-129, H. Ito, K. Fujimoto, H. Tanaka, and D. Lockner, editors, pp.
147-152.
[6] Lomnitz-Adler, J. 1991. Model for steady friction, J. Geophys. Res., 96, 6121-
6131.
Luo, W., and T. F. Stepinski. 2006. Topographically derived maps of valley
networks and drainage density in the Mare Tyrrhenum quadrangle on Mars.
Geophysical Research Letters, 33, L18202, doi:10.1029/2006GL027346.
[32] Lutter, W.J., Fuis,G.S., Ryberg,T., Okaya, D.A., Clayton, R.W., Davis, P.M.,
Prodehl, C., Murphy, J.M., Langenheim, V.E., Benthien, M.L., Godfrey,
N.J., Christensen, N.I., Thygesen, K., Thurber, C.H., Simila, G., and Keller,
G.R. 2004. Upper crustal structure from the Santa Monica Mountains to the
Sierra Nevada, Southern California: tomographic results from the Los
Angeles Regional Seismic Experiment, Phase II (LARSE). Bull. Seismol.
Soc. Amer., 94, 619-632.
Lyakhovsky, V., Ben-Zion, Y. & Agnon, A. (1997). Distributed damage, faulting
and friction, J. Geophys. Res., 102, 27 635–27 649.
Lyakhovsky, V., Ben-Zion, Y. & Agnon, A. (2005). A visco-elastic damage
rheology and rate- and state-dependent friction, Geophys. J. Int., 161, 179–
190, doi: 10.1111/j.1365-246X.2005.02583.x.
Magistrale, H. and Sanders, C. 1995. P wave image of the Peninsular Ranges
batholith, southern California. Geophys. Res. Lett., 22, 2549-2552.
Mai, P.M. 2004. SRCMOD: a database of finite-source rupture models,
www.seismo.ethz.ch/srcmod.
Matthews, V. 1976. Correlation of the Pinnacles and Neenach Volcanic Formations
and their bearing on the San Andreas Fault problem. American Assoc.
Petro. Geol. Bull. V. 60, 2128-2141.
268
McGarr, A., Spottiswoode, S.M., Gay, N. C. and Ortlepp, W. D. (1979).
Observations relevant to seismic driving stress, stress drop, and efficiency,
J. Geophy. Res. B, 84, 2251-2261.
McGuire, J. and Ben-Zion, Y. 2005. High-resolution imaging of the Bear Valley
section of the San Andreas Fault at seismogenic depths with fault-zone head
waves and relocated seismicity, Geophys. J. Int., 163, 152-164, doi:
10.1111/j.1365-246X.2005.02703.x.
McGuire, J. J., Zhao, L., and Jordan, T.H. 2002. Predominance of Unilateral Rupture
for a Global Catalog of Large earthquakes. Bull. Seism. Soc. Am., 92, 3309-
3317.
McKenzie, D. 1972. Active Tectonics of the Mediterranean Region. Geophys. J. Int.
30 (2), 109–185. doi:10.1111/j.1365-246X.1972.tb02351.x.
Mekel, J. F. M., 1977. The use of Aerial Photographs and other Images in Geological
Mapping. ITC Textbook of Photointerpretation ITC, Enschede, The
Netherlands.
[5] Melosh, H. J. 1979. Acoustic fluidization: A new geological process?, J.
Geophys. Res., 84, 7513-7520.
Merifield, P.M., Rockwell, T.K., and Loughman, C.C. 1991, A slip rate based on
trenching studies, San Jacinto fault zone near Anza, California: in
Engineering Geology and Geotechnical Engineering, no. 27 (James
McCalpin, ed.), pp. 28-1 - 28-21.
[49] Michael, A. and Y. Ben-Zion. 1998. Inverting Fault Zone Trapped Waves with
a Genetic Algorithm, EOS, 79, F584.
Miller, W. E. and Downs, T. 1971. A middle Pliocene fauna from Hungry Valley,
southern California. Geological Society of America, Abstracts with
programs, 3, 2, 160-161.
[9] Miller, S. A., A. Nur, and D. L. Olgaard 1996. Earthquakes as a coupled shear
stress high pore pressure dynamical system, Geophys. Res. Lett., 23(2),
197–200.
269
Mizoguchi K., Hirose, T. and Shimamoto, T. 2000. Internal and permeability
structures of Nojima fault zone: data correlation from surface and core
samples.
Mizuno, K., Hattori, H., Sangawa, A. and Takahashi, Y. 1990. Geology of the
Akashi district, quadrangle-series. (in Japanese with English abstract), scale
1:50,000, 90pp., Geol. Surv. Jpn., Tsukuba, Jpn.
Monzawa, N. and Otsuki, K. 2003. Comminution and fluidization of granular fault
materials: implications for fault slip behavior. Tectonophysics
367, 1-2, 127-143. doi:10.1016/S0040-1951(03)00133-1
Moody, J.A. and Kinner, D.A. 2005. Spatial structures of stream and hillslope
drainage networks following gully erosion after wildfire. Earth Surface
Processes and Landforms Earth Surf. Process. Landforms. Published online
in Wiley InterScience (www.interscience.wiley.com). DOI:
10.1002/esp.1246.
Moore, I. D., Grayson, R. B. and A. R. Ladson. 1992. Digital terrain modeling: a
review of hydrological, geomorphological, and biological applications. In
Beven, K.J., and Moore, I.D. (eds.) Terrain Analysis and Distributed
Modeling in Hydrology, John Wiley & Sons, New York. p.7-34.
Moore, D.E. Lockner, D.A. 1995. The role of microcracking in shear-fracture
propagation in granite. J. Struct. Geol. 17, 95-114.
Morton, D.M. and Matti, J.C. 1987. The Cucamonga fault zone, geologic setting and
Quaternary history. USGS Professional Paper, 1339.
MTA.2002.1/500000 scale Geological Map of Sinop Quadrangle, TURKEY.
MTA.2002.1/500000 scale Geological Map of Zonguldak Quadrangle, TURKEY.
Noble, L.F. 1954. Geology of the Valyermo Quadrangle and vicinity, Map GQ-50,
USGS.
270
Nogami, M. 1995. Geomorphometric measures for digital elevation models.
Zeitschrift für Geomorphologie, Supplementband, v. 101, 53-67,
Oakeshott, G.B. 1971. Geology of the epicentral area. Calif. Div. Mines Geol. 196,
19-30.
Oglesby, D. D., R. J. Archuleta, and S. B. Nielsen. 1998. Earthquakes on dipping
faults: the effects of broken symmetry, Science 280, 1055–1059.
Oglesby, D. D., R. J. Archuleta, and S. B. Nielsen. 2000. Dynamics of dip-slip
faulting: explorations in two dimensions, J. Geophys. Res.105, 13,643–
13,653.
Ohtani, T., Fijimoto, K., Ito, H., Tanaka, H., Tomida, N. and Higuchi, T. 2000. Fault
rocks and past to recent fluid characteristics from the borehole survey of the
Nojima rupture in the 1995 Kobe earthquake, southwest Japan. J. Geophys.
Res.,105, B7, 16,161 – 16,171.
[43] Oskin, M., and J. Stock. 2003. Marine incursion synchronous with plate-
boundary localization in the Gulf of California, Geology, Vol. 31, p. 23-26.
Peng, Z. and Y. Ben-Zion. 2004. Systematic analysis of crustal anisotropy along the
Karadere-Düzce branch of the north Anatolian fault, Geophys. J. Int., 159,
253-274, doi: 10.1111/j.1365-246X.2004.02379.x.
Peng, Z. and Ben-Zion, Y. 2006. Temporal changes of shallow seismic velocity
around the Karadere-Duzce branch of the north Anatolian fault and strong
ground motion, Pure Appl. Geophys., 163, 567-599, DOI 10.1007/s00024-
005-0034-6.
[50] Peng, Z., Ben-Zion, Y., Michael, A. J. and Zhu, L. 2003. Quantitative analysis
of seismic trapped waves in the rupture zone of the 1992 Landers,
California earthquake: Evidence for a shallow trapping structure, Geophys.
J. Int., 155, 1021-1041.
[24] Poliakov, A. N., B. R. Dmowska and J. R. Rice. 2002. Dynamic shear rupture
interactions with fault bends and off-axis secondary faulting, J. Geophys.
Res., 107, B11, doi:10.1029/2001JB000572.
271
[42] Powell, R.E. and Weldon, R.J. 1992. Evolution of the San Andreas Fault. Annu.
Rev. Earth Planet. Sci., 20, 431-468.
[22] Power, W. L., T. E. Tullis and J. D. Weeks. 1988. Roughness and wear during
brittle faulting, J. Geophys. Res., 93, 15,268-15,278.
Priest, S.D. 1993. Discontinuity analysis for rock engineering. Chapman and Hall,
New York, 473 pp.
Ramirez, V.R. 1983. Hungry Valley Formation; evidence for 220 kilometers of post
Miocene offset on the San Andreas fault. In Andersen and Rymer (eds)
Tectonics and sedimentation along faults of the San Andreas System, Soc.
Econ. Paleontol. and Mineral., 33-44,.
[13] Ranjith, K. and Rice, J.R. 2001. Slip dynamics at an interface between
dissimilar materials. J. Mech. Phys. Solids.49 (2): 341-361.
Ravenaugh, J. 2000. The relation of crustal scattering to seismicity in southern
California, J. Geophys. Res. 105, 25,403-25,422.
Reches, Z. and Lockner, D.A. 1994. The nucleation and growth of faults in brittle
rocks, J. Geophys. Res., v. 99, p. 18,159-18,173.
[44] Reches, Z. and T.A. Dewers. 2005. Gouge Formation by Dynamic Pulverization
During Earthquakes. Earth Planet. Sci. Lett., 235, 1-2, 361-374.
Rengers, N. 1981. Special Application of Photographs in Engineering Geology.
International Institute for Aerospace Surveys and Earth Science (ITC)
Pamphlet Gol., 11, 51-70.
[28] Rice, J. R., Sammis, C. G. and Parsons, R. 2005. Off-fault secondary failure
induced by a dynamic slip pulse. Seismol. Soc. Am. Bull., 95, 109-134.
[7] Rice, J. R. 1992. Fault stress states, pore pressure distributions, and the weakness
of the San Andreas fault, in Fault Mechanics and Transport Properties of
Rocks, pp. 475- 503, Academic, San Diego, Calif..
272
Ritter, D.F., Kochel, R.C. and Miller, J.R. 2002. Process Geomorphology. 4
th
Edition. Waveland Press, Inc.
Rockwell, T.K., Loughman, C. and Merifield, P. 1990. Late Quaternary rate of slip
along the San Jacinto fault zone, Southern California. J. Geophys. Res., 95,
6, 8593-8605.
Rockwell, T., Seitz, G., Ragona, D., Dawson, T. Geoff Faneros1, Danielle Verdugo,
and O. Altangerel1, 2005, Investigation of segment controls on the rupture
history of the southern San Jacinto fault, Seismo. Res. Letters, v. 76 p. 254
[37] Ross, D.C. 1969. Map showing recently active breaks along the San Andreas
Fault between Tejon Pass and Cajon Pass, Southern California. USGS
Miscellaneous Geologic Investigation, map I-553.
Rubin, C., Lindvall, S. and Rockwell, T.K. 1998. Evidence for large earthquakes in
metropolitan Los Angeles. Science. 281. 298-402.
Rubin, A. 2002. Aftershocks of microearthquakes as probes of the mechanics of
rupture, J. Geophys. Res., 107, 10.1029/2001JB0000496.
Rubin, A. and D. Gillard. 2000. Aftershock asymmetry/rupture directivity along
central San Andreas fault microearthquakes, J. Geophys. Res., 105, 19,095-
19,109.
Sagy, A., Brodsky, E.E., Axen, G.J. 2007. Evolution of fault-surface roughness with
slip. Geology: Vol. 35, No. 3 pp. 283–286.
Sammis, C.G., King, G.C. and Biegel, R.L. 1987. The Kinematics of Gouge
Deformation, Pure and Appl. Geophys.125, 777–812.
Sammis, C. G., and G. C. P. King, Mechanical origin of power law scaling in fault
zone rock, Geophys. Res. Lett, 34, L04312, doi:10.1029/2006GL028548,
2007.
Sammis, C. G. and Y. Ben-Zion. 2007. The Mechanics of Grain-Size Reduction in
Fault Zones, J. Geophys. Res., accepted for publication.
273
Sanders, C.O. and Kanamori, H. 1984. A seismotectonic analysis of the Anza
seismic gap, San Jacinto fault zone, Southern California. J. Geophys. Res.,
89, 5873-5890.
Scholz, C.H. 1987. Wear and gouge formation in brittle faulting. Geology 15, 493–
495.
Scholz, C.H., Dawers, N.H., Yu, J.Z., Anders, M.H., Cowie, P.A. 1993. Fault growth
and fault scaling laws: preliminary results. J. Geophys. Res. 98, 21951–
21961.
Scholz, C.H. 2002. The mechanics of earthquakes and faulting. New York,
Cambridge University Press, 2
nd
edition, 471 pp.
[39] Schubnel, A., personal comm., (2005).
Schulz, S. E. and Evans, J.P. 1998. Spatial variability in microscopic deformation
and composition of the Punchbowl fault, Southern California: implications
for mechanisms, fluid-rock interaction, and fault morphology.
Tectonophysics., 295, 223-244.
[53] Schultz, S.E. and J.P. Evans. 2000. Mesoscopic structure of the Punchbowl
Fault, Southern California and the geologic and geophysical structure of
active strike slip faults. J. Struct. Geol., 913-930.
Schwartz, D.P. and Weldon, R.J. 1986. Late Holocene slip rate on the Mojave
segment of the San Andreas fault zone, Littlerock, Ca; preliminary results.
EOS Trans. Amer. Geophys. Union., 67. Abstract.
[30] Scott, J.S., Masters, T.G. and Vernon, F.L. 1994. 3-D velocity structure of the
San Jacinto fault zone near Anza, California – I. P waves. Geophys. J. Int.,
119, 611-626.
Sengor, A.M.C., Tuysuz, O., Imren, C., Sakınc¸ M., Eyidogan, H, Gorur, N., Le
Pichon, X. and Rangin, C. 2005. The North Anatolian Fault: A new look.
Annu. Rev. Earth Planet. Sci. 33, 37–112.
274
[33] Shapiro, N.M., Campillo, M., Stehly, L., and Ritzwoller, M.H. 2005. High-
resolution surface-wave tomography from ambient seismic noise. Science,
307, 1615-1618
Sharp, R. 1967. The San Jacinto fault in southern California: Geol. Soc. Amer. Bull.,
78, 705-730.
Shi, B., A. Anooshehpoor, J. N. Brune, and Y. Zeng. 1998. Dynamics of thrust
faulting: 2-D lattice model, Bull. Seism. Soc. Am. 88, 1484–1494.
[17] Shi, Z. and Y. Ben-Zion. 2006. Dynamic rupture on a bimaterial interface
governed by slip-weakening friction, Geophys. J. Int., 164, doi:
10.1111/j.1365-246X.2006.02853.x.
Shreve RL. 1966. Statistical law of stream numbers. Journal of Geology 74: 17–37.
Sibson, R.H. 1977. Fault rocks and fault mechanisms. J. Geol. Soc. London. 133,
191-213.
[21] Sibson, R.H. 1989. Earthquake faulting as a structural process. J. Struct. Geol.,
11, 1-14.
Sibson, R. H. 2003. Thickness of the seismic slip zone, Seismol. Soc. Am. Bull., 93,
1169– 1178.
Sibson, R.H.. 1986. Brecciation processes in fault zones: Inferences from earthquake
rupturing. Pure Appl. Geophys, v. 124, 1-2, 159-175.
DOI: 10.1007/BF00875724,
Sieh, K. H. 1978a. Slip along the San Andreas Fault associated with the great 1857
earthquake. Bull. Seism. Soc. Am., 68, 5, 1421-1447.
Sieh, K.H. 1978b. Central California foreshocks of the great 1857 earthquake. Bull.
Seism. Soc. Am., 68, 6, 1731-1749.
275
Sisk, M., M. Stillings, T. Rockwell, G. Girty, O. Dor, and Y. Ben-Zion. 2006.
Pulverized Rock Along Faults of the San Andreas System in Southern
California, SCEC Annual Meeting, Palm Springs, California.
Steck, A. and J. Hunziker, 1994. The Tertiary structural and thermal evolution of the
Central Alps; compressional and extensional structures in an orogenic belt.
Tectonophysics, 238, 1-4, 229-254.
Stein, R.S., Barka, A.A. and Dieterich, J. H. 1997. Progressive failure on the North
Anatolian fault since 1939 by earthquake stress triggering. Geophys. J. Int.,
128, 594-604.
Stierman, D. J. 1984. Geophysical and geological evidence for fracturing, water
circulation, and chemical alteration in granitic rocks adjacent to major
strike-slip faults, J. Geophys. Res., 89, 5849-4857.
Strahler, A. N. 1952. Dynamic basis of geomorphology. Geol. Soc. Am. Bull. 63:
923-938.
Strahler, A. N. 1957. Quantitative analysis of watershed geomorphology.
Transactions of the American Geophysical Union (38):913-920.
Strahler, A. N. 1964. Quantitative geomorphology of drainage basins and channel
networks. in Handbook of Applied Hydrology, Ven Te Chow (Editor).
McGraw Hill.
Sylvester, A. G. 1988. Strike-slip faults: Geol. Soc. Am. Bull. v. 100, p. 1666–1703.
Sylvester, A. G., and R. R. Smith. 1976. Tectonic transpression and basement-
controlled deformation in the San Andreas fault zone, Salton trough,
California: AAPG Bulletin, v. 60, p. 74–96.
Tanaka, H., Fujimoto, K., Ohtani, T. and Ito, H. 2001. Structural and chemical
characteristics of shear zones in the freshly activated Nojima Fault, Awaji
Island, southwest Japan. J. Geophys. Res.106, B5, 8789 – 8810.
276
Templeton, E. L. and Rice, J. R. 2006. Extent and Distribution of Off-Fault Plasticity
during Seismic Rupture Including Bimaterial Effects. Eos Trans. AGU,
87(52), Fall Meet. Suppl., Abstract S34A-01.
Tucker, Allan Z., Dolan, James F. 2001. Paleoseismologic Evidence for a >8 Ka Age
of the Most Recent Surface Rupture on the Eastern Sierra Madre Fault,
Northern Los Angeles Metropolitan Region, California. Bulletin of the
Seismological Society of America. 91: 232-249.
Tuttle, O.F. 1949. Structural petrology of planes of liquid inclusions, J. Geology, 57,
331-356.
van der Pluijm, B.A., Hall, C.M., Vrolijk, P.J., Pevear, D.R. and Covey, M.C. 2001.
The dating of shallow faults in the Earth's crust. Nature., 412, 6843, 172-
175.
Vermilye, J.M., and Schulz C.H. 1998. The Process zone: a microstructural view of
fault growth. J. Geophys. Res., 103, 6, 12,223-12,237.
Vermilye, J.M., and Schulz C.H. 1999. Fault propagation and segmentation: insight
from the microstructural examination of a small fault. J. Struc. Geol., 21,
1623-1636.
Weber, F.H.Jr. 1999. Right-lateral displacement of Pleistocene sedimentary deposits
along the San Andreas Fault, Palmdale to Cajon Pass, Southern California.
USGS, Final technical report, 103 p. 4 sheets.
[10] Weertman, J. 1980. Unstable slippage across a fault that separates elastic media
of different elastic constants. J.Geophys. Res., 85, 1455-1461.
Weertman, J., 2002. Subsonic type earthquake dislocation moving at approximately
2 ×shear wave velocity on interface between half spaces of slightly
different elastic constants, Geophys. Res. Lett., 29(10),
doi:10.1029/2001GL013916.
Weldon, R.J. and Fumal, T.E. 2005. Slip rate of the San Andreas Fault near
Littlerock, California. 2005 SCEC Annual meeting, Proceedings and
Abstracts, XV.
277
Weldon, R. J., II, Fumal, T. E., Powers, T. J., Pezzopane, S. K., Scharer, K. M.,
Hamilton, J. C. 2002. Structure and Earthquake Offsets on the San Andreas
Fault at the Wrightwood, California, Paleoseismic Site. Bull. Seis. Soc.
Am., 92: 2704-2725.
Weldon, R.J., Meisling, K.E. and Alexander, J. 1993. A speculative history of the
San Andreas fault in the central Transverse ranges, California. In: Powell,
R.E., Weldon, R.J. and J.C. Matti (Eds). The San Andreas Fault System:
Displacement, Palinspastic Reconstruction, and Geologic Evolution.
Geological Society of America Memoir., 178, 161-198.
Wesnousky, S.G. 1988. Seismological and structural evolution of strike-slip faults.
Nature 335, 340 – 343. doi:10.1038/335340a0.
Wibberley C.A.J. and T. Shimamoto (2003). Internal structure and permeability of
major strike-slip fault zones: the Median Tectonic Line in Mie Prefecture,
Southwest Japan. J. Struct. Geol. 25, 1, 59-78(20).
Wilson, J. E., Chester, J. S. & Chester, F. M. 2003. Microfracture analysis of fault
growth and wear processes, Punchbowl Fault, San Andreas System,
California. J. Struct. Geol. 25, 1855–-1873.
[2] Wilson, B., Dewers, T., Reches, Z., and Brune, J. 2005. Particle size and
energetics of gouge from earthquake rupture zones. Nature, 434, 749-752.
[49] Wilson, B. 2004. Meso- and Micro-Structural Analysis of the San Andreas Fault
at Tejon Pass, California, Unpublished Masters Thesis, University of
Oklahoma, Norman, Oklahoma, 279 pp.
Woodburne, M.O. 1975. Cenozoic stratigraphy of the transverse ranges and adjacent
areas, Southern California. GSA, Special Paper.
Xia, K., Rosakis, A.J., Kanamori, H. and Rice, J.R. 2005. Laboratory earthquakes
along inhomogeneous faults: directionality and supershear, Science, 308,
681-684.
278
[23] Yamashita, T. 2000. Generation of microcracks by dynamic shear rupture and
its effects on rupture growth and elastic wave radiation, Geophys. J. Int.,
143, 395–406.
Zhang, H.P., Liu, S.F., Yang, N., Zhang, Y.Q., Zhang., G.W. 2006. Geomorphic
characteristics of the Minjiang drainage basin (eastern Tibetian) and its
tectonic implications: New insights from a digital elevation model study.
Island Arc 15, 239-250.
279
Appendices
Appendix 1: Shapiro-Wilk and Shapiro-Francia W tests for normality of the
distribution of FIPL results in Chapter 4.
Shapiro-Wilk W test for normal data
Variable ObsW V z Prob>z
8E-b_104 104 0.82 15.08 6.03 0
set_10 40 0.88 4.55 3.19 0.00071
set_9 40 0.87 5.03 3.40 0.00034
set_8 40 0.83 6.54 3.95 0.00004
set_7 40 0.85 5.94 3.75 0.00009
set_6 40 0.83 6.54 3.95 0.00004
set_5 40 0.76 9.47 4.73 0
set_4 40 0.87 5.09 3.42 0.00031
set_3 40 0.86 5.52 3.60 0.00016
set_2 40 0.75 9.69 4.78 0
set_1 40 0.82 7.00 4.10 0.00002
8E-a 43 0.94 2.64 2.05 0.02006
8E-c 43 0.74 10.89 5.05 0
31 47 0.88 5.58 3.65 0.00013
28 47 0.83 7.63 4.32 0.00001
27 46 0.75 10.87 5.06 0
6a 41 0.92 3.18 2.44 0.00737
280
Shapiro-Francia W' test for normal data
Variable ObsW' V' z Prob>z
8E-b_104 104 0.82 16.60 5.35 0.00001
set_10 40 0.88 5.16 3.01 0.00132
set_9 40 0.88 5.43 3.10 0.00098
set_8 40 0.83 7.36 3.63 0.00014
set_7 40 0.85 6.58 3.43 0.0003
set_6 40 0.83 7.27 3.61 0.00015
set_5 40 0.76 10.69 4.27 0.00001
set_4 40 0.87 5.47 3.11 0.00093
set_3 40 0.86 6.23 3.34 0.00042
set_2 40 0.75 10.99 4.32 0.00001
set_1 40 0.82 7.94 3.76 0.00008
8E-a 43 0.94 2.62 1.80 0.0359
8E-c 43 0.73 12.23 4.51 0.00001
31 47 0.88 6.15 3.35 0.00041
28 47 0.83 8.30 3.87 0.00005
27 46 0.75 11.94 4.49 0.00001
6a 41 0.92 3.61 2.38 0.00871
281
Appendix 2: Bitmap images used for FIPL measurements in Chapter 4.
6a
282
27
283
28
284
31
285
8E-a
286
8E-c
287
8E-b (set 1)
288
Abstract (if available)
Abstract
Structural symmetry properties were mapped across faults of the San Andreas Fault (SAF) and North Anatolian Fault (NAF) systems, at various scales and several sites on each fault. Fractures on a fault-core scale, subsidiary faults and fault rocks on a fault-zone scale and pulverized rocks on a damage-zone scale show systematically asymmetry. On the SAF, San Jacinto and Punchbowl faults the northeast side is more damaged. On the NAF 1943 and 1944 rupture sections the south and north sides, respectively, are more damaged. Asymmetric erosion patterns along the NAF including locations of river valleys with respect to the fault and contrast in drainage density and other morphometric parameters across the fault, are consistent with the geologically mapped structural asymmetry. These asymmetric patterns are compatible with preferred rupture directions northwestward on faults of the SAF system, and eastward and westward on the 1943-1944 rupture sections of the NAF, respectively (as occurred in these two earthquakes). Tomographic studies show that the northeast side of the SAF and the San Jacinto fault have faster seismic velocities at depth. Significant damage content in sedimentary rocks of the Juniper Hills formation near the SAF in the central Mojave section indicates that dynamic generation of damage can occur close to the Earth surface, in agreement with other indications for minimal exhumation of damaged fault zone rocks. An asymmetric shallow damage structure correlated with the velocity structure at depth is a predicted outcome for rupture along a bimaterial interface (Ben-Zion and Shi, 2005). Microfractures in the Juniper Hills rocks near the fault, orientated preferably normal to its strike, are compatible with the transient stress field associated with seismic slip events on frictional rough surfaces (Chester and Chester, 2000). The damage fabric is anisotropic, rich with compressional features, and therefore not compatible with an absolute tension.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Multi-scale damage signatures across major strike-slip faults
PDF
Multi-scale imaging of major fault zones in Southern California
PDF
Detailed properties of seismic waveforms and earthquake sources in Southern California
PDF
Multi-scale imaging and monitoring of crustal and fault zone structures in southern California
PDF
Dynamic rupture processes and seismic radiation in models of earthquake faults separating similar and dissimilar solids
PDF
Determining fault zone structure and examining earthquake early warning signals using large datasets of seismograms
PDF
High-resolution imaging and monitoring of fault zones and shallow structures: case studies in southern California and on Mars
PDF
Elements of seismic structures near major faults from the surface to the Moho
PDF
Volumetric interactions between major ruptures and fault zones illuminated by small earthquake properties
PDF
Heterogeneity of earthquake stress drops, focal mechanisms and active fault zones
PDF
Applying automated techniques to large seismic datasets for systematic analyses of phases, source, and structure
PDF
Stress-strain characterization of complex seismic sources
PDF
Analysis of waveform and catalog data of aftershocks for properties of earthquakes and faults
PDF
Multi-scale imaging of the fault zone velocity structure: double-difference tomography, inversion of fault zone headwaves, and fault zone sensitivity kernels
PDF
Integration and validation of deterministic earthquake simulations in probabilistic seismic hazard analysis
PDF
Interaction between dynamic ruptures and off-fault yielding characterized by different rheologies
PDF
Seismic velocity structure and spatiotemporal patterns of microseismicity in active fault zones of Southern California
PDF
Statistical analyses of ambient seismic noise spectra with applications to detecting imperfectly diffuse wave field and deriving attenuation and phase velocity information
PDF
Spatiotemporal variations of stress field in the San Jacinto Fault Zone and South Central Transverse Ranges
PDF
The impact of faulting and folding on earthquake cycles in collisional orogens
Asset Metadata
Creator
Dor, Ory
(author)
Core Title
Symmetry properties, pulverized rocks and damage architecture as signatures of earthquake ruptures
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Geological Sciences
Publication Date
07/20/2007
Defense Date
05/08/2007
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
dynamic rupture,Earthquakes,faults,Fractures,geological mapping,OAI-PMH Harvest,pulverized rocks
Language
English
Advisor
Ben-Zion, Yehuda (
committee chair
), [illegible] (
committee member
), [illegible], J. (
committee member
), Brune, James M. (
committee member
), Rockwell, Thomas K. (
committee member
)
Creator Email
dor@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m608
Unique identifier
UC1313368
Identifier
etd-Dor-20070710 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-520146 (legacy record id),usctheses-m608 (legacy record id)
Legacy Identifier
etd-Dor-20070710.pdf
Dmrecord
520146
Document Type
Dissertation
Rights
Dor, Ory
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
dynamic rupture
geological mapping
pulverized rocks