CONTROLS ON FAULT GEOMETRY DURING EARLY STAGES OF ...
Transcript of CONTROLS ON FAULT GEOMETRY DURING EARLY STAGES OF ...
CONTROLS ON FAULT GEOMETRY DURING EARLY STAGES OF EXTENSION
IN THE LARKSPUR HILLS, NORTHWEST BASIN AND RANGE
__________________________________
A Thesis
Presented to
The Graduate Faculty
Central Washington University
___________________________________
In Partial Fulfillment
of the Requirements for the Degree
Master of Science
Geology
___________________________________
by
Diana Jean Strickley
May 2014
ii
CENTRAL WASHINGTON UNIVERSITY
Graduate Studies We hereby approve the thesis of
Diana Jean Strickley Candidate for the degree of Master of Science APPROVED FOR THE GRADUATE FACULTY _________________________________________ Dr. Anne Egger, Committee Chair _________________________________________ Dr. Jeff Lee _________________________________________ Dr. Walter Szeliga _________________________________________ Dean of Graduate Studies
iii
ABSTRACT
CONTROLS ON FAULT GEOMETRY DURING EARLY STAGES OF EXTENSION
IN THE LARKSPUR HILLS, NORTHWEST BASIN AND RANGE
by
Diana Jean Strickley
May 2014
Detailed analyses of normal faults in the Larkspur Hills, CA-NV, northwest Basin
and Range, offer insight into factors controlling normal fault initiation, growth, and
distribution. N-trending faults in the southern portion of the study area share trends of
major range-bounding structures and Pliocene linear volcanic vents; in contrast, NNW-
and NNE- trending faults dominate further north and into south-central Oregon. Stress
analyses and comparison with experimental and field data suggest that preexisting
structures control faults in the northern Larkspur Hills, while faults form perpendicular
to 3 in the southern hills. The change in fault orientations is abrupt, occurring across a
major NNE fault. A regional transition is thus captured within the Larkspur Hills,
suggesting they overlie a structural boundary at depth that separates isotropic crust
from crust with a pre-existing NW-trending fabric. This has implications for better
understanding of local and regional structural controls on subsurface hydrothermal flow
paths.
iv
TABLE OF CONTENTS
Chapter Page
I INTRODUCTION ............................................................................................. 1
II JOURNAL ARTICLE ......................................................................................... 7
Introduction ............................................................................................ 8 Background ........................................................................................... 14 Methods ................................................................................................ 23 Results ................................................................................................... 27 Discussion .............................................................................................. 44 Conclusions ........................................................................................... 57 Acknowledgements ............................................................................... 59 References ............................................................................................. 60
v
LIST OF FIGURES
Figure Page Page
1 Map of the Basin and Range. Simplified from Colgan (2013) .............................. 9
2 Map of the northwest Basin and Range, with current and previous
study areas noted .............................................................................................. 10
3 Map of Larkspur Hills, northwest Basin and Range ........................................... 13
4 Block diagrams and associated map views illustrating linking of en echelon
faults (4a & 4b) and the similarity in surface traces for branching faults
vs. individual faults (4c) ..................................................................................... 16
5 Close-up map of domain I. Legend for domain close-up map follows
(5a) ..................................................................................................................... 20
6 Rose diagrams showing fault segment orientations for each domain .............. 30
7 Graphs of one prominent fault per domain and domain boundary. Circles
indicate maximum scarp height and azimuth for each segment, plotted
at the midpoint of each segment ...................................................................... 31
8 Close-up map of domain II ................................................................................. 32
9 Close-up map of domains III and IV ................................................................... 34
10 Close-up map of domain V ................................................................................. 37
11 Rose diagrams showing fault segment orientations for domain boundary
faults .................................................................................................................. 39
12 Cross-sections A-A’ through E-E’ corresponding to section lines shown on
Larkspur Hills map .............................................................................................. 42
13 Comparison of Andersonian and orthorhombic fault geometry ....................... 45
vi
LIST OF FIGURES (CONTINUED)
Figure Page
14 Simplified fault maps comparing results of sandbox models of Bellahsen
and Daniel (2005) and fault traces in the study area ........................................ 48
15 Simplified map showing drill hole locations (grey circles) and hot
springs (white circles) relative to faults ............................................................... 5
1
CHAPTER I
INTRODUCTION
A number of factors influence the formation and evolution of normal faults,
including: the orientation of principal stresses (Anderson, 1951); changes in stress
orientations over time (e.g. Henza et al., 2011; Giba et al., 2012); heterogeneities in the
crust, such as differences in lithology, layering, or grain size (Ackermann et al., 2001;
Crider and Peacock, 2004; Schopfer et al., 2006); and pre-existing structures, e.g. veins,
joints, and faults (Martel et al., 1988; Acocella et al., 2003; Crider and Peacock, 2004;
Bellahsen and Daniel, 2005; Blenkinsop, 2008). Pre-existing structures may promote
faulting, especially in the early stages of extension, by weakening the crust (Buck, 2009,
and references therein). Combined geophysical and geological studies in young active
rifts, such as the East African Rift system, are often able to discern the relative
influences of modern stress orientations vs. pre-existing structures (e.g. Morely, 1995,
and references therein; Modisi et al., 2000).
However, it can be difficult to determine the role of pre-existing structures
during the early development stages in other extensional provinces, such as the North
American Basin and Range, where long-term and recent normal faulting can obscure
earlier structures. The Basin and Range province is characterized by approximately N-S-
striking, normal-fault-bound ranges up to 100 km in length, spaced ~30 km apart, and
separated by similarly-scaled linear basins (Stewart, 1998). Sequential reconstructions of
2
McQuarrie and Wernicke (2005) indicate a total of ~235 km of extension in the Basin
and Range province, with the amount of extension ranging from 200% in the central
region to 50% in the northern region and initiating 15-25 Ma. Extension is most recent
and continues along the eastern and western margins.
As in the rest of the Basin and Range, most of the extension in the northwest
Basin and Range is accommodated by major-offset (vertical offset ≥ 300 m) faults
striking N to NNE (Pezzopane and Weldon, 1993), of which one of the more prominent is
the Surprise Valley fault in northeastern California. Extension is relatively recent, with
evidence that extension along the Surprise Valley fault occurred in two phases, the first
between 14 and 8 Ma and the second phase of rapid uplift beginning at ~3 Ma (Colgan
et al., 2008; Egger and Miller, 2011) and continuing until the present (Personius et al.,
2009). Several studies suggest that the percent of E-W-directed extension decreases
northward along the margin, with a maximum of approximately 15% extension since the
middle Miocene (Pezzopane and Weldon, 1993; Scarberry et al., 2010; Egger and Miller,
2011; Trench et al., 2012). Restored cross-sections indicate 6-7 km of extension across
the 86 km-long Surprise Valley fault, which accommodates most of the 12-15%
extension across the Warner Range region (Egger and Miller, 2011). Extension decreases
northwards, from 3% along the Abert Rim fault (Scarberry et al., 2010) to 0.01% at the
northern termination of the Hart Mountain fault system (Trench et al., 2012), and only
0.004% in the Brothers Fault Zone. Due to the comparatively recent onset and lower
magnitude of extension, as well as the presence of late Miocene to Pliocene volcanic
3
rocks, the northwestern Basin and Range (NWBR) preserves the early stages of
extension, thus is an excellent place to evaluate models for fault initiation and growth in
extensional regimes.
Most of the extension in the northwest Basin and Range is accommodated along
large-offset N- to NNE-striking normal faults. In addition to major offset NNE-striking
faults, the region hosts a pervasive set of NW-NNW-striking faults and fractures with
little to no offset (Donath, 1962; Pezzopane and Weldon, 1993). An early interpretation
of the NW-oriented faults, particularly in the BFZ, was that they were right-lateral strike-
slip forming a transfer zone (Faulds and Varga, 1998) between the northern edge of
Basin and Range extension and unextended area to the north (Lawrence, 1976). This
interpretation was based on mapping lineations from satellite images and not measured
offsets or fault studies in the field. Later field investigations of Pezzopane and Weldon
(1993) found that faults throughout the region were dominantly dip-slip, with negligible
to minor oblique motion. Trench et al. (2012), working in southeast Oregon, also found
abundant evidence for dip-slip motion on the NW-oriented faults, including those within
the BFZ, and sparse evidence of minor strike-slip motion. Scarberry et al. (2010) observe
that the NW-striking faults are more numerous along northern tips of NNE-striking faults
throughout the region, so concluded that the NW-striking set in the Abert Rim area
formed as result of splays along the northward propagating tip of the Abert Rim fault
exploiting a pre-existing NW-oriented fabric. Trench et al. (2012) reached a similar
conclusion by observing cross-cutting relationships between the two fault sets in
4
Oregon, where NNE-striking faults cut the NW-striking set in the south, but the two sets
are mutually cross-cutting in the north. Trench et al. (2012) thus concluded that NW-
striking faults first developed along the fault tips of NNE-striking faults, but the NNE-
striking faults ultimately grew through and cross-cut the NW-striking faults; however,
they also point out that a lack of major NNE-striking faults in or proximal to the Brothers
Fault Zone indicates that the NW-striking faults within that zone developed
independently and were not influenced by NNE-striking structures. Scarberry et al.
(2010) suggest that in the Abert Rim area, the similarity between orientations of the two
distinct middle to late Miocene (16-5 Ma) fault sets and those of early to middle
Miocene (23.8-16.4 Ma) dikes that trend 5-10° and 290-315° indicate that the
development of both fault sets was influenced by a pre-existing fabric .
Unlike the rest of the Basin and Range, the NWBR also hosts a second set of
pervasive, minor-offset (vertical offset < and often << 150 m) normal faults that strike
NW (Pezzopane and Weldon, 1993; Scarberry et al., 2010; Trench et al., 2012). The
origin and nature of these structures is unknown; however, cross-cutting relationships
and fault density along the larger-offset NE-striking structures (Scarberry et al., 2010)
indicates that the NW-striking set may develop as a result of NNE-striking fault growth,
with fault-tip splays reactivating a pre-existing NW-striking structures, similar to the
formation of fault tip dilatational (horsetail) fractures (Trench et al., 2012).
The focus of this study is the Larkspur Hills in northeastern California, which is
overlain by a series of late Miocene to Pliocene basalt flows, with average thicknesses of
5
~1-5 m per flow, which are interbedded with tuffaceous sediments, and cut by normal
faults with strikes ranging from NNW to NNE. Total extension decreases northward from
~4-8% to ≤2%, thus faults are in various early stages of development. Over this relatively
small area of about 550 km2, there are significant differences in fault orientations, from
dominantly N-striking in the south to NNE- and NNW-striking in the north. This
observation leads to the following questions: how do the local fault orientations relate
to the regional stress field over time, and what controls the differences in orientations
from south to north across the field area?
This study involves detailed mapping of faults and linear volcanic features, and
numerical and graphical analyses of fault segmentation, fault and segment orientations,
horizontal offset, and scarp heights, as well as analyses of cross-cutting or truncating
relationships in order to determine possible controls, such as extension direction,
change in extension direction, pre-existing structures, or heterogeneities (e.g.
lithological differences), on the growth and development of normal faults during early
stages of extension in this area.
Better characterization of controlling structures in the Larkspur Hills may provide
an analogue for nearby buried intra-basin faults which control locations of hot springs
throughout the valley. Both the Surprise Valley and the southern Larkspur Hills have
been the object of geothermal exploration (Barker et al., 2005; Benoit et al., 2005;
Bertani, 2007) and studies to define structural features in the subsurface (Egger et al.,
2010; Glen et al., 2013). Local studies (Benoit et al., 2005; Egger, et al. 2014) point to
6
the importance of fault controls on subsurface fluid circulation and the surface location
of hot springs in the Surprise Valley (Glen et al., 2013); this is supported by other studies
that emphasize the importance of fault controls in geothermal systems (Curewitz and
Karson, 1997; Faulds et al., 2010). While hot springs occur on both sides of the basin in
the Surprise Valley, along the east side they do not appear to be located along faults;
however, faults are known to be a primary control on fluid flow (Caine et al., 1996;
Rowland and Sibson, 2004; Faulds et al., 2010). Faults expressed at the surface in the
adjacent Larkspur Hills may serve as an analog for faults buried in Surprise Valley, and
may lend insight into their geometry and controlling structures.
7
CHAPTER II
JOURNAL ARTICLE
8
Controls on fault geometry during early stages of extension in the Larkspur Hills,
northwest Basin and Range
1. Introduction
A number of factors influence the formation and evolution of normal faults,
including: the orientation of principal stresses (Anderson, 1951); changes in stress
orientations over time (e.g. Henza et al., 2011; Giba et al., 2012); heterogeneities in the
crust, such as differences in lithology, layering, or grain size (Ackermann et al., 2001;
Crider and Peacock, 2004; Schopfer et al., 2006); and pre-existing structures, e.g. veins,
joints, and faults (Martel et al., 1988; Acocella et al., 2003; Crider and Peacock, 2004;
Bellahsen and Daniel, 2005; Blenkinsop, 2008). Pre-existing structures may promote
faulting, especially in the early stages of extension, by weakening the crust (Buck, 2009,
and references therein). Combined geophysical and geological studies in young active
rifts, such as the East African Rift system, are often able to discern the relative
influences of modern stress orientations vs. pre-existing structures (e.g. Morely, 1995,
and references therein; Modisi et al., 2000). However, it can be difficult to determine
the role of pre-existing structures during the early development stages in other
extensional provinces, such as the North American Basin and Range (Fig. 1), where long-
term and recent normal faulting can obscure earlier structures.
The Basin and Range province is characterized by approximately N-S-striking,
normal-fault-bound ranges up to 100 km in length, spaced ~30 km apart, and separated
9
Fig. 1. Map of the Basin and Range. Simplified from Colgan (2013).
by similarly-scaled linear basins (Stewart, 1998). Sequential reconstructions of
McQuarrie and Wernicke (2005) indicate a total of ~235 km of extension in the Basin
and Range province, with the percent extension ranging from 200% in the central region
to 50% in the northern region, initiating from 15-25 Ma, with the most recent extension
along the eastern and western margins.
As in the rest of the Basin and Range, most of the extension in the northwest
Basin and Range is accommodated by major-offset (vertical offset ≥ 300 m) faults
striking N to NNE (Pezzopane and Weldon, 1993), of which one of the more prominent is
the Surprise Valley fault in northeastern California. Extension is relatively recent, with
evidence that extension along the Surprise Valley fault occurred in two phases, the first
10
Fig. 2. Map of the northwest Basin and Range, with current and previous study areas noted. Quaternary faults (USGS, 2006) that are discussed in the text are shown in bold and edited to show slip sense. Locations of hot springs in the Surprise Valley are shown with white dots.
11
between 14 and 8 Ma and the second phase of rapid uplift beginning at ~3 Ma (Colgan
et al., 2008) and continuing until the present (Personius et al., 2009). Several studies
suggest that the percent of E-W-directed extension decreases northward along the
margin, with a maximum of approximately 15% extension since the middle Miocene
(Pezzopane and Weldon, 1993; Scarberry et al., 2010; Egger and Miller, 2011; Trench et
al., 2012). Restored cross-sections indicate 6-7 km of extension across the 86 km-long
Surprise Valley fault, which accommodates most of the 12-15% extension across the
Warner Range region (Fig. 2) (Egger and Miller, 2011). Percent extension along single
faults or fault zones decreases northwards, from 3% along the Abert Rim fault
(Scarberry et al., 2010) to 0.01% at the northern termination of the Hart Mountain fault
system (Trench et al., 2012), and only 0.004% in the Brothers Fault Zone (Fig. 2). Due to
the low magnitude of extension, as well as the presence of late Miocene to Pliocene
volcanic rocks, the northwestern Basin and Range (NWBR) preserves the early stages of
extension, thus is an excellent place to evaluate models for fault initiation and growth in
extensional regimes.
Unlike the rest of the Basin and Range, the NWBR hosts a second set of
pervasive, minor-offset (vertical offset < and often << 150 m) faults that strike NW;
while NW-striking faults are not absent in the Basin and Range, in the NWBR they
comprise a distinct set of faults that dominate in number, while the NE-striking faults
are fewer but have the most offset (Fig. 2) (Pezzopane and Weldon, 1993; Scarberry et
al., 2010; Trench et al., 2012). The origin and role that these NW-striking structures play
12
in accommodating strain is unclear; however, cross-cutting relationships and fault
density along the larger-offset NE-striking structures (Scarberry et al., 2010) indicate
that the NW-striking set may develop as a result of NNE-striking fault growth
reactivating pre-existing NW-striking structures, similar to the formation of fault tip
dilatational (horsetail) fractures (Trench et al., 2012).
The focus of this study is the Larkspur Hills in northeastern California (Figs. 2 and
3), which are made up of a series of late Miocene to Pliocene basalt flows interbedded
with tuffaceous sediments cut by normal faults with strikes ranging from NNW to NNE.
The percent extension decreases northward from ~4-8% to ≤2%, thus normal faults are
in various early stages of development. In addition, NW-oriented faults are absent in the
southern Larkspur Hills but become dominant in the north (Fig. 3).
This study involves detailed mapping of faults and linear volcanic features, and
numerical and graphical analyses of fault segmentation, fault and segment orientations,
horizontal offset, and scarp heights, as well as analyses of cross-cutting or truncating
relationships. The results of these analyses were used to determine possible controls,
such as extension direction, change in extension direction, pre-existing structures, or
heterogeneities (e.g. lithological differences), on the growth and development of normal
faults during early stages of extension in this area. Better characterization of controlling
structures in this region is of particular interest because the northwest Basin and Range
is an area of active geothermal exploration, and fault geometries and interactions (e.g.
linkage or overlap of faults) are a primary control on the locations and flow paths of
13
Fig. 3. Map of Larkspur Hills, northwest Basin and Range. Roman numerals denote fault domains. Buried basin faults from Egger and Miller (2011) are inferred from potential field data.
14
hydrothermal fluids in the subsurface (Caine et al., 1996; Rowland and Sibson, 2004;
Faulds et al., 2010).
2. Background
2.1. Normal fault initiation and growth
In an extensional environment, as finite strain accumulates over time it changes
from being diffuse to more localized (e.g. Meyer et al. 2002). Some faults become
inactive while others connect and evolve into larger, through-going structures
(Gawthorpe et al., 2003; Crider and Peacock, 2004; Moriya et al., 2005). As faults evolve,
they undergo changes in growth mechanisms: isolated faults first grow through tip-line
propagation and mode I (open) fracturing (Reches and Lockner, 1994; White and Crider,
2006; Rowland et al., 2007; Herman, 2009). As faults overlap, they begin to link (Fig. 4);
linking is the dominant factor in the development of larger faults rather than growth
along tip-lines (Ferrill et al., 1999; Mansfield and Cartwright, 2001; Crider and Peacock,
2004).
The linking process commonly involves the formation of a relay ramp within the
zone of overlap between segments (Fig. 4a) (Peacock, 2002); further growth of the
faults causes the ramp to fracture, and ultimately break (Soliva and Benedicto, 2004).
Initial breaching of the ramp can occur anywhere along its vertical extent, so faults can
link at depth even when the ramp is intact at the surface (Trudgill and Cartwright, 1994;
Peacock and Parfitt, 2002; Long and Imber, 2012). Once the ramp is fully breached, the
15
two original faults connect to become segments of one longer fault, a process described
in detail by Peacock and Sanderson (1994).
The linking process can also involve fault segments propagating towards each
other, with tension at the tip lines causing the segment tips to at first diverge, then
converge, and ultimately intersect the other fault trace, resulting in a hook shape in map
view (Fig. 4b) (Pollard and Aydin, 1984; Acocella et al., 2000). This type of linking process
also involves a ramp developing in the zone of overlap; however, the faults link directly
rather than through breaching of a ramp (Acocella et al. 2000).
Pre-existing or precursory structures can also influence the orientation of normal
faults and linking structures as they initiate and evolve (Giba et al., 2012; Bellahsen and
Daniel, 2005; Hus et al., 2005; Crider and Peacock, 2004; McClay et al., 2002). Crider and
Peacock (2004) define precursory structures as early structures formed in the same
stress field as current faulting, and pre-existing structures as those which formed in a
stress field unrelated to current faulting; later extensional events can reactivate such
structures at depth, and the rupturing of a large structure through subsequent cover
may appear as segmented normal faults at the surface (Giba et al., 2012) or as non-
optimally oriented normal faults (Morely et al., 2004).
Without three-dimensional data at depth, it can be difficult to determine how
large faults evolved in mature regions, as resulting surface traces can be the same for
different growth mechanisms (Fig. 4c). For example, high-resolution 3D seismic data
from the Taranaki Basin, New Zealand (Giba et al., 2012) and the Bonaparte Basin, NW
16
Australia (Long and Imber, 2012) have shown that faults have complex subsurface
geometries associated with fault growth, linkage, and reactivation. In the absence of
Fig. 4. Block diagrams and associated map views illustrating linking of en echelon faults (4a & 4b) and the similarity in surface traces for branching faults vs individual faults (4c). Diagrams adapted from Ferrill et al. (1999), Acocella et al. (2000), and Giba et al. (2012).
17
seismic data, analogue and numerical models can provide insights into controls on fault
geometries during initial stages of extension (e.g. Crider, 2001; McClay et al., 2002;
Schopfer et al., 2006; Kaven and Martel, 2007; Holland et al., 2011).
2.2. Field studies of the northwest Basin and Range
Most of the extension in the northwest Basin and Range is accommodated along
large-offset N- to NNE-striking normal faults. In addition to major offset NNE-striking
faults, the region hosts a pervasive set of NW to NNW-striking faults and fractures with
little to no offset (Donath, 1962; Pezzopane and Weldon, 1993). An early interpretation
of the NW-oriented faults, particularly in the Brothers Fault zone (BFZ) (Fig. 2), was that
they were right-lateral strike-slip forming a transfer zone (Faulds and Varga, 1998)
between the northern edge of Basin and Range extension and unextended area to the
north (Lawrence, 1976). This interpretation was based on mapping lineations from
satellite images, not measured offsets or fault studies in the field. Later field
investigations of Pezzopane and Weldon (1993) found that faults throughout the region
were dominantly dip-slip, with negligible to minor oblique motion. Trench et al. (2012)
found abundant evidence for dip-slip motion on NW-oriented faults, including those
within the BFZ, and sparse evidence of strike-slip motion.
Scarberry et al. (2010) observed that NW-striking faults are more numerous
along northern tips of NNE-striking faults throughout the region, so concluded that the
NW-striking set in the Abert Rim area (Fig. 2) formed as result of splays along the
northward propagating tip of the Abert Rim fault exploiting a pre-existing NW-oriented
18
fabric. Trench et al. (2012) reached a similar conclusion by observing cross-cutting
relationships between the two fault sets in Oregon, where NNE-striking faults cut the
NW-striking set in the south, but the two sets are mutually cross-cutting in the north.
Trench et al. (2012) thus concluded that NW-striking faults first developed along the
fault tips of NNE-striking faults, but the NNE-striking faults ultimately grew through and
cross-cut the NW-striking faults; they also suggest that a lack of major NNE-striking
faults in or proximal to the Brothers Fault Zone indicates that the NW-striking faults
within that zone developed independently and were not influenced by NNE-striking
structures. Scarberry et al. (2010) suggested that in the Abert Rim area (Fig. 2), the
similarity between orientations of two distinct middle to late Miocene (16-5 Ma) fault
sets and those of early to middle Miocene (23.8-16.4 Ma) dikes that trend 5-10° and
290-315° indicate that the development of both fault sets was influenced by a pre-
existing fabric.
In northeastern California, two relatively isolated and poorly understood faults
strike parallel to the regional NW-striking fabric. The Fandango fault (Fig. 2) borders a
linear valley within the northern Warner Range in the footwall of the Surprise Valley
fault (SVF) and appears to be cut by the SVF (Egger and Miller, 2011). It may be an
accommodation zone between the east-dipping SVF and the west dipping faults along
the Goose Lake graben (Surpless and Egger, 2006), but the exact timing of motion and
amount of slip on the fault is poorly constrained. The Likely fault system (Fig. 2) has
been mapped as a strike-slip fault in the USGS Quaternary fault and fold database
19
(2006), but evidence for this fault is minimal, based on references to early mapping by
Gay and Aune (1958), and observations by Jenkins (1959) of sag ponds and apparent
dextral offset of some topographic features.
2.4 The Larkspur Hills
The Larkspur Hills study area, which lies east of Surprise Valley and north of the
Hays Canyon Range, straddles the border between California and Nevada (Figs. 2 and 3).
The area is characterized by low-relief hills formed by numerous normal faults, most of
which record minor down-to-the-east offset (Fig. 3). The faults cut a series of Late
Miocene to Pliocence volcanic and volcaniclastic rocks, primarily a group of distinctive, 3
-8 Ma low-potassium olivine tholeiites (LKOTs) (Carmichael et al., 2006). Basalt flows are
thin, with average thicknesses of ~1-5 m, and are interbedded with tuffaceous
sediments. The thickness of the entire unit is not well constrained as there are no other
units exposed within the study area. Where observed, the minimum thickness of the
interbedded basalts and tuffaceous sediments is 80 m, but it likely varies throughout the
study area (Egger and Miller, 2011). The LKOT basalts onlap the Oligocene unit along the
Hays Canyon range (Fig. 5) and cover the northern trace of the Hays Canyon fault,
indicating that this fault has been inactive since the late Miocene to Pliocene (Egger and
Miller, 2011).
The study area is characterized by predominantly west-facing dip-slopes of
basalt with east-facing normal faults; in the southern area, minor basins between these
hills are filled with Pleistocene lake sediments of pluvial Lake Surprise (Figs. 3 and 5)
20
Fig. 5. Close-up map of domain I. Legend for all domain close-up maps follows (5a). Scarp heights for each fault are noted. Solid white line is pluvial highstand level (Irwin and Zimbelman, 2012). Dotted white line outlines an E-W magnetic low (Egger et al., 2009).
21
Fig. 5a. Legend for all domain close-up maps.
(Egger and Miller, 2011). In contrast, the northern and eastern portions of the field area
are characterized by primarily flat-lying basalts, fewer faults, and small playas in the
intervening basins (Fig. 3), consistent with a lower magnitude of extension. Faults show
no signs of activity in the Holocene: there are no Holocene fault scarps, and Pleistocene
shorelines below the high-stand of pluvial Lake Surprise (Fig. 5) do not appear to be
deformed.
Over this relatively small area of about 550 km2, however, there are significant
differences in fault orientations, from dominantly N-striking in the south to NNE- and
NNW-striking in the north (Fig. 3). How do these local fault orientations relate to the
5a
22
regional stress field over time, and what controls the differences in orientations from
south to north across the field area?
These questions have relevance to the characterization of nearby buried intra-
basin faults, which control locations of hot springs throughout the valley (Fig. 2). Both
the Surprise Valley and the southern Larkspur Hills have been the object of geothermal
exploration (Barker et al., 2005; Benoit et al., 2005; Bertani, 2007) and studies to define
structural features in the subsurface (Egger et al., 2010; Glen et al., 2013). Local studies
(Benoit et al., 2005; Egger et al., 2014) point to the importance of fault controls on
subsurface fluid circulation and the surface location of hot springs in the Surprise Valley
(Glen et al., 2013); this is in agreement with other studies that emphasize the
importance of fault controls in geothermal systems (Curewitz and Karson, 1997; Faulds
et al., 2010). While hot springs occur on both sides of the basin in the Surprise Valley,
along the east side they do not appear to be located along faults; however, faults are
known to be a primary control on fluid flow (Caine et al., 1996; Rowland and Sibson,
2004; Faulds et al., 2010). Faults expressed at the surface in the adjacent Larkspur Hills
may serve as an analog for faults buried in Surprise Valley, and may lend insight into
their geometry and controlling structures.
3. Methods
3.1. GIS-based methods
Surface structures in the Larkspur Hills were mapped in the ESRI GIS program
ArcMap 10.1, using hillshade maps generated from two digital elevation models (DEMs)
23
in combination with Digital Orthophoto Quadrangles (DOQs) at 1 m resolution. One
DEM covers the southern part of the field area, and was generated from high resolution
(1 pixel = 0.5 m) LiDAR point cloud data collected from aerial swath mapping conducted
in July 2012; data were processed at the National Center for Airborne Laser Mapping
(NCALM) at UC Berkeley. To extend the field area farther north, a DEM generated from
digitized topographic maps with 3 m spacing was obtained for the northern part of the
field area from the USGS National Elevation Dataset (Gesch et al., 2002; Gesch, 2007).
We initially identified fault traces by locating prominent cliffs on the hillshade model.
Where surface traces were not well defined, visual analysis was augmented by drawing
profiles across the faults. This involved interpolating a line perpendicular to strike, and
generating an elevation profile. This was repeated along strike to identify signs of fault
offset, and where the offset tapered off at the tip-lines. Google Earth imagery provided
3D perspective, and analysis of the imagery aided fault analysis and selection of field
sites.
The orientation of linking segments due to breaching of relay ramps (Fig. 4a) can
be influenced by pre-existing structures (Trudgill and Cartwright, 1994). Also, faults can
develop as linking structures when there is a change in extension direction (Henza et al.,
2011) or when a pre-existing fault set at an angle to the extension direction is
reactivated (Bellahsen and Daniel, 2005). In order to test for common orientations in
fault segments that may reflect the influence of either pre-existing structures or a
change in extension direction, we determined the azimuth of individual segments by
24
manually drawing straight best-fit lines along mapped faults at the 1:24,000 scale. The
term “segment” is commonly used to refer to fault segmentation at a wide range of
scales and based on varying criteria from one study to the next (DePolo et al., 1991); for
the purpose of this analysis, the term “segment” refers to these best-fit line segments,
which varied in azimuth from adjacent segments by an average of 25° with a standard
deviation of 15°. A plugin for ArcGIS, EasyCalculate 10.0 by ET Spatial Techniques, was
used to calculate the azimuth for each segment, which was then exported to an Excel
spreadsheet. Segments were normalized to total fault length, which was calculated by
summing the lengths of the segment lines. Since dip direction is dominantly to the east
and this analysis was not affected by dip direction, all azimuths were plotted in the
upper compass quadrants, exported to Stereonet (Allmendinger et al., 2012), and
plotted on rose diagrams in bins of 10°.
Segments were then analyzed according to percent of fault length. Hus et al.
(2005) found that faults interacted and formed linking structures when the overlap was
~1/8 of the combined fault lengths, which is ~12.5% of fault length. Linking structures
that form as a result of change in extension direction (Henza et al., 2011) or the
reactivation of pre-existing structures (Bellahsen and Daniel, 2005) do not have a
predictable ratio to overall fault length, so grouping was also based on the average
number of segments per fault. For the entire study area, the average was 5 segments
per fault, so segments that made up a quarter, or more, of the total fault length were
grouped on one diagram, as those segments tended to dominate overall fault
25
orientation and most were associated with shorter faults. Segments less than 25% of
fault length were grouped on another, as they tended to be associated with very
segmented faults, so were of relatively equal dominance in controlling the fault
orientation. The faults lying along domain boundaries were plotted separately, but with
all segments on one diagram for each boundary, since these faults were the most
segmented. Mean segment orientation for each group, along with the corresponding σ3,
was also plotted. The likely paleostress orientation, σ3, for the study area was computed
at 90° from the mean segment orientation for all segments within the study area.
Extension was calculated along topographic profiles from section lines (Fig. 3)
generated in ArcGIS. The tilt of fault blocks was measured directly from the topographic
profile where slopes represented dip-slopes of basalt, which are well preserved
throughout the study area. The dips of faults, however, could not be measured due to
scarp retreat, which we assumed to be significant due to the interbedded tuffs,
prominent vertical fractures in the basalt, and large talus slopes at the base of cliffs. For
extension calculations, dip was assumed based on the ~68° dip on the Surprise Valley
fault as exposed in a trench (Personius et al., 2009) and the 60°-90° range of fault dips
throughout the region as summarized by Scarberry et al. (2010). Extension was
calculated using a minimum dip of 65° and a maximum of 80° in order to remain within
the range for the region, and allow for uncertainty in the assumption. Fault dips on the
cross-sections were corrected for apparent dip and amount of block tilt measured from
the topographic profile. Extension was calculated by measuring the horizontal
26
displacement for each fault, and summing the displacements across the section line.
Percent extension was calculated across a distance of 10 km for each line for consistent
comparisons across the study area. The values reported are minima due to the lack of
correlation between capping flows of individual faults, and later flows may have filled in
topographic lows (Ritzinger et al., 2013). Block tilt was reported after correcting for
apparent tilt.
Scarp height was measured at the location of maximum elevation change along
each segment, identified by using an overlay of USGS published topographic maps with
contour lines at 20 ft intervals. Scarp height was correlated to fault segment data. For
each fault, the scarp height of each segment was plotted at the midpoint of the segment
(so the spacing of the points would reflect the segment spacing along strike) using a
circle that represents the segment azimuth, with larger empty circles indicating more
westward trends and larger grey circles indicating more eastward trends. Graphing
using this procedure allowed for comparisons of several parameters: maximum scarp
height, segment length, segment azimuth, number and spacing of segments per fault,
and variations in azimuth from segment to segment. Each fault was graphed for analysis.
3.2. Field methods
Field analyses focused on areas where the resolution of the aerial imagery and
DEMs was insufficient for distinguishing minor offset faults and possible volcanic
features, and for examining cross-cutting and truncating relationships of faults,
especially where NNW striking faults or fractures were identified in the footwalls or tip-
27
lines of N-NNE faults. Based on these targeted analyses, locations of fault traces,
linkages, and linear volcanic features were revised or mapped.
4. Results
4.1 Larkspur Hills domains
The Larkspur Hills study area was divided into domains based on fault density
and orientation (Fig. 3), and thus the results of these fault analyses are reported initially
within domains, and summarized in Table 1. Larger scale maps of domains I-V have been
included; Fig. 6 compares segment orientations across all domains; and Fig. 7 compares
fault orientations, segmentation, and offsets across all domains. Domain boundaries
occur primarily along major-offset NNE- and N-striking faults or basins (Fig. 3), and data
from domain-boundary faults are treated separately.
4.1.1 Domain I
The primary characteristic of domain I (Fig. 5) is closely-spaced, N-S-striking fault blocks.
Domain I exposes 20 faults (Table 1) with overall northward trends, dip directions to the
east, and lengths ranging from 0.7-11.4 km. Average fault spacing is 0.8 km; spacing is
more regular in the western half of the domain than in the east. Fault blocks are tilted 5-
13° to the west, with the degree of tilt decreasing across the domain towards the east.
One fault has a scarp height of 224 m, the rest range from 18-128 m, with an average of
64 m.
Faults are segmented, with an average of 6 segments per fault and mean
segment length of 0.5 km. Segments that make up ≥25% of fault length trend
Table 1. Summary of fault domain characteristics. Tilt measured from profile line and corrected for apparent tilt.
Domain Number
of Faults
Fault
length
(km)
Mean fault
length (km)
Spacing
(km)
Mean
spacing
(km)
Tilt of
fault blocks
(westward)
Scarp
height
(m)
Mean scarp
height (m)
Segments
per fault
Number of
Segments
Segment
length (km)
Mean
segment
length (km)
I 20 0.7-11.4
2.1 0.1-1.7
0.8 5-13° 18-224
64 7 132 0.08-1.9 0.5
II 24 1.0-8.4 2.4 0.2-3.6
1.3 0-6.5° 6-156 59 6 136 0.08-2.1 0.6
III 27 1.6-9.3 1.9 0.3-5.8
1.7 3-8° 5-186
42 6 80 0.08-3.3 1.7
IV 9 0.8-4.1 2.0 0.3-5.0
2.7 0-4.5° 17-183
48 2 20
0.4-2.4 1.8
V 6 1.4-4.5 4.1 1.2-6.2
3.7 0-2° 37-177
63 3 28 0.2-3.9 3.2
29
310°- 060°, with an mean trend of 359° and standard deviation of 23°; the segments
form two groups with mean trends of 345° and 015° (Fig. 6). Segments <25% of fault
length trend 300°- 070°, with a mean of 005° and standard deviation of 24°. Adjacent
segments of the same fault vary in orientation by as much as 40° (Fig. 7); faults have
arcuate or zig-zag traces. Fault traces in domain I display hook-shaped linking structures
(Figs. 4b and 5); this style of linkage is not observed in the fault traces in domains II-V.
4.1.2. Domain II
The primary characteristic of domain II (Fig. 8) is closely spaced, NNW-striking
faults and fault blocks. There are 24 faults (Table 1) with overall NNW trends and NE-
facing scarps. The average fault spacing is 1.3 km. Spacing is more regular in the
southern portion of the domain; the northern portion has closely spaced faults along
the basin margin, and a wide (4.6 km) un-faulted section in the NE. Fault blocks are
tilted 0-7° to the west, with the degree of tilt decreasing across the domain towards the
north. Fault lengths range from 1.0-8.4 km. One fault has a scarp height of 151 m, the
rest range from 6-98 m, with an average of 59 m.
Faults are segmented, with an average of six segments per fault and mean
segment length of 0.6 km. Segments ≥25% of fault length have trends ranging from
300°- 040°, with an mean of 347° and standard deviation of 26° (Fig. 6). Segments <25%
of fault length have trends ranging from 300°- 030°, with a mean trend of 344° and
standard deviation of 24°. Most adjacent segments of the same fault vary in orientation
by about 10°-20° (Fig. 7); the fault traces are more linear than in domain I.
30
Fig. 6. Rose diagrams showing fault segment orientations for each domain. Segments are normalized to fault length and plotted in 10° bins. For each plot, n=number of segments displayed. Length of petals represents percent of n segments. Segments >25% of fault length are likely dominant structures, while segments <25 % of fault length are likely linking structures or of relatively equal influence on overall fault orientation. Heavy grey line shows mean segment orientation; dashed grey line shows least compressive stress perpendicular to mean segment orientation.
31
Figure 7. Graphs of one prominent fault per domain and domain boundary. Circles indicate maximum scarp height and azimuth for each segment, plotted at the midpoint of each segment along strike. Grey circles represent orientations 001°-090°, and open circles 270°-359°. The size of the circle indicates the magnitude of deviation from north, with larger grey circles for more eastern azimuths, and larger open circles for greater western azimuths, as indicated by key (above).
32
Figure 8. Close-up map of domain II. See figure 5a for legend. Scarp heights for each fault are noted. Solid white line is pluvial highstand level (Irwin and Zimbelman, 2012).
8
33
Domain II has five linear vents with trends that are subparallel the normal faults.
Three are proximal and appear to be cut by a fault along the same trend.
4.1.3. Domain III
Domain III (Fig. 9) is characterized by fewer faults than domains I and II, and
several fissure vents that parallel faults. There are nine faults (Table 1), with overall
trends to the N and the NNW; unlike domains I and II, fault dip directions vary. While
the dominant dip direction is eastward, two faults dip to the W. This domain also has 12
linear volcanic vents that parallel fault trends. Fault lengths range from 1.6-9.3 km, with
an average length of 1.9 km. Scarp heights range from 5-108 m, with an average height
of 42 m. Fault spacing ranges from 0.3-5.8 km, with an average spacing of 1.7 km. Fault
spacing is more variable than in domains I and II, and while most faults have similar
spacing as other domains, there are also wide un-faulted areas between the domain
boundaries (up to 5.8 km wide). Fault blocks are tilted 3-8° to the west, with the degree
of tilt decreasing across the domain towards the east.
Faults average six segments per fault, with a mean segment length of 0.6 km.
Segments ≥25% of fault length trend 300°-030°, with a mean trend of 347° and standard
deviation of 25°; there are two modes at 330° and 20° (Fig. 6). Segments <25% of fault
length trend 270°-070°, with a mean of 003° and standard deviation of 26°; this plot also
has two modes but at slightly different orientations at 340° and 000° (Fig. 6). In contrast
to the other domains, segment orientations in domain III vary more along longer faults
34
Figure 9. Close-up map of domains III and IV. See figure 5a for legend. Scarp heights for each fault are noted. Boundary faults are labeled.
35
(Fig. 7) than along the shorter faults (Fig. 9); long faults are more arcuate and shorter
faults are relatively linear.
Numerous fissure vents parallel the faults throughout the domain, and some
relief is caused by the exposed margins of basalt flows rather than minor offset faults or
vents. In the southern part of domain III, a vent with a NNE stike parallels the major-
offset NNE-striking fault along the domain boundary (Fig. 9). Other vents in the southern
part of the domain parallel a N-striking fault that may cross-cut the boundary fault. The
combination of little block tilting, the significant playa sediments along Crooks Lake, and
relief due to primary volcanic features complicated the offset calculations for faults in
this domain, as well as adding uncertainty in measurements of scarp heights.
4.1.4. Domain IV
Domain IV (Fig. 9) is characterized by a relatively few faults. There are eight
faults (Table 1) that trend predominantly NNW. Five of the faults are in the footwall of a
major-offset fault along the eastern boundary. Fault lengths range from 0.8-4.1 km.
Scarp heights range from 17 to ~48 m; though determining maximum topographic relief
on four of the faults is complicated by cross-cutting relationships (Fig. 9). Seven of the
faults dip E to NNE, and one dips S to SSE. Westward slope of 1°-5° along the eastern
margin is likely due to footwall flexure of the major offset fault along Long Valley rather
than to overall tilting of fault blocks. Fault spacing is irregular, with sets of 2-3 closely
spaced (0.3-0.9 km) faults separated by wide (2-5 km) un-faulted areas.
36
Faults in this domain are the least segmented, with an average of two segments
per fault and average segment length of 0.9 km. Segments ≥25% of fault length trend
310°-070°, with an average trend of 340° and standard deviation of 29°, and segments
<25% of fault length trend 310°-350° with an average of 324° and standard deviation of
27° (Fig. 6). Segments of the same fault have little variation in orientation (Fig. 7),
though there is variation between different faults; faults have relatively linear traces.
4.1.5. Domain V
Domain V (Fig. 10) is characterized by a narrow zone of faults and wide un-
faulted areas. There are six faults (Table 1) that trend predominantly north, and dip east
and west. Fault lengths range from 1.0-3.4 km. Scarp heights range from 37-115 m, with
an average of 63 m. Fault spacing is irregular, with 4 of the 6 faults spaced <2 km apart
along one narrow zone, and most of the domain consists of un-faulted areas over 6 km
wide. The footwall of one major offset fault slopes ~2° to the east, likely a reflection of
footwall flexure rather than tilting of the fault block.
Faults are less segmented than domains I-III, with an average of only three
segments per fault and average segment length of 1.2 km. Segments ≥25% of fault
length trend 300°-020°, with an average trend of 345° and standard deviation of 31°;
segments <25% of fault length have trends from 280°-060°, with an average trend of
335° and standard deviation of 41° (Fig. 6). Fault segments have little variation in
azimuth along strike (Fig. 7), so have relatively linear traces.
37
There is little relief in domain V. While examination of the DEM and Google
earth imagery indicated the possibility of minor offset faults, field observations
confirmed much of the minor topography is created by margins of individual basalt
flows, which in some cases appear steep due to drainages that follow this topography.
There are six linear vents that trend consistent with the faults, and five appear to have
been subsequently offset by faults (Fig. 10).
Fig. 10. Close-up map of domain V. See Fig. 5a for legend. Scarp heights for each fault are noted. Inset map shows close-up of structurally controlled linear vent, with the fault trace dashed where it offsets the vent; see text for discussion.
38
4.2. Major boundary faults
Differences between the fault characteristics among the domains aided
determination of the boundary locations, and faults lying along boundaries were
analyzed separately from the domain faults using the same methods. The eastern
boundary of domain II (Fig. 9) consists of three faults with overall NNE trends and
eastward dips. Westward tilting of fault blocks is 12° on the southernmost fault,
decreasing to 4° on the next fault, and no tilt was noted on the northern-most fault.
The faults are highly segmented, with an average of 12 segments per fault. Fault lengths
range from 10-11.4 km, and segment lengths average 0.6 km. Two of the faults have
maximum scarp heights of 185 m and 100 m. The third fault may be linked to the fault
along Crooks Meadow (Fig. 9), and this fault likely cuts an older NW trend fault in the
northern part of the domain. The scarp height is ~43 m. Segments trend 270°-080°, with
a mean of 022° and standard deviation of 28° (Fig. 11); the faults have arcuate to zig-zag
traces, with the greatest variation of segment orientations along strike (Fig. 7). The NNE-
striking fault, D2B-1, that makes up the southern part of the boundary (Fig. 9) has minor
offsets in the footwall which trend NNW, especially along northern segment of the fault.
Several of these were also observed in the field, but were below the resolution of the
DEM and the scale of the mapping; they are noted here due to their consistent trend
with other footwall faults. The offset on most of these footwall faults tapers off quickly
from the main fault, but a few appear to be continuous with NNW-striking domain II
faults. These NNW-striking footwall faults are not continuous across the fault in the
39
hanging wall, thus supporting the designation of a boundary along the major NNE-
striking fault.
Similarly, the eastern boundary of domain III (Fig. 9) consists of five faults with
overall NNE trends and eastward dips. Westward tilting of fault blocks is 8° on the
southern-most fault, and there is no tilt noted on the rest of the boundary faults. The
faults are less segmented than the eastern domain II boundary, with an average of five
Fig. 11. Rose diagrams showing fault segment orientations for domain boundary faults. Segments are normalized to fault length and plotted in 10° bins. For each plot, n=number of segments displayed. Length of petals represents percent of n segments. Each plot shows all the segments for the boundary. Solid grey line shows mean segment orientation; dashed grey line shows least compressive stress perpendicular to mean segment orientation.
segments per fault. Fault lengths range from 0.7-8.9 km, and segment lengths average
0.8 km. Scarp heights range from 13-145 m (Fig. 9). Segments trend 340°-070° (Fig. 11),
with a mean of 023° and a standard deviation of 23°; the fault traces are arcuate, but
segment orientations vary less along strike than along the eastern domain II boundary
(Fig. 7).
40
Fault D3B-1, a NNE-striking fault along the southern part of the boundary (Fig. 9)
has prominent fractures which appear in the field as broad “notches” along the top of
the scarp that extend from the scarp edge and taper off into rubble-filled prominent
fractures that trend roughly NNW. These fractures were below the resolution of the
DEM and the scale of the mapping, but are noted here due to trends consistent with the
faults. Along the northern portion of the fault, there are a few minor NNW-striking
footwall faults. Overall, the fault is shorter with less offset than the major offset NNE
fault in the domain II eastern boundary, and it appears to be cut by a N-striking fault at
its southern tip, though this is difficult to constrain due to a major drainage, volcanic
vent, and landslides that complicate identification of the D3B-1 fault trace.
4.3 Extension direction and magnitude
Averaging the mean least compressive stress orientation (3) for the domains
(Fig. 6) resulted in an average orientation of 263°. While the average σ3 for the
boundary faults is oriented 292°, when the boundary faults are grouped with the
domain faults, the resulting orientation is 267° ±3°. This is consistent with ~E-W
principal stress direction for the northwest Basin and Range region determined by
Crider (2001), and 267° for south eastern Oregon determined by Pezzopane and Weldon
(1993). This ~E-W least compressive stress orientation is assumed to be consistent with
the regional extension direction, based on Anderson’s classification of tectonic stress
(Anderson, 1951) and the seismic and geologic evidence for regional extension in an E-
W to ENE-WSW direction (Crider, 2001). Section lines for extension calculations were
41
drawn due E-W to remain within the range of uncertainty and also minimize the
differences between apparent and actual dip on the majority of faults along the section
lines.
Extension was calculated along five EW-striking cross-section lines to compare
the magnitude and percent of extension from south to north across the study area (Figs.
3 and 12). Cross-sections are included to show dominant style of faulting; however,
were not restored due to lack of correlation of units. The percent of extension was
calculated over 10 km for each section line for purposes of comparison, though the
faults along section lines A and B are concentrated along a shorter distance. Since the
magnitude of extension was calculated assuming a minimum dip of 65° and maximum of
80°, the following results are reported as a range. The magnitude of extension across
section line A is 222-424 m across a narrow zone of 6 down-to-the-east bookshelf style
faults, resulting in 2.4-4.6% extension. Extension may be slightly greater along section
line B, with 3.5-6.8% extension calculated across the same zone of faults. Extension of
237-569 m (2.4-6%) across section line C is accommodated primarily along four east-
dipping faults, with one additional minor offset fault. Extension decreases across line D,
with 151-327 m (1.5-3.4%) distributed across 9 faults, including one west dip fault. All
but two of the faults have only minor offset. Only 70-166 m (0.4-0.9%) of extension was
accommodated across line E, mainly on one major offset west-dipping fault. All of these
values for extension should be considered minima, as later basalt flows may have filled
in basins, decreasing the apparent offset along individual faults.
42
Fig. 12. Cross-sections A-A’ through E-E’ corresponding to section lines shown on Larkspur Hills map (Fig. 3). Cross-sections show dominant down-to-the-east bookshelf faults with assumed original dips of 65°. Lines are organized moving northwards from line A to line E.
43
4.3 Timing of faulting and extension rates
Calculating extension rates is complicated by lack of correlation between flows
(Ritzinger et al., 2013), the range of possible horizontal offset values due to
uncertainties in fault dip, and a lack of constraint on the period of fault activity. All
faults are likely younger than 3 Ma, but we cannot definitively rule out the possibility
that some faults were active during the period of magmatism from 3-8 Ma. Faults in the
Larkspur Hills show no signs of activity in the Holocene: there are no Holocene fault
scarps, and the Pleistocene shorelines do not appear to be disturbed. Assuming the
faults were active from the start of the period of extension beginning ~3 Ma (Colgan et
al., 2008) until just before the Pleistocene shorelines were developed ~ 0.02 Ma (Ibarra
et al., 2014), the extension rates (in mm-yr-1) across each line are: 0.14-0.07 for section
line A, 0.21-0.11 for line B, 0.19-0.08 along line C, 0.11-0.05 along line D, and 0.06-0.02
across line E. This results in an average of 0.1 mm-yr-1 across the field area.
5. Discussion
5.1 Controls on fault geometry
The analysis described above suggests that the main factors that influence the
geometry of normal faults in the Larkspur Hills are extension direction and pre-existing
structures. The relative importance of each of these is discussed below, followed by a
brief discussion of the two other controls that were considered, a change in stress
orientation and effect of differences in lithology.
44
5.1.1 Extension direction
Normal faults in a homogeneous medium form perpendicular to the least
compressive stress, 3, according to assumptions outlined in Anderson (1951) (Fig. 13a).
Non-Andersonian mechanics have been theorized and tested both in the laboratory
(Reches and Dieterich, 1983) and in the field (Aydin and Reches, 1982; Krantz, 1988;
Krantz, 1989), and result in separate coeval fault sets (Fig. 13b).
Fig. 13. Comparison of Andersonian and orthorhombic fault geometry. The symbol σ represents
stress, and ε represents the associated strain, with 1 indicating maximum stress/strain, 2
intermediate, and 3 minimum. Fig. 13a shows conjugate faults that form by plane strain as predicted by Anderson (1951). Fig. 13b shows two conjugate pairs that form by non-planar strain as illustrated by Reches (1983). Arrows indicate slip directions. Diagram adapted from Reches (1983).
45
Assuming Anderson mechanics (Anderson, 1951), where the maximum
compressive stress in an extensional regime is vertical, and the minimum and
intermediate stresses are both horizontal and unequal (Fig. 13a), the faults in the
Larkspur Hills are all favorably oriented to slip with a minimum stress oriented
approximately E-W, or slightly to the N or S of this orientation, assuming some oblique
component of motion on the NNW- and NE-striking faults. While it is not possible to
measure stress directly at the time when the faults formed, we infer an E-W paleostress
orientation based on:
1. The dominantly N-S orientation of faults in domain I (Figs. 3 and 5),
which are consistent with orientations of the proximal major structures,
the Hays Canyon fault and Surprise Valley fault (Fig. 2). Additionally, the
domain I faults have hook-shaped linking structures (Fig. 4b). This form
of linking structure results from interaction of tensions at fault tips
during fault growth, and is not observed in analogue models where pre-
existing structures ruptured through subsequent cover (e.g. Bellahsen
and Daniel, 2004). The seismic study of Giba et al. (2012) indicated that a
large structure at depth can rupture through subsequent cover, resulting
in en echelon segments at the surface, which are characterized by relay
structures in the zones of overlap (Figs. 4a and 4c); however, hook-
shaped structures were not evident (Giba et al., 2012). This suggests that
the faults traces in domain I resulted from individual N-S segments that
46
grew along tip-lines and connected, rather than resulting from
reactivation of pre-existing structures at depth.
2. The average σ3 direction of 267° determined by a least compressive
stress perpendicular to average segment orientations (Fig. 6 and 7). As
indicated in Wang (2000), the movement of several faults can
approximate the regional stress, so while there are variations in the
inferred σ3 among the different fault domains and domain boundaries
due to the variations in fault orientations (Fig. 3, 6, and 7), it is
reasonable to infer that they work together to accommodate an overall
motion consistent with a single regional stress direction; however, in the
absence of sense-of-slip indicators for the faults in the study area, this
cannot be determined conclusively. Strain partitioning along faults with
varying orientations and pure dip-slip motion would also be more
consistent with non-planar strain (Fig. 13b) than plane strain (Fig. 13a).
3. The N-S trend of linear vents. Those that have NNW-SSE and NNE-SSW
orientations are at tip-lines of, or have been subsequently offset by,
faults along the same trend, so likely reflect the influence of pre-existing
structures (discussed in section 5.1.2), while many of the N-S oriented
vents are not proximal to faults so are interpreted as responding to
extension direction. Since dikes intrude along planes perpendicular to
the least principal stress (Anderson, 1951), dikes and linear volcanic
47
features (e.g. aligned cinder cones, elongated volcanic edifices) can be
used as good geologic indicators of the regional orientation of the least
principal stress (Nakamura, 1977; Zoback et al., 1994).
Our conclusion that extension during fault activity was oriented E-W is supported
by previous work. Crider (2001) used a variety of methods and regional data to
determine a minimum principal stress direction for the northwest Basin and Range
ranging from 263°-269°. She concluded that an extension direction due E-W (270°) or
slightly north of east (264°) would result in sinistral slip on NNE-striking faults, and
dextral slip on NW-striking faults, which is consistent with field evidence for minor to
negligible oblique motion on some faults in the region (Pezzopane and Weldon, 1993;
Trench et al., 2012); evidence of minor oblique motion was largely absent along the
surface traces throughout the northwest Basin and Range, but detected in trenches
(Pezzopane and Weldon, 1993). Other faults, such as the Surprise Valley fault, show no
evidence of lateral motion (Personius et al., 2009).
5.1.2 Pre-existing structures
Lacking seismic data at depth in the Larkspur Hills, comparison with sandbox
models provides a means to assess the role of pre-existing structures. In particular,
sandbox models that tested extension direction with varying orientations of pre-existing
weaknesses provide insight. The surface traces of the NNW-striking faults in domain II
compare favorably to those formed in sandbox models of Bellahsen and Daniel (2005)
(Fig. 14a and 14b). In the model, the size, dip, spacing, and orientation of pre-existing
48
faults was controlled by inserting a card through a sand layer. Pre-existing faults
oriented 70° to the extension direction were reactivated and continued as the dominant
fault set due to the high angle of the pre-existing structures to extension direction.
Assuming the same plane strain conditions of the model, the pre-existing NNW-striking
faults in the study area could therefore have evolved under E-W extension; however, a
component of oblique motion would be required. While sense of slip indicators are not
observed in the Larkspur Hills, minor oblique motion on other faults in the region has
been observed (Pezzopane and Weldon, 1993; Trench et al., 2012), even where
evidence of minor oblique motion was largely absent along the surface traces it was
detected in trenches (Pezzopane and Weldon, 1993).
Fig. 14. Simplified fault maps comparing results of sandbox models of Bellahsen and Daniel (2005) (a and c) and fault traces in the study area (b). Diagram 14a shows model with pre-existing faults at a high angle to extension, and diagram 14c shows pre-existing structures at a low angle to extension. Grey line: fault trace. Dashed line: pre-existing structure - high angle. Black line: pre-existing structure - low angle. Pre-existing structures for diagram 14b are inferred. See text for discussion.
49
The major offset NNE fault traces along the eastern boundaries of domains II and
III (Figs. 3 and 10) also have similarities to the sandbox models of Bellahsen and Daniel
(2005), in this case where pre-existing faults were oriented 45° to the extension
direction (Fig. 14c). In this model, due to the highly oblique angle of the structures to
extension direction, the faults were reactivated as linking structures between newly
formed N-striking segments. The surface traces in the field have long segments oriented
10-20° (Fig. 14b) rather than directly N as seen in the model, perhaps due to differences
in length, spacing, or lower angle of pre-existing structures. An alternative explanation is
that the NE-striking structures are more developed than the NNW-striking set. This may
also be why the NE-striking faults truncate the NNW-striking set, as well-developed pre-
existing faults can be barriers to fault rupture, which has been observed both in
analogue models (Henza et al., 2011) and in field studies (Machette et al., 1991; Milano
and Di Giovambattista, 2011).
NNW-striking faults and fractures are present in the footwalls of major offset N-
NNE-striking structures in domains II-IV, especially along the northern segments of the
longer faults. The NNW-striking structures may have reactivated initially as dilatational
fractures along fault tips as the larger-offset NNE-striking faults propagated through a
pre-existing fabric, as has been interpreted in the Abert Rim area (Fig. 2) (Scarberry et
al., 2010). This would account for the consistently NNW trend of the footwall faults and
fractures. The NNW-striking faults likely only dominate as an independent fault set in
domain II due to a lack of N-NNE structures, similar to faults in the Brothers Fault Zone
50
(Fig. 2) (Trench et al., 2012). Scarberry et al. (2010) noted that mid-Miocene dike
alignments in the Abert Rim area shared the same trends as the two fault groups, and
concluded that both the dikes and the late Miocene faults were exploiting a pre-existing
fabric. Similar evidence for exploitation of pre-existing fabrics may be seen in the linear
volcanic features in the Larkspur Hills as well.
The trend of the NNW and NNE linear vents in the study area is sub-parallel to
fault and fracture strikes suggesting that magma likely exploited both to reach the
surface (Silver et al., 2004; Valentine and Krogh, 2006; Gaffney et al., 2007; Thomson
and Schofield, 2008). In domains II and V, linear vents are proximal to or cut by faults
along the same orientation (Figs. 9 and 11), suggesting that they exploited the same pre-
existing structures as the faults, while in domains I and III, most of the NS-striking linear
vents are not faulted, but parallel the faults (Figs. 5 and 10), suggesting they responded
to extension direction (as discussed in 5.1.1 above) rather than pre-existing structures.
5.1.2 Change in extension direction
A mid-Miocene rotation of extension direction from 245°-250° to the current
direction of 295° in the Basin and Range has been proposed (Zoback and Thompson,
1978; Zoback et al., 1994), with the rotation attributed to the influence of right-lateral
shear resulting from the evolution of the San Andreas fault system. Colgan (2013)
argues that long term Basin and Range extension has been oriented 290°-300°, with only
a temporary rotation to the northeast in response to the impingement of the
Yellowstone Hotspot ~16 Ma, with the overall long-term spreading occurring as a result
51
of gravitational collapse perpendicular to the axis of thickest crust resulting from the
Sevier fold and thrust belt (Figure 1 of Colgan, 2013). The timing of the events proposed
by these studies pre-dates the recent period of faulting in the Larkspur Hills, which are
located in an area that has not experienced a change in its tectonic setting since the
mid-Miocene (Egger and Miller, 2011). Differences in fault orientations in the study area
as a result of rotation of the stress field during faulting should be consistent throughout
the field area, but this is not observed; while, the possibility of a local rotation cannot be
completely ruled out, it is unlikely and the cause of rotation would be only speculation.
It is also worth noting that the orientation of the axis of thickest crust, as illustrated and
discussed in Colgan (2013), is consistent with ~E-W extension at the latitude of the
northwest Basin and Range.
5.1.2 Lithological differences
While there are studies examining the effect of differences in lithology on fault
initiation and growth in normal fault systems, both along the down-dip extent (e.g.
Shopfer et al., 2006) and on surface traces (Ackermannn et al., 2001), any relationship
between the fault geometries in the Larkspur Hills and the nature or the thicknesses of
the underlying layers would be merely speculation, due to lack of constraint on
thicknesses and composition of the units in the study area, which likely vary throughout
the field area. The exposed basalt flows vary in thickness, but most have an average
thickness of ~1-5 m, and the flows are interbedded with tuffs and tuffaceous sediments.
There are no other units exposed within the study area, and the thickness of the entire
52
unit is only measureable along part of the eastern boundary (Fig. 9), where the major
normal fault along Long Valley exposes LKOTs directly overlying an Oligocene volcanic
unit (Egger and Miller, 2011). The approximate thickness of the LKOT unit at that
location is 80 m. The LKOT basalts onlap the Oligocene unit along the Hays Canyon
range to the south (Fig. 3). Logs from an exploratory drill hole located in the basin
adjacent to the southern portion of domain I indicates the presence of dense altered
shaley-sand at a depth of 1784 m (depth of hole), while drill-hole logs on the other side
of the basin at approximately the same latitude indicates lake sediments to a depth of at
least 1998 m (depth of hole) (Fig. 15). The presence of pillow basalts along the western
margin of the study area supports the likelihood of lake sediments directly underlying
the basalt unit along the valley margin, but again, the thickness of the basalt unit is
unconstrained along this margin, as is the lateral extent of underlying sedimentary
layers. Due to the lack of constraint on underlying units, the control they may exert on
the surface traces is undetermined.
5.2 Evidence for a structural or lithological boundary separating fault domains
Although pre-existing structures appear to control fault development in much of
the field area, namely the NNW- and NNE-striking faults in domains II-V, they do not
appear to occur everywhere in the subsurface, especially in domain I where faults
strike primarily N-S, and are located north of the Hays Canyon fault splays inferred from
potential field data (Egger and Miller, 2011) (Figs. 3 and 5). Specifically, the change in
fault orientations from domain I to domain II is abrupt (Figs. 3, 7, and 8), occurring along
53
Fig. 15. Simplified map showing drill hole locations (grey circles) and hot springs (white circles) relative to faults. See Figs. 2 and 3 for detailed maps of faults.
an approximately E-W trend that coincides with the major NNE-striking boundary faults
DB2-1 and DB3-1 (Fig. 9).
The dominantly N-strike of domain I faults indicates E-W extension, which is the
same as along the Surprise Valley and Hays Canyon faults (Egger and Miller, 2011).
While NNW and NNE-striking segments are present in domain I, the NNW-striking
54
segments dominate in domain II and NNE-striking segments dominate the boundary
faults. These two groups become the primary fault orientations north of the field area,
in south-central Oregon. The N-striking faults, including the inferred termination of the
major Hays Canyon fault (Egger and Miller, 2011), die out as a dominant trend across
the two boundary structures. The abrupt change in orientations between domain I and
domains II-IV thus suggests a structural or lithological boundary at depth, indicating a
border between relatively isotropic crust in the south to heterogeneous crust in the
north.
While the primary line of evidence for a boundary is the abrupt change in fault
orientations, there are two other items worth noting that may indicate domain I has
different crustal characteristics than domains II-V. First, during a recent aeromagnetic
survey using low-flying UAV, an E-W oriented magnetic low was detected extending
from the basin into the southern portion of domain I (Glenn et al., 2013; Egger et al.,
2010). This magnetic anomaly lies along the same trend as a few minor E-W striking
footwall faults in the southern part of domain I (Fig. 5). These E-W faults are not present
elsewhere in the study area, but there are similar features in the footwall of the Surprise
Valley fault (Egger and Miller, 2011); there are no other prominent E-W striking
anomalies indicated by the magnetic data collected in the basin. The E-W low anomaly
separates N-S striking magnetic highs in the basin, which have been interpreted as dikes,
at least one of which may have exploited a pre-existing fault plane (Glenn et al., 2013).
The magnetic low may represent a structural feature that resulted from the formation
55
of the dikes, or which subsequently altered them after emplacement (Glenn et al.,
2013); alternatively, it may represent a structural feature that pre-dated, and therefore
influenced, the dike intrusions.
Another feature worth noting that is particular to the fault geometries in domain
I, but absent in other domains, is a pronounced but relatively short arc in an otherwise
fairly linear fault trace. While these may be due erosional effects related to the hook
style of linkage (Pollard and Aydin,1984; Acocella et al., 2000) (Fig. 4b), this pattern in
the fault traces correlates from one fault to another along ~E-W lines, one of which
corresponds to the magnetic low noted above. This pattern bears a marked similarity to
one of the analogue sandbox models of McClay et al. (2002) where abrupt, nearly 90°-
changes in the orientation of surface fault traces reflected the locations of lateral offsets
in the basement. Again, while the similarity between the fault traces in the field and
those in the model is speculative and may be the result of different processes, it is
nevertheless another possible indication that the crust underlying domain I differs from
that underlying domains II-V.
5.3 Implications for geothermal fluid flow
While the purpose of this study is to examine controls on fault geometries in the
Larkspur Hills, those faults are useful analogues for the geometries of buried intrabasin
faults, and so the results of this study has implications for projects to better understand
geothermal fluid pathways and barriers in this area (Barker et al., 2005; Benoit et al.,
2005; Glen et al., 2013; Egger et al., 2014). Faulds et al. (2010) noted that while
56
geothermal fields can be present along major range bounding faults, they are also found
along less prominent fault zones. This is the case in the Surprise Valley (Figs. 2 and 15),
where hot springs are located along both sides of the valley. Areas of fault intersection,
interaction, and termination create areas of concentrated stress resulting in highly
fractured, thus more permeable, regions (Curewitz and Karson, 1997; Faulds et al.,
2010). The size and shape of these areas are controlled by differences in fault geometry
and kinematics (Curewitz and Karson, 1997), so it is important to understand the
controls on the fault geometries in a particular area in order to predict the subsurface
geothermal potential related to hot springs at the surface. Geochemical differences
exist in the valley hot springs (Barker et al., 2005; Cantwell and Fowler, 2014) and this
may be explained by the structures controlling the distribution of those springs. Helium
isotope ratios measured in hot springs at the north end of the valley range from 2.28-
2.58 Ra (where Ra is 3He/4He ration in the atmosphere), differing from the 0.95-1.49 Ra
measured in hot springs further south, which indicates a trend of increasing mantle
influence to the north, possibly due to higher permeability in the north allowing for
transportation of heat and mantle volatiles from depth (Barker et al., 2005). Cantwell
and Fowler (2014) measured lower concentrations of B and Cl in the northern springs
than those to the south, and interpreted this as the result of colder meteoric water
mixing with hot thermal waters. They support this interpretation by silica polymorph
geothermometry, which also indicates a hot fluid at depth diluted by cold water.
Cantwell and Fowler (2014) speculated that the Surprise Valley fault was the main
57
conduit for fluid flow, but that the Fandango fault may be contributing to meteoric
water circulation. This seems unlikely, as the Fandango fault is south of the hot springs.
It may be that NNW-striking structures exposed in the hills to the east are present in the
basin as well, and that these may be influencing the hot springs to the north. Springs to
the south, along the intrabasin high (Fig. 15) and southward are likely influenced by N-S
striking faults, which is supported by potential field data (Egger et al., 2010; Glen et al.,
2013) indicating basin structures share the same trends as faults in domain I.
6. Conclusions
Based on detailed analyses of surface fault traces using a variety of remote sensing data,
coupled with field observations and comparisons to existing models, we conclude:
In the Larkspur Hills, preexisting structures in rocks underlying late Miocene-
Pliocene units exert a strong control on surface fault geometries.
NNW-striking structures are favorably oriented to slip under the current
stress field; NNW-striking faults therefore have longer segments and less
arcuate fault traces.
NE-striking structures are less favorably oriented to slip under the most
recent stress field and thus formed primarily as linking structures; these
NE-striking structures link newer segments that form faults with an
overall orientation that is optimal to slip.
Orientations of linear volcanic vents that range in age from 8-4 Ma (Carmichael
et al., 2006) are consistent both with the orientations of inferred pre-existing
58
structures and with E-W extension, suggesting that extension direction has
remained relatively constant since at least 8 Ma.
North of the field area, NNW- and NE-striking faults are dominant, while N-
striking faults are not present. This regional transition is captured within the field
area between domain I and domains II- IV, suggesting that the Larkspur Hills
overlies a significant structural or lithological boundary at depth.
59
Acknowledgements
Funding for this project was provided as part of NASA grant 10-UAS10-0021
awarded to: Jonathan Glenn, USGS; Anne Egger, CWU; and Corey Ippolito, NASA Ames.
We thank Dr. Heezin Lee of the National Center for Airborne Laser Mapping (NCALM) at
UC Berkeley for processing LiDAR as well as providing instruction on processing
methods. Brent Ritzinger’s help in the field with logistics and conversations about field
and data interpretations is much appreciated. Lisa Ely provided useful insights regarding
geomorphic features, especially drainages and landslides where they complicated
interpretations of fault terminations or interactions. D.S. would also like to thank to
Diane and Ken Austin for offering a sounding board for ideas and for input on initial
drafts of figures, and to Ken for discussions of geodetic data and geologic
interpretations.
60
References
Ackermannn, R. V., Schlische, R. W., & Withjack, M. O., 2001. The geometric and statistical evolution of normal fault systems: an experimental study of the effects of mechanical layer thickness on scaling laws. Journal of Structural Geology 23, 1803-1819. doi: 10.1016/s0191-8141(01)00028-1
Acocella, V., Gudmundsson, A., & Funiciello, R., 2000. Interaction and linkage of extension fractures and normal faults: examples from the rift zone of Iceland. Journal of Structural Geology 22, 1233-1246. doi: 10.1016/s0191-8141(00)00031-6
Acocella, V., Korme, T., & Salvini, F., 2003. Formation of normal faults along the axial zone of the Ethiopian Rift. Journal of Structural Geology 25, 503-513. doi: 10.1016/s0191-8141(02)00047-0
Acocella, V., Morvillo, P., & Funiciello, R., 2005. What controls relay ramps and transfer faults within rift zones? Insights from analogue models. Journal of Structural Geology 27, 397-408. doi: 10.1016/j.jsg.2004.11.006
Allmendinger, R. W., Cardozo, N., & Fischer, D., 2012. Structural geology algorithms: Vectors and tensors in structural geology. Cambridge University Press, Cambridge.
Anderson, E. M., 1951. The Dynamics of Faulting and Dyke Formation with Application to Britain, 2nd ed. Oliver & Boyd, Edinburgh.
Aydin, A., & Reches, Z., 1982. Number and orientation of fault sets in the field and in experiments. Geology 10, 107-112. doi: 10.1130/0091-7613(1982)10<107:NAOOFS>2.0.CO;2
Barker, B., Kennedy, M., Hoversten, M., van Soest, M. C., & Williams, K., 2005. Geothermal Exploration at Fort Bidwell, California. In: Proceedings of the Thirtieth Workshop on Geothermal Reservoir Engineering, Stanford, California.
Bellahsen, N., & Daniel, J. M., 2005. Fault reactivation control on normal fault growth: an experimental study. Journal of Structural Geology 27, 769-780. doi: 10.1016/j.jsg.2004.12.003
Benoit, D., Moore, J., Goranson, C., & Blackwell, D. D., 2005. Core hole drilling and testing at the Lake City, California, geothermal field. In: Geothermal Resources Council Transactions 29, 203-208.
61
Bertani, R., 2007. World Geothermal Generation in 2007. In: Proceedings of the European Geothermal Congress, Unterhaching, Germany.
Blenkinsop, T. G., 2008. Relationships between faults, extension fractures and veins, and stress. Journal of Structural Geology 30, 622-632. doi: 10.1016/j.jsg.2008.01.008
Buck, W. R., 2009. Dynamic Processes in Extensional and Compressional Settings: The Dynamics of Continental Breakup and Extension. In: A. B. Watts (Ed.), Crust and Lithosphere Dynamics: Treatise on Geophysics, 1st ed., vol. 6, Elsevier B.V., Amsterdam, 335-376.
Caine, J. S., Evans, J. P., & Forster, C. B., 1996. Fault zone architecture and permeability structure. Geology 24, 1025-1028.
Cantwell, C. A., & Fowler, A. P. G., 2014. Fluid Geochemistry fo the Surprise Valley Geotherml System. In: Proceedings of the Thirty-ninth Workshop on Geothermal Reservoir Engineering, Stanford, California, SGP-TR-202.
Carmichael, I. S. E., Lange, R. A., Hall, C. M., & Renne, P. R., 2006. Faulted and tilted Pliocene olivine-tholeiite lavas near Alturas, NE California, and their bearing on the uplift of the Warner Range. Geological Society of America Bulletin 118, 1196-1211. doi: 10.1130/b25918.1
Colgan, J. P., 2013. Reappraisal of the relationship between the northern Nevada rift and Miocene extension in the northern Basin and Range Province. Geology 41, 211-214. doi: 10.1130/g33512.1
Colgan, J. P., Shuster, D. L., & Reiners, P. W., 2008. Two-phase Neogene extension in the northwestern Basin and Range recorded in a single thermochronology sample. Geology 36, 631-634. doi: 10.1130/g24897a.1
Crider, J. G., 2001. Oblique slip and the geometry of normal-fault linkage: mechanics and a case study from the Basin and Range in Oregon. Journal of Structural Geology 23, 1997-2009. doi: 10.1016/s0191-8141(01)00047-5
Crider, J. G., & Peacock, D. C. P., 2004. Initiation of brittle faults in the upper crust: a review of field observations. Journal of Structural Geology 26, 691-707. doi: 10.1016/j.jsg.2003.07.007
Curewitz, D., & Karson, J. A., 1997. Structural settings of hydrothermal outflow: Fracture permeability maintained by fault propagation and interaction. Journal of
62
Volcanology and Geothermal Research 79, 149-168. doi: 10.1016/s0377-0273(97)00027-9
DePolo, C.M., Clark, D.G., Slemmons, D.B. and Ramelli, A.R., 1991. Historical surface faulting in the Basin and Range province, western North America: implications for fault segmentation. Journal of Structural Geology 13, 123-136.
Donath, F.A., 1962. Analysis of Basin-Range structure, south-central Oregon. Geological Society of America Bulletin 73, 1-16.
Egger, A. E., Glen, J. M. G., & McPhee, D. K., 2014. Structural controls on geothermal circulation in Surprise Valley, California: A re-evaluation of the Lake City fault zone. Geological Society of America 126, 523-531. doi: 10.1130/B30785.1
Egger, A. E., Glen, J. M. G., & Ponce, D. A., 2010. The northwestern margin of the Basin and Range province Part 2: Structural setting of a developing basin from seismic and potential field data. Tectonophysics 488, 150-161. doi: 10.1016/j.tecto.2009.05.029
Egger, A. E., & Miller, E. L., 2011. Evolution of the northwestern margin of the Basin and Range: The geology and extensional history of the Warner Range and environs, northeastern California. Geosphere 7, 756-773. doi: 10.1130/ges00620.1
Esri ArcGIS 10.1, 2012. ArcGIS for Desktop. Environmental Systems Research Institute: Redlands, CA. http://www.esri.com/software/arcgis/
Faulds, J.E. and Varga, R.J., 1998. The role of accommodation zones and transfer zones in the regional segmentation of extended terranes. In: J.E. Faulds and J.H. Stewart (Eds.), Accommodation Zones and Transfer Zones: The Regional Segmentation of the Basin and Range Province: Boulder, Colorado. Geological Society of America Special Paper 323, pp. 1-45.
Faulds, J., Coolbaugh, M., Bouchot, V., Moeck, I., & Oguz, K., 2010. Characterizing Structural Controls of Geothermal Reservoirs in the Great Basin, USA, and Western Turkey: Developing Successful Exploration Strategies in Extended Terranes. In: Proceedings of the World Geothermal Congress 2010, Bali, Indonesia, 1-11.
Ferrill, D. A., Stamatakos, J. R., & Sims, D., 1999. Normal fault corrugation: implications for growth and seismicity of active normal faults. Journal of Structural Geology 21, 1027-1038. doi: 10.1016/s0191-8141(99)00017-6
63
Gaffney, E. S., Damjanac, B., & Valentine, G. A., 2007. Localization of volcanic activity: 2. Effects of pre-existing structure. Earth and Planetary Science Letters 263, 323-338. doi: 10.1016/j.epsl.2007.09.002
Gawthorpe, R. L., Jackson, C. A. L., Young, M. J., Sharp, I. R., Moustafa, A. R., & Leppard, C. W., 2003. Normal fault growth, displacement localisation and the evolution of normal fault populations: the Hammam Faraun fault block, Suez rift, Egypt. Journal of Structural Geology 25, 883-895. doi: 10.1016/s0191-8141(02)00088-3
Gay, T. E., & Aune, Q. A., 1958. Geologic map of California, Alturas Sheet, California Dept. of Natural Resources, Div. Mines, scale 1:250,000.
Gesch, D., Oimoen, M., Greenlee, S., Nelson, C., Steuck, M., & Tyler, D., 2002. The National Elevation Dataset. Photogrammetric Engineering and Remote Sensing 68, 5-11.
Gesch, D. B., 2007. The National Elevation Dataset. In: D. Maune (Ed.), Digital Elevation Model Technologies and Applications: The DEM Users Manual, 2nd ed., American Society for Photogrammetry and Remote Sensing, Bethesda, 99-118.
Giba, M., Walsh, J. J., & Nicol, A., 2012. Segmentation and growth of an obliquely reactivated normal fault. Journal of Structural Geology 39, 253-267. doi: 10.1016/j.jsg.2012.01.004
Glen, J. M. G., Egger, A. E., Ipplito, C., & Athens, N., 2013. Correlation of geothermal springs with sub-surface fault terminations revealed by high-resolution, UAV-acquired magnetic data. In: Proceedings of the Thirty-Eighth Workshop on Geothermal Reservoir Engineering, Stanford, California.
Henza, A. A., Withjack, M. O., & Schlische, R. W., 2011. How do the properties of a pre-existing normal-fault population influence fault development during a subsequent phase of extension? Journal of Structural Geology 33, 1312-1324. doi: 10.1016/j.jsg.2011.06.010
Herman, G. C., 2009. Steeply-dipping extension fractures in the Newark basin, New Jersey. Journal of Structural Geology 31, 996-1011.
Holland, M., van Gent, H., Bazalgette, L., Yassir, N., Strating, E. H. H., & Urai, J. L., 2011. Evolution of dilatant fracture networks in a normal fault - Evidence from 4D model experiments. Earth and Planetary Science Letters 304, 399-406. doi: 10.1016/j.epsl.2011.02.017
64
Hus, R., Acocella, V., Funiciello, R., & De Batist, M., 2005. Sandbox models of relay ramp structure and evolution. Journal of Structural Geology 27, 459-473. doi: 10.1016/j.jsg.2004.09.004
Irwin, R. P., III, & Zimbelman, J. R., 2012. Morphometry of Great Basin pluvial shore landforms: Implications for paleolake basins on Mars. Journal of Geophysical Research-Planets 117, E07004. doi: 10.1029/2012je004046
Jenkins, H. O., 1959. Geology of northeastern California. Mineral Information Service 12, 1-8.
Kaven, J. O., & Martel, S. J., 2007. Growth of surface-breaching normal faults as a three-dimensional fracturing process. Journal of Structural Geology 29, 1463-1476. doi: 10.1016/j.jsg.2007.05.007
Krantz, R. W., 1988. Multiple fault sets and three-dimensional strain theory and application. Journal of Structural Geology 10, 225-237.
Krantz, R. W., 1989. Orthorhombic fault patterns: the odd axis model and slip vector orientations. Tectonics 8, 483-495.
Lawrence, R. D., 1976. Strike-slip faulting terminates the Basin and Range province in Oregon. Geological Society of America Bulletin 87, 846-850. doi: 10.1130/0016-7606(1976)87<846:SFTTBA>2.0.CO;2
Long, J. J., & Imber, J., 2012. Strain compatibility and fault linkage in relay zones on normal faults. Journal of Structural Geology 36, 16-26.
Machette, M. N., Personius, S. F., & Nelson, A. R., 1991. The Wasatch fault zone, Utah-segmentation and history of Holocene earthquakes. Journal of Structural Geology 13, 137-149.
Mansfield, C., & Cartwright, J., 2001. Fault growth by linkage: observations and implications from analogue models. Journal of Structural Geology 23, 745-763. doi: 10.1016/s0191-8141(00)00134-6
Martel, S. J., Pollard, D. D., & Segall, P., 1988. Development of simple strike-slip fault zones, Mount Abbot quadrangle, Sierra Nevada, California. Geological Society of America Bulletin 100, 1451-1465.
McClay, K. R., Dooley, T., Whitehouse, P., & Mills, M., 2002. 4-D evolution of rift systems: Insights from scaled physical models. AAPG Bulletin 86, 935-959.
65
McQuarrie, N., & Wernicke, B. P., 2005. An animated tectonic reconstruction of southwestern North America since 36 Ma. Geosphere 1, 147-172. doi: 10.1130/ges00016.1
Meyer, V., Nicol, A., Childs, C., Walsh, J. J., & Watterson, J., 2002. Progressive localisation of strain during the evolution of a normal fault population. Journal of Structural Geology 24, 1215-1231. doi: 10.1016/s0191-8141(01)00104-3
Milano, G., & Di Giovambattista, R., 2011. Seismicity at the border between Central and Southern Apennines (Italy): Re-evaluation of the early 1984 instrumental earthquake. Tectonophysics 499, 92-104. doi: 10.1016/j.tecto.2010.12.008
Modisi, M. P., Atekwana, E. A., Kampunzu, A. B., & Ngwisanyi, T. H., 2000. Rift kinematics during the incipient stages of continental extension: Evidence from the nascent Okavango rift basin, northwest Botswana. Geology 28, 939-942. doi: 10.1130/0091-7613(2000)28<939:rkdtis>2.0.co;2
Moriya, S., Childs, C., Manzocchi, T., & Walsh, J. J., 2005. Analysis of the relationships between strain, polarity and population slope for normal fault systems. Journal of Structural Geology 27, 1113-1127. doi: 10.1016/j.jsg.2005.01.017
Morley, C. K., 1995. Developments in the structural geology of rifts over the last decade and their impact on hydrocarbon exploration. In J. J. Lambiase (Ed.), Hydrocarbon Habitat in Rift Basins, Geological Society Special Publication No. 80 (pp. 1-32).
Morley, C. K., Haranya, C., Phoosongsee, W., Pongwapee, S., Kornsawan, A., & Wonganan, N., 2004. Activation of rift oblique and rift parallel pre-existing fabrics during extension and their effect on deformation style: examples from the rifts of Thailand. Journal of Structural Geology 26, 1803-1829. doi: 10.1016/j.jsg.2004.02.014
Nakamura, K., 1977. Volcanoes as possible indicators of tectonic stress orientation – principle and proposal. Journal of Volcanology and Geothermal Research 2, 1-16.
Peacock, D. C. P., 2002. Propagation, interaction and linkage in normal fault systems. Earth-Science Reviews 58, 121-142. doi: 10.1016/s0012-8252(01)00085-x
Peacock, D. C. P., & Parfitt, E. A., 2002. Active relay ramps and normal fault propagation on Kilauea Volcano, Hawaii. Journal of Structural Geology 24, 729-742. doi: 10.1016/s0191-8141(01)00109-2
66
Peacock, D. C. P., & Sanderson, D. J., 1994. Geometry and development of relay ramps in normal fault systems. AAPG Bulletin 78, 147-165.
Personius, S. F., Crone, A. J., Machette, M. N., Mahan, S. A., & Lidke, D. J., 2009. Moderate rates of late Quaternary slip along the northwestern margin of the Basin and Range Province, Surprise Valley fault, northeastern California. Journal of Geophysical Research-Solid Earth 114, 1-17. doi: 10.1029/2008jb006164
Pezzopane, S. K., & Weldon, R. J., 1993. Tectonic role of active faulting in central Oregon. Tectonics 12, 1140-1169. doi: 10.1029/92TC02950
Pollard, D. D., & Aydin, A., 1984. Propagation and linkage of oceanic ridge segments. Journal of Geophysical Research 89, 10,017-010,028.
U.S. Geological Survey, Nevada Bureau of Mines and Geology, and California Geological Survey, 2006. Quaternary fault and fold database for the United States, accessed April 2014, from USGS web site: http//earthquakes.usgs.gov/regional/qfaults/.
Reches, Z., & Dieterich, J. H., 1983. Faulting of rocks in three-dimensional strain fields I. Failure of rocks in polyaxial, servo-control experiments. Tectonophysics 95, 111-132.
Reches, Z., 1983. Faulting of rocks in three-dimensional strain fields II. Theoretical analysis. Tectonophysics 95, 133-156.
Reches, Z., & Lockner, D. A., 1994. Nucleation and growth of faults in brittle rocks. Journal of Geophysical Research 99, 18,159-118,173.
Ritzinger, B., Egger, A. E., & Glen, J., 2013. Paleomagnetic mapping of late Miocene-Pliocene basalt flows in the northwestern Basin and Range: revealing the relationship between faulting and volcanism in an emergent extensional environment. Geological Society of America Abstracts with Progams 45, 170.
Rowland, J. V., Baker, E., Ebinger, C. J., Keir, D., Kidane, T., Biggs, J., Hayward, N., Wright, T. J., 2007. Fault growth at a nascent slow-spreading ridge: 2005 Dabbahu rifting episode, Afar. Geophysical Journal International 171, 1226-1246. doi: 10.1111/j.1365-246X.2007.03584.x
Rowland, J. V., & Sibson, R. H., 2004. Structural controls on hydrothermal flow in a segmented rift system, Taupo Volcanic Zone, New Zealand. Geofluids 4, 259-283. doi: 10.1111/j.1468-8123.2004.00091.x
67
Scarberry, K. C., Meigs, A. J., & Grunder, A. L., 2010. Faulting in a propagating continental rift: Insight from the late Miocene structural development of the Abert Rim fault, southern Oregon, USA. Tectonophysics 488, 71-86. doi: 10.1016/j.tecto.2009.09.025
Schopfer, M. P. J., Childs, C., & Walsh, J. J., 2006. Localisation of normal faults in multilayer sequences. Journal of Structural Geology 28, 816-833. doi: 10.1016/j.jsg.2006.02.003
Silver, P. G., Fouch, M. J., Gao, S. S., & Schmitz, M., 2004. Seismic anisotropy, mantle fabric, and the magmatic evolution of Precambrian southern Africa. South African Journal of Geology 107, 45-58. doi: 10.2113/107.1-2.45
Soliva, R., & Benedicto, A., 2004. A linkage criterion for segmented normal faults. Journal of Structural Geology 26, 2251-2267. doi: 10.1016/j.jsg.2004.06.008
Stewart, J. H., 1998. Regional characteristics, tilt domains, and extensional history of the later Cenozoic Basin and Range Province, western North America. GSA Special Papers 323, 47-74. doi: doi: 10.1130/0-8137-2323-X.47
Surpless, B., & Egger, A. E., 2006. The nature of extension on the western edge of the Basin and Range: Evolutoin of the Surprise Valley Fault System. Paper presented at the American Geophysical Union, Fall Meeting. http://adsabs.harvard.edu/abs/2006AGUFM.T23F..05S
Thomson, K., & Schofield, N., 2008. Lithological and structural controls on the emplacement and morphology of sills in sedimentary basins. In: K. Thomson & N. Petford (Eds.), Structure and Emplacement of High-Level Magmatic Systems, Vol. 302, Geological Society Publishing House, Bath, 31-44.
Trench, D., Meigs, A., & Grunder, A., 2012. Termination of the northwestern Basin and Range province into a clockwise rotating region of transtension and volcanism, southeast Oregon. Journal of Structural Geology 39, 52-65. doi: 10.1016/j.jsg.2012.03.007
Trudgill, B., & Cartwright, J., 1994. Relay-ramp forms and normal-fault linkages, Canyonlands National Park, Utah. Geological Society of America Bulletin 106, 1143-1157. doi: 10.1130/0016-7606(1994)106<1143:RRFANF>2.3.CO;2
Valentine, G. A., & Krogh, K. E. C., 2006. Emplacement of shallow dikes and sills beneath a small basaltic volcanic center - The role of pre-existing structure (Paiute Ridge,
68
southern Nevada, USA). Earth and Planetary Science Letters 246, 217-230. doi: 10.1106/j.epsl.2006.04.031
Wang, K., 2000. Stress-strain 'paradox', plate coupling, and forearc seismicity at the Cascadia and Nankai subduction zones. Tectonophysics 319, 321-338. doi: 10.1016/s0040-1951(99)00301-7
White, I. R., & Crider, J. G., 2006. Extensional fault-propagation folds: mechanical models and observations from the Modoc Plateau, northeastern California. Journal of Structural Geology 28, 1352-1370. doi: 10.1016/j.jsg.2006.03.028
Zoback, M.L., McKee, E.H., Blakely, R., & Thompson, G.A., 1994. The northern Nevada rift: Regional tectono-magmatic relations and middle Miocene stress direction. Geological Society of America Bulletin 106, 371-382. doi:10.1130/0016-7606(1994)106<0371:TNNRRT>2.3.CO;2
Zoback, M.L., & Thompson, G.A., 1978. Basin and Range rifting in northern Nevada: clues from a mid-Miocene rift and its subsequent offsets. Geology 6, 111-116.