The eastern California shear zone (ECSZ) and Walker Lane represent an evolving segment of the Pacific–North America plate boundary in the western United States. Understanding temporal variations in strain accumulation and release along plate boundary structures is critical to assessing how deformation is accommodated throughout the lithosphere. Late Pleistocene displacement along the Lone Mountain fault suggests that the Silver Peak–Lone Mountain (SPLM) extensional complex is an important structure in accommodating and transferring strain within the ECSZ and Walker Lane. Using geologic and geomorphic mapping, differential global positioning system surveys, and terrestrial cosmogenic nuclide (TCN) geochronology, we determined rates of extension across the Lone Mountain fault in western Nevada. The Lone Mountain fault displaces the northwestern Lone Mountain and Weepah Hills piedmonts and is the northeastern component of the SPLM extensional complex, a series of down-to-the-northwest normal faults. We mapped seven distinct alluvial fan deposits and dated three of the surfaces using 10Be TCN geochronology, yielding ages of 16.5 ± 1.2 ka, 92 ± 9 ka, and 137 ± 25 ka for the Q3b, Q2c, and Q2b deposits, respectively. The ages were combined with scarp profile measurements across the displaced fans to obtain minimum rates of extension; the Q2b and Q2c surfaces yield an extension rate between 0.1 ± 0.1 and 0.2 ± 01 mm/yr and the Q3b surface yields a rate of 0.2 ± 0.1–0.4 ± 0.1 mm/yr, depending on the dip of the fault. Active extension on the Lone Mountain fault suggests that it helps partition strain off of the major strike-slip faults in the northern ECSZ and transfers deformation to the east around the Mina deflection and northward into the Walker Lane. Combining our results with estimates from other faults accommodating dextral shear in the northern ECSZ reveals an apparent discrepancy between short- and long-term rates of strain accumulation and release. If strain rates have remained constant since the late Pleistocene, this could reflect transient strain accumulation, similar to the Mojave segment of the ECSZ. However, our data also suggest a potential increase in strain rates between ca. 92 ka and ca. 17 ka, and possibly to present day, which may also help explain the mismatch between long- and short-term rates of deformation in the region.
Understanding the temporal and spatial constancy of strain accumulation and release rates is a critical component to deciphering how deformation is accommodated in the lithosphere. For most plate boundaries, comparisons of short-term (decadal) geodetic and long-term (e.g., 103–106 yr) geologic plate motion data indicate that rates of strain storage and release are relatively constant over a wide variety of time scales (e.g., Sella et al., 2002). However, a growing number of studies along the Pacific–North America plate boundary suggest that distinct sections of the plate boundary record different spatial and temporal deformation rate patterns (e.g., Argus and Gordon, 2001; Peltzer et al., 2001; Lee et al., 2001a, 2001b, 2009a, 2009b; Oskin and Iriondo, 2004; Kylander-Clark et al., 2005; Walker et al., 2005; Kirby et al., 2006, 2008; Le et al., 2007; Frankel et al., 2007a, 2007b, 2008a; Oskin et al., 2007, 2008; Andrew and Walker, 2009; Ganev et al., 2010). Differential plate motion across the Pacific–North America plate boundary is primarily accommodated along the dextral San Andreas fault system, and much of the remaining motion is thought to be taken up by structures in the eastern California shear zone (ECSZ) and Walker Lane belt in the western United States (e.g., Burchfiel, 1979; Dokka, 1983; Stewart, 1988; Dokka and Travis, 1990; Reheis and Dixon, 1996; Reheis and Sawyer, 1997; Hearn and Humphreys, 1998; Dixon et al., 2000, 2003; Oldow et al., 2001; Bennett et al., 2003; Wesnousky, 2005a, 2005b; Kirby et al., 2006; Frankel et al., 2007a, 2007b, 2008a; Lee et al., 2009b).
Recent geodetic studies suggest that the transtensional ECSZ and Walker Lane accommodate ∼20% (9–14 mm/yr) of the cumulative Pacific–North America plate boundary deformation (Fig. 1; e.g., Humphreys and Weldon, 1994; Hearn and Humphreys, 1998; Thatcher et al., 1999; Dixon et al., 2000, 2003; Flesch et al., 2000; McClusky et al., 2001; Miller et al., 2001; Bennett et al., 2003; Hammond and Thatcher, 2007; Kreemer et al., 2009). However, at different latitudes in the ECSZ, previous work documents a match between long-term geologic and short-term geodetic rates in some regions (McClusky et al., 2001; Bennett et al., 2003; Frankel et al., 2007a; Lee et al., 2009b), while other areas are characterized by apparent transient strain accumulation (Peltzer et al., 2001; Oskin and Iriondo, 2004; Oskin et al., 2007, 2008; Frankel et al., 2007b).
Across the southern portion of the ECSZ, in the Mojave Desert, the geologic record of right-lateral slip is at odds with the geodetic rates of dextral shear (Oskin et al., 2008). At this latitude, a summation of dextral shear since the late Pleistocene provides a cumulative right-lateral deformation rate of ≤6.2 ± 1.9 mm/yr, representing only about half of the short-term rate of 12 ± 2 mm/yr (Oskin et al., 2008). However, north of the Garlock fault, at latitude ∼37°N, geologic and geodetic data agree with each other; a summation of slip rates across the main right-lateral fault structures (Owens Valley, Hunter Mountain–Saline Valley, northern Death Valley, and Stateline faults) shows a cumulative late Pleistocene strain release rate of ∼9–10 mm/yr, comparable to the region-wide 9.3 ± 0.2 mm/yr of dextral shear oriented at 323° ± 2° determined with global positioning system (GPS) data (Lee et al., 2009b; Bennett et al., 2003; Frankel et al., 2007a).
Geodetic studies also suggest that a majority of differential plate motion between latitude 37°N and 38°N is accommodated as right-lateral shear and the remainder as ∼1 mm/yr of extension accumulating normal to the right-lateral motion (e.g., Savage et al., 2001; Bennett et al., 2003; Wesnousky, 2005a). The two major right-lateral strike-slip faults at this latitude are the north-northwest–trending dextral-oblique White Mountain and northern Death Valley–Fish Lake Valley (DVFLV) faults (Fig. 1). A summation of slip rates across the Fish Lake Valley and White Mountains faults at latitude ∼37.5°N suggest that <3.5 mm/yr of the region-wide rate of right-lateral shear is accommodated on these structures (Frankel et al., 2007b; Kirby et al., 2006; Bennett et al., 2003). Geodetic models predict that right-lateral shear rates should increase toward the northern end of the DVFLV fault (Reheis and Dixon, 1996; Dixon et al., 2000), but geologic data indicate the opposite (Frankel et al., 2007b, 2008b, 2010). Together, these observations suggest that only about one-third of the short-term rate of dextral shear in this region can be accounted for in the geologic record on strike-slip faults.
Farther north, there is an eastward (right) step in the locus of active deformation where a belt of active left-lateral strike slip and down-to-the-northwest normal faults extends ∼80 km east of the DVFLV fault before continuing northward into the active right-lateral faults of the Walker Lane (Nielsen, 1965; Stewart, 1985, 1988; Oldow et al., 2001; Wesnousky, 2005b). This eastward step, known as the Mina deflection, connects the predominantly dextral northern ECSZ and central Walker Lane fault systems. Faults within the Mina deflection are predominantly east-west trending and are thought to accommodate clockwise block rotation via left-lateral strike-slip deformation (Stewart, 1985; Cashman and Fontaine, 2000; Faulds et al., 2005; Wesnousky, 2005a, 2005b). Lee et al. (2009a) showed that strain is transferred from the White Mountains fault into the Mina deflection via the Queen Valley fault; the Emigrant Peak fault and faults of the Silver Peak–Lone Mountain (SPLM) extensional complex also help partition strain away from the DVFLV fault zone north and east around the Mina deflection and into the Walker Lane (e.g., Reheis and Sawyer, 1997; Ganev et al., 2010; Oldow, 2003; this study).
In this study we address the apparent discrepancy between short- and long-term rates of deformation at the ECSZ–Walker Lane transition. We investigate the Lone Mountain fault, a prominent down-to-the-northwest normal fault within the SPLM extensional complex located to the south of the Mina deflection and east of the DVFLV and Emigrant Peak fault systems (Fig. 1). We document the amount of extension on fault scarps cutting alluvial fans, and calculate late Pleistocene extension rates for this fault by dating offset alluvial fans using 10Be terrestrial cosmogenic nuclide (TCN) geochronology. Our results suggest temporal variations in strain release rates in the late Pleistocene along the Lone Mountain fault and have important implications for understanding the dynamics and evolution of the Pacific–North America plate boundary deformation within the ECSZ–Walker Lane.
Deciphering the ECSZ–Walker Lane fault system requires interpretation of earlier structures as well as currently active faults. Transform displacement initiated along the Pacific–North America plate boundary and northeast-southwest extension in the Basin and Range was underway by the Early Miocene (e.g., Burchfiel, 1979; Atwater and Stock, 1998). As the transform plate boundary evolved, faults along the western margin of the Basin and Range began accommodating part of the transform motion, which is now manifest as the ECSZ and Walker Lane (Burchfiel, 1979; Stewart, 1988). By Late Miocene to early Pliocene time, the ECSZ and Walker Lane played a significant role in plate boundary dynamics and displacement was primarily taken up on a series of northwest-southeast–striking, right-lateral faults (Atwater and Stock, 1998; Stockli et al., 2003; Kylander-Clark et al., 2005; Bartley et al., 2008; Lee et al., 2009b; Burchfiel et al., 1987).
The SPLM extensional complex is a series of northeast-striking, down-to-the-northwest normal fault–bounded basins in the central ECSZ–Walker Lane, which includes the Lone Mountain, Clayton Valley, and Lida faults (Fig. 1; e.g., Oldow et al., 2009). This region of extension is bound to the north by a structural boundary in the Excelsior Mountains and to the south by the Sylvania and Palmetto Ranges (Fig. 1; Oldow et al., 2008). During the Late Miocene, the SPLM extensional complex accommodated the structural eastward step, known as the Mina deflection, between the north end of the DVFLV fault system and the Benton Springs and Bettles Well–Petrified Springs faults of the central Walker Lane (Fig. 1; Stewart, 1988; Oldow et al., 1989,1994, 2001, 2008; Oldow, 2003). Miocene extension occurred along on a low-angle detachment fault, the core of which is now exposed in Lone Mountain, the Weepah Hills, and Silver Peak (Kirsch, 1971). The lower plate exhibits amphibolite facies rocks that record a complex history of metamorphism and intrusive deformation (Oldow et al., 2008). Based on paleomagnetic analysis of volcanic rocks, Petronis et al. (2002, 2007, 2009) suggested that Silver Peak and Lone Mountain have undergone 20°–30° of clockwise vertical-axis rotation since the Oligocene. It is thought that this gradual rotation throughout the Miocene helped lock the detachment faults by the middle Pliocene, shifting the locus of strain transfer farther west and north to the Mina deflection (Oldow et al., 2008, 2009).
The Lone Mountain fault is located at the northeast part of the SPLM extensional complex along the southern edge of Big Smoky Valley. The fault is expressed as a series of prominent synthetic and antithetic normal fault scarps cutting across alluvium deposited along the northwest Weepah Hills and Lone Mountain piedmonts (Dohrenwend et al., 1992; dePolo, 2008). We mapped and dated offset alluvial fans in this region to determine late Pleistocene rates of extension across the Lone Mountain fault.
ALLUVIAL FAN STRATIGRAPHY
Geologic mapping of late Quaternary alluvial fan deposits and fault-related features was completed at a scale of 1:10,000 using U.S. Department of Agriculture 1-m-resolution color orthorectified aerial photographs (Figs. 2, 3, 4, and Supplemental Fig. 11). Alluvial fan deposits were identified and correlated across the field area based on standard criteria such as bar and swale morphology, soil development, desert pavement development, rubification and varnish hues, height above active channel, and degree of fan dissection (e.g., Bull, 1968, 1991; Ritter et al., 1993; Frankel and Dolan, 2007). The fan deposits identified along the northwest Lone Mountain piedmont are consistent with the previously defined North American alluvial lithostratigraphic framework (Bull, 1991) and include seven distinct units: Q4, Q3c, Q3b, Q3a, Q2c, Q2b, and Q2a. Active and recently active channels are represented by Q4 and occupy the lowest topographic positions in the landscape. All older surfaces are progressively higher in elevation relative to Q4; a brief description of each deposit is included in Table 1. Although we identified and correlated seven alluvial fan surfaces, here we focus on three of them, Q3b, Q2c, and Q2b, because these surfaces preserve the best expression of late Pleistocene deformation (Fig. 5).
LATE PLEISTOCENE FAULT DISPLACEMENT
The Lone Mountain fault is characterized by numerous down-to-the-northwest and several down-to-the-southeast (antithetic) fault strands striking between 210° and 235° that prominently displace the Q3b, Q2c, and Q2b alluvial fan deposits. Fault scarps were surveyed with differential GPS equipment to document the amount of vertical separation across the various fan units. We collected 83 profiles using a Trimble GeoXH GPS unit to measure fault perpendicular topographic profiles, and post-processing of data resulted in decimeter accuracy (Hoeft, 2010).
Following a surface-rupturing earthquake, normal fault scarps typically have free faces that rapidly reach a stable angle of repose. Scarp slopes are then thought to degrade into the landscape via diffusive processes, including rain splash and soil creep (e.g., Hanks et al., 1984; Arrowsmith et al., 1998; Bucknam and Anderson, 1979; Wallace, 1977). We measured scarp height at the scarp center point, using the Scarp Dater software of Hilley and Arrowsmith (2003), as illustrated in Figure 6. Errors associated with the displacement measurements include variations in surface roughness (e.g., amplitude of bar and swale; ≤25 cm), and uncertainty with GPS measurements (∼10 cm), but do not reflect the possibility of off-fault deformation.
Accurately determining the total vertical offset across the Q3b, Q2c, and Q2b alluvial fan surfaces is critical to determining extension rates; to document the total vertical displacement of the deposits, locations with well-preserved fan surfaces and fault scarps with the largest surface displacement were selected. Figures 3 and 4 show aerial views of profile locations, and Figures 7, 8, and 9C show the resulting topographic profiles with cumulative vertical offset. Figure 10 shows field photographs of scarps displacing the different alluvial fan surfaces. The Q3b topographic profile records a cumulative vertical offset of 6.0 ± 1.2 m with a maximum single scarp height of 4.0 ± 0.3 m (Figs. 7 and 9A). The Q2c topographic profile yields a total vertical offset of 14.3 ± 1.2 m and a maximum single scarp height of 7.2 ± 0.3 m (Figs. 8 and 9B). Vertical displacements of Q2b were measured along the largest identified fault scarp and result in an offset of 17.0 ± 0.3 m (Fig. 9C). Note that all scarp measurements are minimum estimates; while we are confident that we captured all measurable displacement across these fan units, we cannot completely rule out off-fault deformation or the potential for unidentified fault strands. For example, the hanging wall of Q2b is not preserved, so it is possible there is additional displacement that is no longer reflected in the surficial geology. To determine extension rates we dated the displaced Q3b, Q2c, and Q2b fan surfaces using TCN geochronology.
The accumulation of terrestrial cosmogenic nuclides, produced through the interaction of cosmic rays with minerals at Earth's surface, can be used to determine the age of alluvial landforms (e.g., Lal, 1991; Gosse and Phillips, 2001). TCN methods provide distinct advantages when determining surface exposure ages where datable organic materials are scarce or nonexistent, or where deposits are beyond the age limit of radiocarbon dating (ca. 40 ka). Here, we use TCN 10Be geochronology to determine the age of the offset Q3b, Q2c, and Q2b alluvial fans.
Beryllium-10 is produced through spallation and muon-induced reactions with Si and O in quartz (Gosse and Phillips, 2001). Quartz is the preferred sample material due to the relative stoichiometric simplicity and propensity to retain 10Be, as well as its resistance to weathering. We collected 21 samples: 8 surface samples each on the Q3b and Q2c surfaces, and 5 samples from a single depth profile on the Q2b surface. We selected surface samples that appeared to be in their original depositional geometry, remaining exposed at the surface since emplacement. Two potential problems that can be encountered with surface samples include inheritance and erosion. Samples ideally have no prior exposure history (inheritance), which could yield an artificially old surface age. Conversely, the effects from erosion or exhumation can provide artificially young ages; following surface abandonment, a clast that has been recently exhumed from beneath the surface or a clast that has had extensive surface erosion will have a lower TCN concentration.
For samples collected from the Q3b and Q2c deposits we targeted flat-lying quartzite cobbles with well-developed varnish and rubification located along flat, stable portions of fan surfaces (Figs. 3 and 4). All samples were surrounded by moderately to well-developed desert pavements with at least 20% of the cobble height exposed above the fan surface. Cobbles with obvious weathering or evidence of disturbance or recent exhumation were avoided. The upper 3–5 cm of each sample was used (Table 2). Sample locations are shown in Figures 3 and 4.
There were no suitably large, undisturbed cobble-sized or larger clasts located on the Q2b fan surface. As a result, the age of this deposit was determined with a depth profile. We dug an ∼2-m-deep pit in a stable portion of the fan surface (Fig. 3) and collected ∼1500 g of 0.25–1-mm-sized clasts at depth intervals of 25 cm, 55 cm, 85 cm, 125 cm, and 195 cm. The concentration of TCNs diminishes quasi-exponentially as a function of depth beneath the ground surface (e.g., Anderson et al., 1996; Gosse and Phillips, 2001), so by projecting the depth versus 10Be concentration curve to 0 cm depth a model surface age can be determined (e.g., Anderson et al., 1996; Hancock et al., 1999; Hidy et al., 2010). One advantage of using depth profiles to determine TCN surface exposure ages is that levels of inheritance can be quantified. At depths exceeding ∼2 m in alluvium the accumulation of TCNs is negligible. Thus, by extrapolating the depth versus TCN concentration curve to >2 m, inherited nuclide concentrations can be determined (e.g., Anderson et al., 1996).
Following collection, samples were processed at the Georgia Institute of Technology Cosmogenic Nuclide Geochronology Laboratory. Samples were crushed, ground, and sieved to the 250–500 μm grain-size fraction. Quartz was then isolated using procedures described in Kohl and Nishiizumi (1992). Beryllium was extracted from the samples by ion exchange chromatography, precipitated as Be(OH)2, and oxidized to BeO (e.g., Bierman et al., 2002). The BeO was ignited and mixed with niobium powder during packing in stainless steel cathodes. Samples were analyzed by accelerator mass spectrometry at the Lawrence Livermore National Laboratory Center for Accelerator Mass Spectrometry. Measured 10Be/9Be ratios were converted to 10Be concentrations and model ages were calculated with the CRONUS-Earth online calculator, version 2.2 (http://hess.ess.washington.edu/; Balco et al., 2008; Table 2).
Currently there is no consensus regarding the best way to model time-variable TCN production rates (e.g., Lal, 1991; Stone, 2000; Dunai 2001; Desilets and Zreda, 2003; Pigati and Lifton, 2004; Desilets et al. 2006; Lifton et al. 2005; Balco et al. 2008; Staiger et al., 2007). Because of this, we use a time-invariant production model to determine our fan ages (Lal, 1991; Stone, 2000). Age variability between different production rate models is low for the range of ages reported in this study; for a given sample, ages from the different production rate models are within ∼10% of each other (Hoeft, 2010; Supplemental Table 12).
Exposure ages for the Q3b surface have a range of ∼73 k.y., from ca. 15 to 88 ka (Table 2). A probability density plot (Fig. 11A) of the sample distribution shows that the largest peak is ca. 17 ka, where 4 of the 8 samples are tightly clustered; to the right of this peak there is a tail, which signifies longer exposure histories for the remaining samples. The higher 10Be concentrations in the four oldest samples are likely due to inheritance generated from prolonged exposure on hillslopes or in older fan units prior to erosion or due to extended residence time in channels and floodplains during transport from the source area to depocenter. Based on these observations, we believe that the four youngest samples likely represent a maximum age for the Q3b surface. The tight cluster of ages among the four youngest samples suggests limited inheritance in these samples. The weighted mean of these samples yields an age and standard deviation of 16.5 ± 1.2 ka. When compared to correlative fan deposits in the region there is good agreement between this age and ages reported for similar Q3 surfaces in Death Valley (Sohn et al., 2007; Frankel et al., 2008b, 2010; Owen et al., 2010).
The samples collected from the displaced Q2c surface have a large variance, ranging from ca. 89 ka to 311 ka (Table 2). Upon inspection of the probability distribution function (PDFs) in Figure 11B, it is apparent that at least six of the samples have some level of inheritance. The two youngest samples are in close agreement with ages of ca. 92 ka. Although inheritance is possible, we consider these to best approximate a maximum age for the surface. Because there is a large range of values older than ca. 92 ka, we feel it is appropriate to exclude the six oldest samples from surface age determination. Samples composing the primary peak of the PDF in Figure 11 range from 89.5 ± 8.3 ka to 95.4 ± 8.8 ka and yield a mean age of 92 ± 9 ka.
We chose the Q2c sampling site because the fan surfaces are the best expressed geomorphically, and have the most fault displacement across them. However, we note that the source area for this Q2c fan is very large, with numerous drainages coming off both Lone Mountain and the Weepah Hills (Supplemental Fig. 1 [see footnote 1]). A detailed investigation of these areas is beyond the scope of this study, but the likelihood of prolonged exposure in older sedimentary sequences prior to deposition during Q2c time is likely. This observation helps to explain the broad range of 10Be concentrations observed in the Q2c surface samples and is not uncommon in arid and semiarid environments (e.g., Owen et al., 2010; Armstrong et al., 2010).
When the Q2c dates are compared to correlative fan deposits across the region, the 92 ± 9 ka age is similar to those reported in previous studies. In Fish Lake Valley equivalent surfaces have 10Be ages of 94 ± 11 ka (Furnace Creek fan) and 71 ± 8 ka (Indian Creek fan) (Frankel et al., 2007b); similar fan deposits in Death Valley have 10Be and 36Cl ages of ca. 70–90 ka (Frankel et al., 2007a; Machette et al., 2008). Based on the close proximity of these locations, it is probable that climatic events controlling alluvial fan deposition are similar. Given the relative agreement between surface ages, we believe that the exclusion of the six oldest samples to determine the Q2c surface age is reasonable. We therefore take the offset Q2c surface age to be 92 ± 9 ka, but recognize that if the samples used for this age also have an inherited component, the fan could be somewhat younger.
The 10Be model age determined from the depth profile on the offset Q2b surface yields an age of 137 ± 25 ka (Fig. 12). To determine this age, we used the geologically constrained Monte Carlo approach developed by Hidy et al. (2010). The age was calculated from the best model fit of 100,000 Monte Carlo depth profile solutions through the top 4 samples in the depth profile (Fig. 12). The lowest sample in the profile was not used in the simulations because it does not strictly obey the theoretical depth versus 10Be concentration profile relationship (e.g., Gosse and Phillips, 2001). Individual sample 10Be concentrations from the Q2b depth profile are listed in Table 3. Although field evidence suggests that erosion of the fan surface is minimal, we allowed the model to simulate denudation rates ranging from 0 to 1 cm/k.y.; the best‑fit profile solution yielded an erosion rate of 0.18 cm/k.y. We assumed a constant bulk density through the fan profile, but permitted this value to vary randomly between 2.0 and 2.4 g/cm3 in the model. The best-fit model profile also yields an estimated inherited 10Be concentration of ∼58 +9/–12 × 104 at/g SiO2. When compared to similar fan deposits in Death Valley and Fish Lake Valley, our surface age of ca. 137 ka is within the middle of the ca. 120–180 ka age range reported by Machette et al. (2008) and Ganev et al. (2010).
LATE PLEISTOCENE EXTENSION RATES
Critical to our assessment of extension is accurately determining fault dip. At the northern end of the field area, we identified an exposed fault plane cutting alluvium in a channel wall incised orthogonal to the Lone Mountain fault (Fig. 13). We measured a dip of 40° toward 300° at this location (Fig. 13B). We use 60° as an upper bound for the dip of the Lone Mountain fault, which is a commonly assumed normal fault angle based on Andersonian mechanics. Two values of extension are reported for each offset fan deposit using the observed 40° dip and the more conservative 60° dip (Table 4). To illustrate the effect of dip on extension, a plot of extension versus dip angle is shown in Figure 14A. As the dip angle becomes shallower, the extension rapidly increases as the inverse to the tangent of the dip.
Q3b and Q2c
The vertical displacements across the Q3b and Q2c surfaces are 6.0 m ± 1.2 m and 14.3 ± 1.2 m, respectively. In both cases, several fault strands offset the alluvial fans, and the surfaces are continuous and identifiable in both the hanging wall and footwall (Figs. 3 and 4). The vertical displacements correspond to extension ranging from 3.5 ± 0.7 m to 7.2 ± 1.2 m for Q3b, and 8.8 ± 0.7 m to 18.1 ± 1.2 m for Q2c, using dips of 40° and 60°, respectively (Table 4).
Measurements across the Q2b surface yield a vertical offset of 17.0 ± 0.3 m. This distance is a minimum because Q2b is not preserved in the hanging wall. Using the observed dip of 40° and assumed dip of 60°, minimum extension values across Q2b range from 9.8 ± 0.2 m to 20.3 ± 0.4 m (Table 4). Although the full displacement across this unit cannot be measured, we can look at the nearby distribution of fault strands cutting the younger Q3b surface and infer the amount of displacement that may be missing across the Q2b surface. The topographic profile across Q3b shows several scarps of varying size (Fig. 7); the primary offset is approximately two-thirds of the total displacement, and three smaller synthetic scarps compose the remaining deformation (Figs. 3 and 7). If ruptures occur in the same location, and rates have remained constant, then approximately one-third of the total displacement across Q2b may have been on secondary fault strands that are not preserved. This implies that extension across the Q2b surface could be as much as 15 m (assuming a 60° dip) to 30 m (assuming a 40° dip) since deposition.
We use the probabilistic approach of Zechar and Frankel (2009) to calculate rates of extension by combining our results from TCN dating and the two horizontal displacement (extension) distances calculated using dip angles of 40° and 60°. The results of these calculations are reported in Table 4 and extension rate PDFs for each fan surface and the two dip angle scenarios are shown in Figure 15. Uncertainties in extension rates are reported at the 1σ confidence level.
The ca. 17 ka Q3b fan yields an extension rate between 0.2 ± 0.1 mm/yr and 0.4 ± 0.1 mm/yr. The offset ca. 92 ka Q2c deposit results in an extension rate of 0.1 ± 0.1 mm/yr to 0.2 ± 0.1 mm/yr. The Q2b extension rate is ≥0.1 ± 0.1 mm/yr since ca. 137 ka, regardless of fault dip. We show the range of possible extension rates as a function of dip angle in Figure 14B, which also illustrates the faster rate from the Q3b fans compared to the other two surfaces.
Rates of Strain Accumulation and Release
At the northern extent of the ECSZ, there are relatively few structures accommodating right-lateral shear (Fig. 1). Individual faults along a transect drawn perpendicular to the azimuth of differential plate motion (∼323° ± 2°; Bennett et al., 2003) at the latitude of the Lone Mountain fault include the White Mountain fault, the DVFLV fault, the Emigrant Peak fault, and the SPLM extensional complex. Previous studies reported late Pleistocene shear rates for the two main right-lateral strike-slip faults, ∼0.3–0.4 mm/yr for the White Mountain fault and ∼2.2–3.5 mm/yr for the DVFLV fault in Fish Lake Valley (Kirby et al., 2006; Frankel et al., 2007b). Reheis and Sawyer (1997) reported late Pleistocene (ca. 50 ka) and Holocene (ca. 6.5 ka) vertical slip rates across the Emigrant Peak fault of 0.4–1.3 mm/yr and 2.5–5.2 mm/yr, respectively. These vertical rates correspond to a late Pleistocene extension rate of 0.1–1.3 mm/yr and a Holocene extension rate of ∼0.9–5.2 mm/yr using fault dips ranging from 45° to 70° based on field observations reported by Reheis and Sawyer (1997). Our results are the first geochronologically determined late Pleistocene rates for the SPLM extensional complex.
The SPLM extensional complex is the most prominent tectonic feature east of the DVFLV and Emigrant Peak faults, and the Lone Mountain fault is the best geomorphically expressed structure within it. Our data show the Lone Mountain fault accommodates right-lateral shear (northwest-directed extension) at a rate of 0.1 ± 0.1–0.2 ± 0.1 mm/yr since at least ca. 92 ka. This rate potentially doubles to 0.2 ± 0.1–0.4 ± 0.1 mm/yr over the past ∼17 k.y. Based on field reconnaissance of the other faults within the SPLM extensional complex, it is likely that the Lida and Clayton Valley faults have comparable rates. Thus, the down-to-the-northwest faults in the SPLM extensional complex may be accommodating right-lateral shear at a rate of ≥1 mm/yr.
When our results are combined with the right-lateral shear rates on nearby faults, they suggest a discrepancy between short- and long-term cumulative strain rates across the region. A summation of late Pleistocene rates of right-lateral shear across the White Mountain, DVFLV, Emigrant Peak, and Lone Mountain fault suggest a right-lateral shear rate of 2.7–5.7 mm/yr (Fig. 16), which is substantially slower than the regional geodetically determined rate of 9.3 ± 0.2 mm/yr (Bennett et al., 2003). Based on our preliminary observations, the Lida and Clayton Valley faults may contribute an amount of strain similar to that of the Lone Mountain fault, but are not likely to account for the >3.5 mm/yr deficit.
The apparent mismatch between these rates can be resolved by three possibilities: (1) the region is undergoing transient strain accumulation, similar to the Mojave segment of the ECSZ; (2) strain rates are increasing, and as a result, the short-term geodetic rates should not be expected to match the longer-term geologic rates; or (3) deformation rates have remained constant, there is no anomalous strain accumulation, and additional structures, such as the Lida and Clayton Valley faults, account for the apparent slip deficit.
The various sections of the ECSZ appear to behave differently. South of the Garlock fault, studies have shown an apparent strain transient in the Mojave block, yet at the latitude of northern Death Valley there is general agreement between geologic and geodetic records of deformation (e.g., Peltzer et al., 2001; Frankel et al., 2007a; Oskin et al., 2008; Lee et al., 2009b). The summations of late Pleistocene geologic slip rates at the northern end of the ECSZ are at odds with the geodetic rate. Our observations suggest the Lone Mountain fault is helping accommodate strain across the ECSZ–central Walker Lane transition zone, but is not solely responsible for the missing slip over late Pleistocene time scales. While our results are consistent with the idea of a strain transient, they also suggest the possibility for an increase in rates of deformation along this part of the Pacific–North America plate boundary. Detecting increases in strain rate can be difficult because discrete intervals preserved in the geologic record that show variable displacement histories are often scarce (e.g., Gardner et al., 1987; Nicol et al., 2009). However, across the Lone Mountain fault, we identify variable rates of strain release over multiple late Pleistocene time scales that suggest that deformation may have been increasing from ca. 140 ka to the present across the northern ECSZ.
The offset Q2b surface yields a rate of ≥0.1 mm/yr over a time scale of ∼137 k.y. The uncertainty in this rate is high due to the fact that the Q2b surface is not preserved in the hanging wall. As a result, this rate could be as much as twice as fast (using a 40° fault dip) if the total offset across Q2b is one-third greater than that observed in the present-day geologic record. Rates of extension across the Q2c deposit range from ∼0.1 to 0.2 mm/yr since ca. 92 ka, but if the surface is younger, similar to the Fish Lake Valley and Death Valley fans with comparable soil development and morphology, the rate could be as much as 0.1 mm/yr faster. The rate derived from the Q3b fan is ∼0.2–0.4 mm/yr, which is faster than the rates from the Q2c or Q2b age offsets. Thus, given our best extension rate estimates and their associated uncertainties, the pace of deformation appears to be increasing over late Pleistocene time scales on the Lone Mountain fault; rates would have to take their extreme values to overlap, which we think is unlikely. Other possibilities, although also unlikely, are that fault dips have changed through time or that individual fault sets have systematically different dips, both of which could contribute to the difference in extension rates. Although the late Pleistocene and Holocene rates that Reheis and Sawyer (1997) reported for the Emigrant Peak fault have relatively high uncertainties, they suggest a similar pattern of increasing rates of deformation from the late Pleistocene to the present. If strain rates are increasing, as our data suggested, then the short-term strain accumulation rates do not necessarily need to match long-term strain release rates. In this case, the faster rates of right-lateral shear observed in the geodetic data may simply reflect an increase in strain accumulation over late Pleistocene to Holocene time scales as deformation becomes focused along this section of the Pacific–North America plate boundary.
Alternatively, if rates have remained constant through time across this region, then either nearby structures, such as the Lida and Clayton Valley faults, may accommodate the remaining strain, or deformation must be distributed throughout the lithosphere, such as along the mid-crustal detachment proposed by Wesnousky (2005b) and Wernicke et al. (2008). Regardless, faults throughout the SPLM extensional complex are not mutually exclusive with respect to other structures in the ECSZ and Walker Lane (e.g., Oldow et al., 1994). This is particularly true when considering the close proximity of the SPLM extensional complex to the northern tip of the DVFLV fault system, which has clear along-strike variations in slip and slip rate (e.g., Frankel et al., 2007b). These structures must therefore act in concert to both accommodate and transfer strain along this part of the Pacific–North America plate boundary.
Strain transfer between the major right-lateral faults of the ECSZ and the central Walker Lane is thought to primarily be accommodated by the left-lateral oblique faults of the Mina deflection and the normal faults of the Emigrant Peak and SPLM extensional complex (e.g., Oldow et al., 1994; Reheis and Sawyer, 1997; Frankel et al., 2007b; Lee et al., 2009a; Ganev et al., 2010;). Lee et al. (2009a) showed that strain from the White Mountain fault is transferred to the Mina deflection via the Queen Valley fault. Ganev et al. (2010) suggested that slip is transferred off the DVFLV fault system in a northward progression via extension and transferred to the Emigrant Peak fault system and around the eastern end of the Mina deflection. These structures, together with the faults of the SPLM extensional complex, transfer strain northeastward to the southern end of the right-lateral oblique Benton Springs and Bettles Well–Petrified Springs faults that make up the southern end of the Walker Lane (Figs. 1 and 16).
It is clear, based on orientation, proximity, and rates of extension along the Lone Mountain fault, that the faults of the SPLM extensional complex must partition at least part of the right-lateral slip in the northern ECSZ onto down-to-the-northwest normal faults. Although further work is needed to address the exact contribution of the SPLM extensional complex in accommodating right-lateral shear, the extension rates from the Lone Mountain fault help explain the gradual northward decrease in right-lateral slip rates observed along the DVFLV fault (e.g., Frankel et al., 2007b, 2008b, 2010) and the northeastward transfer of strain around the right step (i.e., Mina deflection) in the dominantly right-lateral northern ECSZ and southern Walker Lane.
New geologic and geomorphic mapping together with TCN geochronology of three offset alluvial fans along the Lone Mountain and Weepah Hills piedmonts provide the first geochronologically determined late Pleistocene extension rates for the Lone Mountain fault, an important component of the SPLM extensional complex and Pacific–North America plate boundary. Based on detailed field mapping, surveying of normal fault displacements, and TCN 10Be geochronology, we document an extension rate of ≥0.1 ± 0.1 mm/yr over the past 137 ± 25 k.y., a rate of 0.1 ± 0.1–0.2 ± 0.1 mm/yr since 92 ± 9 ka, and a rate of 0.2 ± 0.1–0.4 ± 0.1 mm/yr over the past 16.5 ± 1.2 k.y. These results show that the Lone Mountain fault has been active since at least the late Pleistocene, and extension rates increased between ca. 92 ka (Q2c) and ca. 17 ka and possibly to the present.
When late Pleistocene right-lateral shear rates from other faults at the north end of the ECSZ are combined with our extension rates, the cumulative long-term rate is slower than geodetic rates of deformation. This discrepancy implies that (1) there is a strain transient across this region; (2) strain rates are increasing, and geologic records should not be expected to match geodetic data; or (3) strain rates are constant, there is no transient, and strain is accommodated on yet-to-be documented structures or distributed throughout the lithosphere. Furthermore, our results show that the Lone Mountain fault is an important structure that acts to partition strain off of the major strike-slip faults in the northern ECSZ and helps transfer deformation around the Mina deflection northward into the Walker Lane.
This research was funded by National Science Foundation grants EAR-0929960 and EAR-0948570, the Geological Society of America, and Georgia Institute of Technology. We thank Jaime Convers and Zach Lifton for assistance in the field, Tina Marstellar for help with TCN sample preparation, Dylan Rood and Bob Finkel for TCN sample analyses, James Dolan for insightful discussions, and Alan Hidy for advice on depth profile modeling. Andy Newman and Zhigang Peng provided comments on an early draft of the manuscript, and formal reviews by Doug Walker, Nancye Dawers, and Ramon Arrowsmith helped improve the paper.