Abstract

The Stateline fault system is a 200-km-long zone of active right-lateral shear along the California-Nevada border, United States. Recent identification of 30 ± 4 km of dextral offset since 13.1 Ma on the southern segment of the fault requires significant displacement to extend farther south than has been commonly considered in the past. However, major structures exposed where the fault projects to the south reveal predominantly dip-slip extensional faulting, suggesting that displacement is transferred into substantial northwest-oriented extension in eastern Ivanpah Valley. New (U-Th)/He apatite data from Proterozoic orthogneiss in the southern McCullough Range and northern New York Mountains support this model by recording dates as young as 5 ± 1 Ma in the structurally deepest parts of the footwalls to the range-bounding normal faults. This age is distinctly younger than both the ages of regional extension in surrounding areas and the youngest (U-Th)/He apatite dates reported from the immediately adjacent Colorado River extensional corridor. Late Miocene–Pliocene extension in Ivanpah Valley, contemporaneous with that elsewhere in the Eastern California shear zone, provides an independent line of support that the eastern margin of the Eastern California shear zone extends to the California-Nevada border. If this age marks the onset of deformation on the State-line system, then long-term slip rates on the southern segment may be as high as 5 mm/yr, significantly higher than the present-day estimate of 0.9 mm/yr derived from geodetic observations across the northern segment of this fault system.

INTRODUCTION

The Eastern California shear zone and Walker Lane fault system are estimated to accommodate ~24% of the total relative right-lateral displacement between the Pacific and North American plates (e.g., Minster and Jordan 1987; Dokka and Travis, 1990a; Bennett et al., 2003). Compared with the San Andreas and related strike-slip faults along the plate boundary, the strain pattern in the Eastern California shear zone is remarkably complex, being distributed over a 125-km-wide zone of relatively short fault segments of diverse orientation and sense of slip (Fig. 1A; e.g., Stewart, 1988; Dokka and Travis, 1990a; Wesnousky, 2005). Uncertainties inherent in relating short-term (e.g., geodetic) fault displacement rates and long-term rates from geologic observation (e.g., Friedrich et al., 2003; Niemi et al., 2004; Oskin and Iriondo, 2004; Guest et al., 2007) further complicate regional models that are aimed at understanding the deformation history and seismic risk associated with this major plate boundary region.

An important avenue for improving tectonic models of the Eastern California shear zone is to better define the spatial and temporal boundaries of the active system. Recent analysis of geodetic data from a continuous global positioning system (GPS) network (Wernicke et al., 2004; Hill and Blewitt, 2006) and geologic observations of fault offsets (Guest et al., 2007) indicate that the northwest-striking, right-lateral Stateline fault system is currently active, and has undergone significant post–mid-Miocene to Holocene displacement. Therefore, this fault zone likely marks the eastern boundary of active right-lateral shear in the Eastern California shear zone between lat 37°N and 35.5°N (Fig. 1A). The geometry, kinematics, and displacement history of this structure are important constraints for any model of the evolution of the Eastern California shear zone as well as for seismic risk assessments for nearby urban centers such as Las Vegas and Pahrump.

Recent work on the Stateline fault system documents 30 ± 4 km of right-lateral offset since 13.1 Ma based on the identification and dating of volcanic and proximal landslide deposits as piercing points across the southeasternmost segment of the fault system (Fig. 1B; Guest et al., 2007). The magnitude of offset defined by these piercing points requires significant strain to be accommodated farther southeast than previously considered. In this paper we present new (U-Th)/He apatite thermochronologic data from the footwall blocks of normal faults in the northern New York Mountains and southern McCullough Range. These data support a model whereby right-lateral displacement on the State-line fault system is kinematically linked with extension in Ivanpah Valley, and possibly with northeast-striking left-lateral faults to the south. These data (1) improve current knowledge of the structural history of the Stateline fault, (2) suggest that the long-term slip-rate on the fault may be higher by a factor of two than previously constrained by the age of piercing points, and (3) provide independent support for the southeastern limits of post–6 Ma strain in the Eastern California shear zone.

GEOLOGIC SETTING

Stateline Fault System

The 200-km-long Stateline fault system is a zone of discontinuously exposed active dextral strike-slip faults that closely follows the California-Nevada state line from Amargosa Valley in the north to Ivanpah Valley in the south (Fig. 1B; Guest et al., 2007). Guest et al. (2007) described the system as comprising three main segments. The northern Amargosa segment is ~20 km southwest of the proposed Yucca Mountain nuclear waste repository and is thought to have ruptured during Holocene time based on fault scarp morphology (Piety, 1995). Faults within the central Pahrump segment are also believed to have accommodated late Quaternary to Holocene deformation based on geomorphic observations (Anderson et al., 1995; Niemi et al., 2005) and analysis of high-resolution airborne laser swath mapping of offset late Quaternary sediments (Niemi et al., 2005). Late Quaternary slip rates of ~0.01 mm/yr were assigned to faults in the Amargosa and Pahrump segments by Stepp et al. (2001), whereas Anderson (1998) suggested a late Quaternary slip rate of 0.1 mm/yr for the Pahrump segment. However, fault dislocation modeling of high-resolution continuous GPS data in the region (Basin and Range Geodetic Network [BARGEN]–Yucca GPS networks) suggests a contemporary right-lateral displacement rate on the Stateline fault system of 0.9 mm/yr (Wernicke et al., 2004; Hill and Blewitt, 2006). Evidence for late Quaternary movement on the southern Mesquite segment is scarce, but the fault trace appears to be defined by linear contacts between an active playa surface and late and/or middle Pleistocene playa deposits in Mesquite Valley (Schmidt and McMackin, 2006). These contacts are parallel to the faults in southern Pahrump Valley and project into the bedrock-hosted Stateline fault of Hewett (1956) at Stateline Pass.

Estimates of total right-lateral displacement on the Stateline fault system, based on offset pre-Cenozoic markers, include 25–45 km on the Amargosa segment (Poole and Sandberg, 1977; Cooper et al., 1982; Stevens, 1991; Schweickert and Lahren, 1997), 10–19 km on the Pahrump segment (R.L. Christiansen, 1968, personal commun., inStewart et al., 1968; Burchfiel et al., 1983), and ~3 km on the Mesquite segment (Burchfiel et al., 1983; Walker et al., 1995). Volcanic and proximal landslide deposits at Black Butte, in southernmost Pahrump Valley (BB in Fig. 1B), correlate with the 13 Ma Devil Peak rhyolite plug, in the southern Spring Mountains (DP in Fig. 1B), on the basis of structural and stratigraphic similarities, indistinguishable geochemical profiles, and U-Pb zircon ages (Guest et al., 2007). The limited size of the plug and proximal character of correlative units at Black Butte indicate dextral offset of 30 ± 4 km across the Mesquite segment of the fault system since 13.1 ± 0.2 Ma. These data provide the first robust estimate of minimum total late Cenozoic displacement on the Stateline fault system and correspond to a post–13 Ma average slip rate of 2.3 ± 0.35 mm/yr (Guest et al., 2007).

Along strike northwest of the Mesquite segment, this displacement is likely distributed among multiple known fault splays, possibly including those in the immediate vicinity of Yucca Mountain (e.g., Schweickert and Lahren, 1997), or possibly transferred westward to the Death Valley–Furnace Creek fault zone (Fig. 1B) via a left stepover between the Pahrump and Amargosa segments (Guest et al., 2007). How this displacement is accommodated to the southeast, in the vicinity of Ivanpah Valley, is less clear. We have observed evidence consistent with dextral shear (e.g., Riedel shear fractures) in fault exposures in Paleozoic sedimentary rocks of the southernmost Spring Mountains southeast of Stateline Pass (Fig. 2), and Hewett (1956) inferred the extension of the Stateline fault to the southern margin of the Lucy Gray Range (Fig. 2). Further projection would extend the fault along the southern margin of the southern McCullough Range and into the northern New York Mountains, but significant Tertiary strike-slip displacement across these ranges is unlikely, given the contiguous exposure of Proterozoic crystalline basement of similar age and metamorphic grade (Young et al., 1989; Wooden and Miller, 1990; Miller and Wooden, 1993). Rather, major structures bounding these ranges appear to be predominantly dip-slip extensional faults (Miller and Wooden, 1993; our preliminary mapping, Fig. 2). The latter observation raises the possibility that strike-slip displacement on the Stateline fault may be kinematically linked to extension in Ivanpah Valley and possibly transferred to northeast-trending left-lateral faults in the region. A similar link has been inferred on some regional maps (Burchfiel and Davis, 1988; Snow and Wernicke, 2000).

Ivanpah Valley and Surrounding Areas

Ivanpah Valley is one of the deeper basins in the Mojave Desert region (Blakely et al., 1999), with a depth to basement of ~2.5 km estimated from modeling of gravity data (Carlisle et al., 1980). The basin is asymmetric with the maximum depth to basement axis offset to the southeast from the lowest elevations in the playa (Fig. 2), suggesting a fault-bounded basin (Carlisle et al., 1980). The ranges bounding the east side of Ivanpah Valley, including the Lucy Gray and southern McCullough Ranges and the northern New York Mountains (Fig. 2), are cored by Paleoproterozoic (1.76–1.64 Ga) igneous and older supracrustal metamorphic rocks (e.g., Hewett, 1956; Bingler and Bonham, 1973; DeWitt et al., 1989; Wooden and Miller, 1990; Miller and Wooden, 1993). To the west and south of the valley, the region represents the convergence of the leading edge of the late Mesozoic Sevier fold and thrust belt (Burchfiel and Davis, 1977, 1988) and the Mesozoic magmatic arc (Beckerman et al., 1982; Walker et al., 1995). Late Neoproterozoic to Mesozoic sedimentary rocks crop out in the southern Spring Mountains, much of the Clark Mountains and Mescal Range, and a relatively small portion of the central New York Mountains. South of the Cretaceous Kokoweef and Slaughterhouse faults in the Mescal Range and central New York Mountains, Cretaceous granites (97–90 Ma) of the Teutonia batholith (Beckerman et al., 1982) dominate the ranges flanking southern Ivanpah Valley.

In the southern McCullough Range and New York Mountains, moderately to gently east dipping (55°–15°) Miocene volcanic and volcaniclastic rocks nonconformably overlie Proterozoic rocks and locally crop out along the eastern flanks of these ranges (Fig. 2; Hewett, 1956). Miocene rocks in the southern McCullough Range are composed of locally derived basal conglomerate and volcaniclastic conglomerate overlain by basalt flows, basaltic andesite flows, and pyroclastic breccia. Bingler and Bonham (1973) and DeWitt et al. (1989) interpreted these units as time correlative with the ca. 18–16 Ma Patsy Mine Volcanics of Anderson (1971). As much as 3 km of similar volcanic strata, ranging in age from 18.3 to 12.8 Ma (Faulds et al., 2002), are preserved in the adjacent Highland Spring Range. In the northernmost New York Mountains, sparsely preserved rocks of Tertiary age include arkosic sandstone and conglomerate overlain primarily by andesite flows and breccia. Farther south, in the vicinity of Castle Peaks and Barnwell (Fig. 2), the 18.5 ± 0.2 Ma Peach Springs Tuff (Nielson et al., 1990) is locally preserved as the oldest volcanic unit. The youngest unit is a silicic welded ash-flow tuff correlated with the 17.8–17.7 Ma Wild Horse Mesa Tuff (McCurry et al., 1995). In the adjacent Castle Mountains, as much as 1.5 km of volcanic rocks are preserved, ranging in age from 18.5 Ma (Peach Springs Tuff) to 12.8 ± 0.2 Ma (Turner and Glazner, 1990; Nielson et al., 1999).

North- to northeast-striking and west-dipping Cenozoic normal faults are common in and adjacent to the ranges bounding the east side of Ivanpah Valley (Hewett, 1956; Bingler and Bonham, 1973; Miller and Wooden, 1993). Several nested west-side-down normal faults displace Miocene volcanic rocks within the north-central New York Mountains, but have a net throw on these rocks of only ~100 m (Miller and Wooden, 1993). The inferred west-dipping McCullough fault between the Lucy Gray Range and McCullough Range has ~6100 m of west-side-down normal displacement defined by the projected trace of presumed correlative and tilted Miocene volcanic rocks (Hewett, 1956). A west-dipping range-bounding fault is inferred on the east flank of the southern McCullough Range (Hewett, 1956; Bingler and Bonham, 1973; DeWitt et al., 1989), although Faulds et al. (2002) inferred an eastward dip for this fault. A fault may also exist between the New York and Castle Mountains (Miller and Wooden, 1993; Nielson et al., 1999).

North and east of Nipton, northwest-striking and southwest-dipping normal faults bound the southern margins of the Lucy Gray Range and McCullough Range, whereas a northeast-striking and northwest-dipping normal fault zone bounds Proterozoic basement in the northernmost New York Mountains (Fig. 2). These range-bounding normal faults locally crop out as breccia zones with numerous fault surfaces in Proterozoic basement (Fig. 3A). The largely buried Nipton fault zone is a northeast-trending structure inferred to extend through much of eastern and southern Ivanpah Valley (Hewett, 1956; Burchfiel and Davis, 1977; Carlisle et al., 1980; Swanson et al., 1980; Miller and Wooden, 1993; Miller and Jachens, 1995; Miller, 1995). The fault system extends from east of Nipton, where it is coincident with the range-bounding normal fault mentioned above (indicating both dip-slip and strike-slip components), to south of Cima (Miller and Jachens, 1995). The fault zone appears to have ~15 km of total post- Cretaceous left-lateral offset, based on correlation of two sets of Cretaceous faults, the Kokoweef and Slaughterhouse faults (Burchfiel and Davis, 1971, 1977; Swanson et al., 1980) and the Cima and Pinto shear zones (Fig. 2; Miller et al., 1996; Wells et al., 2005).

Late Miocene gravel deposits are locally preserved within and between the McCullough Range and northern New York Mountains and adjacent ranges to the east. One important locality occurs along the crest of the New York Mountains just north of Barnwell (Fig. 2). Here, remnants of a 1.2-km-wide, ~180-m-deep paleovalley (Willow Wash paleovalley) comprise gravel, sand, and silt with crossbeds indicating fluvial transport to the east-southeast (Miller, 1995). Clast populations include angular to rounded clasts derived locally from the New York Mountains, but also a distinct population whose closest possible source is the present-day Mescal Range 30 km to the northwest (Miller, 1995). The latter population includes large rounded cobbles and boulders of Late Proterozoic quartzite and conglomerate (Figs. 3B, 3C), unmetamorphosed Paleozoic limestone and dolomite, and Cretaceous Delfonte Volcanics, none of which are known to crop out in the New York Mountains. However, the large size of the externally sourced clasts suggests that these fluvial gravels were deposited relatively close to their source area. The gravel deposits overlie and postdate the youngest Miocene volcanics in the New York Mountains (ca. 17 Ma) and are unconformably overlain by late Pliocene and Quaternary deposits. Correlative gravels in the adjacent Castle Mountains and Piute Range overlie 14–12.9 Ma rocks and are interbedded with lava flows dated as 10–8 Ma (Nielson, 1995; Nielson et al., 1999). The clast populations and transport direction of these late Miocene gravels imply that the present-day topographic barriers to their source created by southern Ivanpah Valley and the New York Mountains must have formed since ca. 13–8 Ma (Miller, 1995; Nielson, 1995; Nielson et al., 1999).

Relatively little unambiguous evidence for latest Quaternary and Holocene faulting has been recognized in the Ivanpah Valley region. Some fault segments cut early to late Pleistocene deposits along the north-trending normal faults east and west of the McCullough Range (Schmidt and McMackin, 2006). In a quarry wall within the northeast-trending, oblique-normal fault zone that bounds the northern New York Mountains, we observed a secondary fault splay that cuts Proterozoic basement and several layers of overlying alluvial fan gravel (Fig. 4). The basal gravel layer, which reveals ~20 cm of normal separation across the fault, directly overlies a sheared granitic basement breccia and displays evidence for Stage II soil development (Gile et al., 1966) with abundant pedogenic carbonate (small friable carbonate nodules) in a well-developed red Bk horizon (Birkeland, 1999). Three younger units, the lowermost of which apparently truncated the upper horizons of the soil during transport and deposition, overlie this paleosol. The lowermost overlying unit (conglomerate 1 in Fig. 4) is also cut by the secondary normal fault. The next youngest unit (conglomerate 2) appears to be an alluvial channel deposit that cuts into conglomerate 1 and overlaps the fault with zero apparent offset. The youngest unit is poorly consolidated and probably represents Holocene colluvium. The position of the faulted paleosol, below a well-consolidated gravel unit that has been erosionally truncated by an even younger gravel unit, suggests that it is likely Pleistocene or older in age. This is consistent with descriptions of other similar deposits in the area that have been assigned Pleistocene ages (Schmidt and McMackin, 2006). However, we did not conduct a rigorous soil analysis or CaCO3 assessment (e.g., Machette, 1985), and are therefore uncertain of the minimum age. Activity on the Nipton fault zone as young as early Quaternary is inferred based on the presence of small fault-bounded sedimentary basins filled with deposits of this presumed age (Miller and Jachens, 1995).

(U-TH)/HE APATITE DATA FROM THE IVANPAH VALLEY REGION

(U-Th)/He apatite thermochronology is a potentially powerful tool for investigating exhumation histories in a variety of tectonic settings including extensional systems (Farley and Stockli, 2002; Stockli, 2005). In general, the He partial retention zone (PRZ) in apatite ranges from 70 °C to 30 °C (Wolf et al., 1998; Farley, 2000), although recent studies indicate that radiation damage in apatite can elevate the effective closure temperature (Shuster et al., 2006; Flowers et al., 2007). Wolf et al. (1998) concluded that extreme variations in ambient surface temperature such as that in the Mojave Desert (mean annual temperature at Death Valley, California, is 25 °C) are unlikely to significantly modify He ages. Hence, the system is sensitive to exhumation from depths as shallow as ~1 km, assuming a geothermal gradient of ~25–30 °C/km and annual surface temperatures of 15–30 °C. In extensional fault systems, this behavior can be used to document the onset of rapid cooling related to tectonic exhumation of the footwall and can therefore provide an estimate of the timing of normal fault displacement.

In order to test the hypothesis that extensional faulting in Ivanpah Valley is kinematically linked to the Stateline fault system, we collected (U-Th)/He apatite data from the footwall blocks of normal faults bounding the southern McCullough Range and northern New York Mountains (Fig. 2). The sampling traverses are arranged approximately orthogonal to the surface trace of the normal fault and the nonconformity with overlying Miocene volcanic rocks at each end. The first traverse, comprising 10 samples in the southern McCullough Range, extends northeastward for a horizontal distance of ~12 km. The second traverse, comprising 7 samples in the northern New York Mountains, extends southeastward from the fault trace bounding this range over a horizontal distance of ~13 km. Based on our own measurements during sampling and inspection of the map by Miller and Wooden (1993), consistent trends in solid-state fabric orientation along each traverse attest to the general structural integrity of the basement blocks (Fig. 5). In the southern McCullough Range, southeast-striking and moderately southwest-dipping foliation persists, whereas a south-southwest–striking and shallowly northwest-dipping mylonitic foliation with consistent top-to-east shear sense is characteristic of the northern New York Mountains traverse. The structurally deepest samples in both traverses are 100–200 m across strike from the main trace of the faults in relatively fresh, unaltered K-feldspar-biotite granitic orthogneiss.

Analytical Methods

Analytical work was performed at the California Institute of Technology. Single apatite crystals were selected based on morphology, clarity, and lack of inclusions using a binocular microscope with crossed polarizers. Photomicrographs and grain dimensions were measured prior to placement in platinum packets and laser heating to 1065 °C for 8 min (House et al., 2000). Extracted He gas was spiked with 3He, purified using cryogenic and gettering methods, and analyzed on a quadrupole mass spectrometer. The degassed grains were then dissolved at 90 °C for 1 h in HNO3 with a spike of 235U-230Th-51V tracer and analyzed with a Finnigan Element inductively coupled plasma-mass spectrometer (ICP-MS). Grain mass was computed from the Ca concentration using 51V as an elemental spike and then used to calculate the apatite U and Th concentrations. Rare earth element data were also obtained by ICP-MS for each grain. Fragments of Durango apatite standard were analyzed by the same procedures with the samples. Hexagonal prism morphology was used as a reasonable approximation for the alpha-ejection correction (Farley et al., 1996). Most samples comprise 2–5 individually analyzed apatite grains. No apparent correlation is observed between calculated dates and effective U concentrations that might point to varying closure temperatures due to radiation damage (Shuster et al., 2006; Flowers et al., 2007).

Results

Limited mapping of basement structure and the scarcity of reliable late Cenozoic stratigraphic markers that can be matched across normal faults preclude detailed interpretations of the data profiles. However, the general patterns are clear. In both profiles, mean (U-Th)/He dates range from 65 to 5 Ma, with the oldest dates systematically at the shallowest structural levels near the nonconformity with Miocene volcanic rocks and the youngest dates in the structurally deepest part of the footwall blocks. The analytical results are presented in Table 1. Figures 6A and 6B are topographic profiles along the two sampling traverses showing the dates as a function of horizontal distance from the main basement-bounding normal fault and as a function of paleodepth beneath the projected Miocene unconformity. The latter relationship is also shown in Figures 7A and 7B as x-y plots. Clearly, the estimate of paleodepth becomes progressively more uncertain with increasing depth because of the possibility of differential tilting across the width of the block. Despite this uncertainty, the general consistency of solid-state fabric orientations and monotonic younging of mean dates away from the unconformities indicate that there are no significant structural repetitions within the basement. However, because relatively small faults may dilate the basement terrain in a more penetrative fashion (e.g., Anderson, 1971), the estimated paleodepths should be considered maxima. An estimate of this uncertainty in paleodepth (shown in Fig. 7) is derived by varying the tilt of the unconformity through the range of observed values in outcrop (i.e., 45° ± 10° in the southern McCullough Range, 15° ± 5° in the New York Mountains). This results in an uncertainty of ~±1 km at the structurally deepest levels.

In the southern McCullough Range, paleo-depth isolines are drawn at a tilt angle of 45°, which is the mean tilt between the two large exposures of Miocene volcanics at this end of the traverse. The structurally shallowest samples yield mean dates of 65 and 56 Ma, consistent with residing above the middle Miocene He PRZ. Progressively younger mean dates are seen at structurally deeper paleodepths, including 2 ca. 12 Ma samples at paleodepths of ~4–5 km and the 2 deepest samples at 8.2 ± 3.4 Ma and 8.2 ± 3.8 Ma (2σ standard deviations). One structurally high sample that does not fit this pattern, collected within a few meters of the basal unconformity with Miocene volcaniclastic rocks, yielded a date of 18.5 ± 2.0 Ma, which is nearly identical to the oldest volcanic dates of 18.3 Ma in the adjacent Highland Spring Range (Faulds et al., 2002). Thus, we interpret this sample data as thermally reset at the time of volcanic eruption. The 8 Ma mean dates in the structurally deepest part of the profile are conservatively interpreted as a maximum age for the youngest phase of faulting, assuming that these represent samples only partially reset prior to extensional exhumation. Although not clearly required by the data pattern, the 17–12 Ma dates in the central part of the profile are consistent with a phase of middle Miocene exhumation (Figs. 6A and 7A), which is constrained to have occurred in the interval of 16–11 Ma by the ages and degree of tilting of volcanic rocks in the Highland Spring Range (Faulds et al., 2002).

A similar (U-Th)/He data pattern is seen in the northern New York Mountains, where the Miocene unconformity dips ~15°. In this profile, the structurally shallowest and oldest dates are 60–50 Ma and the two structurally deepest samples are indistinguishable, 5.1 ± 1.0 Ma and 5.0 ± 1.4 Ma (2σ). The 5 Ma samples were no deeper than 3.5 km below the Miocene unconformity prior to the latest phase of unroofing, and were likely even shallower than that, allowing for the possibility of some middle Miocene unroofing. Thus, these samples were likely very close to the base of the partial retention zone at the onset of the latest phase of unroofing. This inference is strongly supported by the abrupt increase in dates for immediately overlying samples, implying that most of the tectonic exhumation of these samples, and therefore the event that caused unroofing, occurred after 6 Ma.

In summary, our results are consistent with an episode of mid-Miocene tilting and exhumation, but also require a significant component of fault-related exhumation in late Miocene to early Pliocene time. One additional sample from farther south yielded a date of 56 Ma (Fig. 2), but is not included as part of either profile.

DISCUSSION AND REGIONAL IMPLICATIONS

New (U-Th)/He apatite dates presented here indicate that a phase of extensional faulting in northeastern Ivanpah Valley occurred post–6 Ma, taking 5 ± 1 Ma from the New York Mountains profile as most representative of the timing of exhumation. While the data as a whole (and particularly those from the southern McCullough Range) permit an interpretation involving continuous extensional exhumation since ca. 16 Ma, we view a punctuated history involving two distinct periods of exhumation at 16–11 Ma and post–6 Ma as more likely. This is based on several lines of reasoning that include (1) major regional extension in all areas surrounding the eastern Mojave having ended by 11 Ma, (2) a direct kinematic link between extension in Ivanpah Valley and the strike-slip Stateline fault system, which is part of the Eastern California shear zone and for which independent evidence points to a ca. 6 Ma inception, (3) (U-Th)/He apatite dates associated with localized extension in other parts of the Eastern California shear zone that are similar to our youngest dates, and (4) the presence of non–locally derived gravel deposits in the New York Mountains that are now cut off from their source and that may be as young as 8 Ma.

Major regional extension in areas surrounding Ivanpah Valley is substantially older than the post–6 Ma extensional exhumation recognized in this study. In the adjacent Colorado River extensional corridor, large-magnitude extension responsible for unroofing metamorphic core complexes occurred between 23 and 11 Ma (e.g., John and Foster, 1993). Major extension in the immediately adjacent Highland Spring Range is constrained to 16–11 Ma (Faulds et al., 2002). Several studies have produced apatite fission track dates that generally range from 22 to 13 Ma in the corridor (John and Foster, 1993; Foster et al., 1993). More recent (U-Th)/He apatite studies in the Sacramento-Chemehuevi Mountains (Carter et al., 2006), Buckskin Mountains (Brady, 2002), and at Gold Butte (Reiners et al., 2000) also yield dates from 17 to 11 Ma with a strong concentration at 15 Ma (Fig. 1A). Extension in the central Mojave metamorphic core complexes farther west began at about the same time (ca. 23 Ma), but ended by 18 Ma (Bartley and Glazner, 1991). East and north of Las Vegas in the central Basin and Range, major extension occurred between 17 and 11 Ma (Axen et al., 1993; Snow and Wernicke, 2000).

The proximity of the southern piercing point defining 30 km of post–13 Ma offset across the Stateline fault (Figs. 1B and 2; Guest et al., 2007) and the relatively large magnitude of this displacement require a significant tectonic expression in Ivanpah Valley. Some displacement may be transferred to the northeast- trending, left- lateral Nipton fault zone that extends through the southern part of the valley. This structure offsets a pair of Cretaceous faults by ~15 km (Burchfiel and Davis, 1977; Swanson et al., 1980), locally cuts Miocene volcanic units, and may have been active into the early Quaternary (Miller and Jachens, 1995). Thus, we suggest that extension and unroofing in Ivanpah Valley represents a local kinematic expression of the termination of the dextral Stateline fault zone (Fig. 1B), and we speculate that this connection has likely existed throughout the history of fault system.

Whereas late extension in Ivanpah Valley clearly postdates regional extension in the adjacent Colorado River extensional corridor, the timing agrees well with the onset of trans-tensional deformation in the Eastern California shear zone. Dokka and Travis (1990a, 1990b) originally considered strike-slip deformation in the Eastern California shear zone to have most likely begun between 10 and 6 Ma. More recent estimates suggest inception as late as 5 Ma, perhaps kinematically coordinated with the opening of the Gulf of California and the birth of the San Andreas fault system (Gan et al., 2003, Oskin and Stock, 2003), although it is conceivable that the suggested 12 to 6 Ma strike-slip deformation in the eastern Gulf extensional province may have been linked to an early Eastern California shear zone (Gans, 1997; Fletcher et al., 2007). Major movement on the Kingston Range–Halloran Hills detachment system, which is the southernmost detachment of the Death Valley extensional system (Fig. 1B), occurred between 13.5 and 7 Ma (Davis et al., 1993).

Apatite (U-Th)/He dates from several widespread localities within the Eastern California shear zone–Walker Lane system record late Miocene to Pliocene fault-related exhumation similar to that identified in this study (Fig. 1A). These include 7–4 Ma dates from the Wassuk Range (Stockli et al., 2002), 5–3 Ma dates from the northern White Mountains (Stockli et al., 2003), and 7–4 Ma dates from the Avawatz Mountains (Reinert et al., 2003).

Late Miocene to Pliocene extension in Ivanpah Valley is also consistent with the presence of non–locally derived Miocene gravel deposits in a paleovalley perched high in the central New York Mountains. The closest possible source for these gravels is the Mescal Range, which is ~30 km to the northwest and across the modern topographic depression of Ivanpah Valley (Miller, 1995). The large clast sizes present in the channel suggest deposition in a proximal position relative to the source area. However, their position in a paleovalley allows the possibility that the large clasts were transported 30 km in a confined channel flow environment. Transport of the detritus could therefore have occurred either across a paleolandscape above a preexisting overfilled Ivanpah Valley, which subsequently transitioned to an underfilled (present) state as subsidence outpaced the fill rate, or the paleovalley was relatively close to the source area and the present separation was achieved by horizontal extension leading to the development of the Ivanpah depression. In either case, the present Ivanpah depression must postdate gravel deposition, which was underway no earlier than 13 Ma (Miller, 1995), and possibly as late as 8 Ma (Nielson, 1995). Furthermore, in either case mentioned above, the most likely source for the transition to a high subsidence rate to form the Ivanpah depression would have been northwest-southeast extension.

Contemporaneous extension in Ivanpah Valley and that elsewhere in the Eastern California shear zone provide an independent line of evidence that the eastern margin of the Eastern California shear zone extends as far east as the Stateline fault zone, a conclusion also reached based on previous geodetic (Wernicke et al., 2004) and geologic observations (Guest et al., 2007). If our (U-Th)/He apatite data record the onset of a distinct post–6 Ma phase of extension, it has important implications for long-term slip rates on the Stateline fault system. A minimum long-term slip rate of 2.3 ± 0.35 mm/yr is based on 30 ± 4 km of offset between 13.1 ± 0.2 Ma volcanic units across the southern fault segment (Guest et al., 2007). However, if all of this displacement has occurred since 6 Ma, then revised long-term slip rates for this segment of the Stateline fault zone would double to ~5 mm/yr. If true, then the discrepancy recognized by Guest et al. (2007) between the long-term geologic slip rate and 0.9 mm/yr estimated from geodetic data is even more severe.

Several possibilities could explain this discrepancy: (1) the geodetic data represent a transient period of slow slip, (2) faults in the Eastern California shear zone developed sequentially from east to west, and the Stateline fault zone is no longer as active as it was in the past, and (3) high geologic slip rates on the southern fault segment are transferred laterally westward to adjacent faults farther north and do not pertain to the northern segment where geodetic measurements have so far been made (Guest et al., 2007). The general lack of evidence for recent (Holocene) fault activity in the Ivanpah Valley area may lend support to the second possibility, and would suggest that strain in the Eastern California shear zone may have migrated westward since its inception (Dokka and Travis, 1990b). Additional geodetic data from across the central and southern segments of the State-line fault zone will help address the third possibility. These data may soon be available from both new permanent GPS sites across the central segment as part of the Yucca Mountain geodetic network and across Ivanpah Valley from Earth-Scope's Plate Boundary Observatory (http://pboweb.unavco.org/) (Figs. 1B and 2).

In summary, field observations and new (U-Th)/He apatite thermochronologic data provide important new insight into the kinematic history of the Stateline fault zone as well as the spatial extent of the modern Eastern California shear zone. An important implication of the results and interpretations presented here is that the eastern limit of Eastern California shear zone dextral shear does not extend farther south than the New York Mountains, and thus appears to be transferred west toward the plate boundary. The cryptic southern extension of the Southern Death valley fault zone may therefore mark the southeastern boundary of the Eastern California shear zone. An obvious question is, What will geodetic data for the southern Eastern California shear zone and southern Mojave region reveal?

This research was supported by U.S. Department of Energy contract FC-08-98NV12081 to Wernicke. We thank Ken Farley for access to his helium lab for the analytical work. Lindsey Hedges and Becky Flowers helped immensely in the sample preparation and analysis and their efforts are greatly appreciated. We are grateful to M. Oskin and T. Wawrzyniec for critical reviews that helped to significantly improve the presentation. Stereonets plotted with Stereonet 6.3 by R.W. Allmendinger.