Post–12 Ma deformation in the lower Colorado River corridor, southwestern USA: Implications for diffuse transtension and the Bouse Formation

The role of tectonism within the lower Colorado River (LOCO) corridor (Fig. 1) following large-magnitude extension in the middle Miocene Epoch remains unclear, especially with regard to the initial birth and evolution of the continental-scale Colorado River system as it integrated to the Gulf of California. Deposits of the ca. 6–4.8 Ma Bouse Formation, the ca. 4.5–3.5 Ma Bullhead Alluvium, and ancestral Colorado River deposits dating from the Pliocene Epoch through the Quaternary Period are preserved both in the subsurface and as patchy erosional remnants along much of the LOCO corridor (Crow et al., 2018). These deposits record the first arrival and integration of the Colorado River to the proto–Gulf of California (Metzger, 1968; Dorsey et al., 2007; Spencer et al., 2013; House et al., 2008; McDougall and Martínez, 2014), followed by Colorado River aggradation (Bullhead Alluvium; Howard et al., 2015) and aggradation-incision cycles (e.g., Chemehuevi Formation; Malmon et al., 2011) as the river continued to evolve.

There are lingering and linked debates about the timing and mechanisms that operated during evolution of the Colorado River system. One uncertainty is the age range of Bouse Formation deposition and whether the southward-advancing Colorado River had flowed into the northward opening Gulf of California by 5.3 Ma (Dorsey et al., 2007, 2011), 4.9 Ma (Spencer et al., 2013), between 4.8 and 4.6 Ma (Crow et al., 2019b), or had a punctuated polyphase history (Dorsey et al., 2018). A related second debate is whether the Colorado River initially met the proto–Gulf of California south of the Chocolate Mountains divide in southern California–Arizona (Spencer and Patchett, 1997; Spencer et al., 2013; Pearthree and House, 2014) or north of it (McDougall and Martínez, 2014; Beard et al., 2016; Dorsey et al., 2018). This latter debate revolves around the marine (Buising, 1990; McDougall and Martínez, 2014; O’Connell et al., 2017; Dorsey et al., 2018) versus nonmarine (Spencer and Patchett, 1997; Pearthree and House, 2014; Bright et al., 2018) versus estuarine (Crossey et al., 2015) depositional setting for the Bouse Formation in the southern Blythe Basin. To date, the most commonly cited model, especially for northern LOCO basins, is that Colorado River integration occurred via lake spillover (“fill and spill”; Pearthree and House, 2014). Our paper addresses a key component of all of these debates: How faithfully do the present positions of Bouse Formation deposits record initial depositional geometries, or the degree to which syn- and post-integration tectonism have influenced the geometry and preservation of the Bouse Formation?

The goals of this paper are to (1) characterize deformational features that postdate major middle Miocene extensional deformation within the region, and (2) present a tectonic model to explain this deformation and its influence on deposition of the Bouse Formation. We integrate new and previously published evidence for post-detachment LOCO corridor deformation. Previous findings include: potential for a slip discrepancy of 30–35 km in the LOCO study area relative to adjacent dextral shear domains (Bennett et al., 2016); post–12 Ma distributed dextral strain across the west-central Arizona metamorphic core complex belt (Singleton, 2015; Singleton et al., 2019); models for young deformation and uplift across the Chocolate Mountains (Ricketts et al., 2011; Beard et al., 2016); and latest Miocene propagation of Gulf of California tectonism into the area immediately south of the LOCO region (Bennett et al., 2015; Umhoefer et al., 2018). In addition, we present a new structural data set that documents diffuse (partitioned or spread out) pre- (>6 Ma), syn- (6–4.8 Ma), and post-Bouse (<4.8 Ma) deformation manifest primarily as normal and/or oblique dextral faulting. Geometric and kinematic characterization of these young structures is presented and interpreted to be the product of a distributed intracontinental strain field generated by dextral transtension inboard of the San Andreas plate-boundary fault. Post–12 Ma dextral transtension played an important role in localizing Bouse Formation depocenters, and in causing significant tectonic modification of original depositional elevations of the Bouse Formation during and after its deposition.

The LOCO corridor and the locations of ca. 13–5.4 Ma basalt, the Bouse Formation, and the Bullhead Alluvium are shown in Figure 1. The northernmost Cottonwood and Mohave Basins have the highest-elevation Bouse Formation outcrops at up to 560 m above sea level (m.a.s.l.), which have been interpreted to represent a lake highstand (House et al., 2008; Pearthree and House, 2014). The highest-elevation Bouse Formation outcrops in Chemehuevi Valley are located at 330–360 m.a.s.l., while the greater Blythe Basin (including Parker and Palo Verde Valleys) has Bouse Formation outcrops at up to 345 m.a.s.l. Bedrock divides between basins were progressively breeched via lake spillover during integration of the Colorado River to the Gulf of California (just south of Fig. 1) at ca. 5 Ma (Dorsey et al., 2018; Crow et al., 2019b).

The extent and geometry of subsurface Bouse Formation have been investigated in several stages by drilling (Metzger, 1968; Metzger and Loeltz, 1973; Metzger et al., 1973), and is currently under comprehensive investigation and synthesis by the U.S. Geological Survey (e.g., Cassidy et al., 2018). Structure contours for the base of Bouse Formation (Fig. 1) show ≥200-m-deep Bouse Formation depocenters in the southern Mohave and Blythe Basins. Also evident is a segmentation of depocenters by divides and modern topography, and depocenters aligned sequentially north-south (Cottonwood Basin), NW-SE (Mohave and Chemehuevi Basins), and NE-SW (Blythe Basin between Parker and Blythe).

The LOCO corridor is part of the southern Basin and Range province and is located between the actively deforming Eastern California shear zone to the west and the western Colorado Plateau margin to the east (Fig. 2). To the south, the Pacific–North America plate boundary is defined by the San Andreas transform fault and Gulf of California transtensional system. Migration of inboard diffuse deformation, manifest as the Walker Lane and Eastern California shear zone, started ca. 13–12 Ma (Dokka and Travis, 1990; Faulds and Henry, 2008). The dextral Stateline fault system near the Nevada-California border has accommodated ∼30 km of dextral offset since ca. 13 Ma (Guest et al., 2007). The Stateline fault system is though by some to form the northeastern boundary of the Eastern California shear zone, and to die out to the southeast 50 km northwest of the Mohave Basin into an oblique sinistral-normal fault system (Fig. 2; Mahan et al., 2009). Together, the Colorado Plateau, San Andreas, Gulf of California, and Stateline systems form boundaries for a group of structural domains by which inboard dextral shear related to Pacific–North America plate margin deformation since ≤13 Ma has been estimated (Bennett et al., 2016). Domains surrounding the LOCO region show 50–60 km of dextral shear (Fig. 2), whereas the LOCO domain shows ∼26 km of estimated dextral shear, which results in a 30–35 km discrepancy. A 30–35 km discrepancy suggests that a significant component of latest to post-Miocene LOCO deformation may remain undocumented (Bennett et al., 2016).

The Colorado River extensional corridor (black lines of Fig. 1; Howard and John, 1987) is well known for early to middle Miocene large-magnitude extension (e.g., Tosdal and Sherrod, 1986; Howard and John, 1987; Spencer and Reynolds, 1991). Northern to central basins (Cottonwood, Mohave, and Chemehuevi) are intrinsically related to early to middle Miocene deformation, with approximately east-west–trending topographic divides in lower Colorado River valleys (Pyramid, Topock, and Parker) spatially associated with metamorphic core complexes and low-angle detachment fault systems (Newberry, Chemehuevi, and Whipple, respectively). Each divide exhibits varying degrees of structural exhumation and extension, likely decreasing to the north toward Pyramid divide at the Newberry Mountains (Fig 1; Faulds et al., 2001).

The northern Blythe Basin (Fig. 1) exhibits classic metamorphic core complex deformation in the Moon Mountains and Plomosa Mountains (Strickland et al., 2017, 2018a; Spencer et al., 2018). Miocene extension in the southern Blythe Basin (Fig. 1) is more difficult to discern and may represent deformation in the upper plate of a low-angle detachment system (Tosdal and Sherrod, 1986). Early to middle Miocene NE-SW extension imparted a pervasive brittle structural fabric prior to deposition of the Bouse Formation. NE-SW extension in the northern LOCO region transitioned to ENE-WSW– and east-west–directed extension between ca. 16 and 12 Ma in the Mohave Valley (Anderson et al., 1972; Campbell-Stone et al., 2000; Faulds et al., 2001, 2002, 2016) and probably ca. 13–11 Ma in the central LOCO region (Singleton, 2015).

We characterize post–12 Ma deformation by analyzing deformation within key rock units (Fig. 3). “Bedrock” for this paper includes complexly deformed Paleoproterozoic basement through middle Miocene (>14–12 Ma) syn-detachment intrusive complexes and extrusive flows (e.g., Peach Springs Tuff and associated units). Bedrock was generally not studied given the poor age constraints on faulting. Exceptions include the Mount Manchester quadrangle (House et al., 2004), where a NW-SE–striking zone of minor and macroscale faults was mapped within the ca. 17–15 Ma Spirit Mountain pluton (Walker et al., 2007); the Plomosa Mountains, where faults and barite and carbonate veins crosscut mylonitic units with a top-to-the-northeast sense of shear in the footwall of the Plomosa detachment fault (e.g., Spencer and Reynolds, 1991; Strickland et al., 2017); and the northwestern mapped segment of the Bill Williams River fault zone (Fig. 1; Sherrod, 1988), where both Precambrian units and Miocene fanglomerate (alluvial fan material) are dextrally offset.

Units that postdate bedrock but predate the Bouse Formation are commonly fanglomerate (Fig. 3A) or finer-grained axial-basin deposits (e.g., Lost Cabin beds: House et al., 2005). Sandy fanglomerate deposits on the southern Whipple Mountains (Fig. 3A) have previously been interpreted as a subaerially deposited shoreline facies of the Bouse Formation (Dickey et al., 1980). Axial-basin deposits are dominantly composed of low-energy mud, silt, and sand beds (House et al., 2005). Interbedded ashes in the Cottonwood Valley area indicate that these deposits are at least as young as 5.5 Ma (House et al., 2008) and likely <5.24 Ma (Crow et al., 2019b).

Basaltic and rhyolitic units near and east of Parker divide (Fig. 1) in the central portion of the study area were extruded during bimodal volcanism (Figs. 1 and 3B; Suneson and Lucchitta, 1983). K-Ar and ⁴⁰Ar/³⁹Ar ages on these flows range from ca. 13 to 7 Ma (Suneson and Lucchitta, 1979, 1983; Howard et al., 1999; Singleton et al., 2014). A megacrystic basalt flow in Mohave Wash (Fig. 1) has yielded a K-Ar age on plagioclase of 5.4 Ma (Suneson and Lucchitta, 1983).

The Bouse Formation represents the inception of the Colorado River system in the region (Pearthree and House, 2014). The Bouse Formation has a thin basal carbonate consisting of travertine/tufa, thin-bedded marl, and calcareous sandstone (Dorsey et al., 2018), and is commonly underlain by an angular to well-rounded locally derived and oxidized gravel commonly referred to as “golden gravel” (House and Pearthree, 2016). Travertine/tufa typically encrusts basement rock and (more rarely) fanglomerate. Synsedimentary draping at a scale of tens of meters is evident at numerous locations (Fig. 3C; e.g., Pearthree and House, 2014). Travertine/tufa is onlapped by and interbedded with marl and locally derived sandstone (Figs. 3D, 3E; Pearthree and House, 2014). A succession of siliciclastic deposits (Fig. 3F; “interbedded unit” of Metzger, 1968) overlies the Bouse Formation carbonate unit, and includes a succession of clay, silt, and sand interpreted to represent deltaic sediment related to filling of water bodies with sediment supplied predominantly by the Colorado River (e.g., Fig. 3G; Pearthree and House, 2014). In the southern Blythe Basin, a green clay layer (Fig. 3F) is alternatively interpreted to represent the first arrival of the Colorado River (Dorsey et al., 2018), possibly via a spillover event (Bright et al., 2016).

The 4.5–3.5 Ma Bullhead Alluvium is an aggradational package that consists of rounded gravel and quartz-rich sand (House et al., 2005; Howard et al., 2015). The Bullhead Alluvium was deposited by a through-flowing river after incision into the Bouse Formation during establishment of a graded profile and is generally found >200 m above the river’s modern profile, but locally it is 75 m below sea level in the Blythe Basin (Howard et al., 2015; Cassidy et al., 2018; Crow et al., 2019a). Colorado River units that were deposited after deposition and incision of the Bullhead Alluvium are present throughout the corridor as well (e.g., Chemehuevi Formation; Malmon et al., 2011; House, 2016).

Plio-Pleistocene piedmont alluvial deposits form the modern tributary valley floors, and commonly cap erosional remnants of Miocene fanglomerate, Bouse Formation, Bullhead Alluvium, and Colorado River deposits throughout the piedmont. Piedmont alluvial deposits are typified as poorly sorted and locally derived alluvial fan and terrace deposits (House, 2016).

At the broadest scale, the following has been noted for post–12 Ma deformation within the LOCO corridor: (1) potential slip discrepancy in the LOCO region of up to 35 km relative to adjacent dextral shear domains (Fig. 2; Bennett et al., 2016), (2) potential for latest Miocene (ca. 8–6 Ma) propagation of Gulf of California tectonism into the southernmost portion of the LOCO region (Bennett et al., 2015; Umhoefer et al., 2018), (3) post–12 Ma distributed dextral shear across metamorphic core complexes in western and central Arizona (Singleton, 2015; Singleton et al., 2019), and (4) ≤9 Ma deformation and uplift across the Chocolate Mountains (Ricketts et al., 2011; Beard et al., 2016). Published evidence also includes outcrop- and map-scale deformation, such as (1) the ∼25-km-long NW-SE–striking Bill Williams River fault zone that accommodates up to 1 km of dextral offset of Proterozoic units and up to 200 m of apparent NE-down normal throw in ca. 10 Ma basalts near Parker divide (Fig. 1; Sherrod, 1988), (2) tilted Bullhead Alluvium and younger units (Metzger and Loeltz, 1973; Howard et al., 2013), (3) normal and dextral faults mapped within sandy fanglomerate on the south side of the Whipple Mountains (Dickey et al., 1980; Carr, 1993), (4) dextral faulting within the southern Plomosa Mountains (Miller and McKee, 1971), and (5) displacement of Pleistocene alluvial surfaces in the Blythe graben and Needles deformation zone, the latter of which includes graben structures, monoclinal folding, and reverse faulting (Menges and Pearthree, 1983; Pearthree et al., 1983, 2009; Thacker et al., 2017). Faulted Bouse Formation has been noted by Gootee et al. (2016) and Dorsey et al. (2017), and basin subsidence after Bouse Formation deposition within the Blythe and Mohave Basins has been suggested by alluvial fan morphometry (Pearthree and House, 2014), deeply buried and thick accumulations of Bullhead Alluvium below sea level in the southern Blythe Basin (Howard et al., 2015), and diffuse small-scale structural features (Thacker et al., 2017). Youngest Quaternary deformation is exhibited by the Blythe graben and Needles deformation zones, which cut Pleistocene surficial deposits with a maximum of 5 m of vertical displacement (Schell et al., 1981), and by modern geodetics (Kreemer et al., 2010) that suggest diffuse east-west extension in the northern part of the study area (Cottonwood Basin area; Fig. 1).

Evidence for continued deformation within and near the LOCO through the latest Miocene and into the Pliocene is shown by faults mapped in ca. 13–5.4 Ma basaltic lava flows as well as in fanglomerate with intercalated 14.0–12.8 Ma tuffaceous units (Cottonwood Basin; Faulds et al., 2001, 2003) and syn- to post-detachment fanglomerate that are locally folded (Sherrod, 1988). In Mohave Wash, basalt flows are as young as ca. 5.4 Ma based on K-Ar dating, a young age that is corroborated by their geometric relationship within modern washes (Suneson and Lucchitta, 1983). North-south–striking dikes associated with these and older basaltic flows suggest that east-west extension was established by the latest Miocene (≤13 Ma; Sherrod, 1988). These units provide constraints on continued deformation within and near the LOCO region through the latest Miocene into the Pliocene.

Significant ≤12 Ma post-detachment structures in the Buckskin-Rawhide Mountains (Fig. 1) show continued deformation to at least the late Miocene (Singleton, 2015), and are dominantly NW-striking dextral to oblique dextral faults with up to 8 km of apparent separation. Many of these dextral faults likely reactivated early to middle Miocene normal faults that are abundant across the range (Singleton, 2015). Similarly, post-detachment dextral and oblique dextral faults are common in other parts of the region (e.g., Miller and McKee, 1971; Singleton et al., 2019).

The Bouse Formation is cut by the “Big fault” in the southernmost Blythe Basin (Gootee et al., 2016). Up to 500 m of vertical distribution between basin-margin Bouse Formation outcrops and basin-axis subsurface Bouse Formation (Fig. 1; Metzger et al., 1973; Metzger and Loeltz, 1973; Olmsted et al., 1973, Turak, 2000) and 300 m of Bullhead Alluvium (Howard et al., 2015; Crow et al., 2019a) are observed in surface and subsurface data (Cassidy et al., 2018). This distribution has been used to suggest that some component of deformation was synchronous with or postdated deposition of the Bouse Formation (Pearthree et al., 2016, 2018; Karlstrom et al., 2017; Thacker et al., 2017). Significantly, the Bullhead Alluvium, originally graded to Pliocene sea level, is found ≥75 m below sea level in the axis of the Blythe Basin, and provides compelling subsurface evidence for deformation within the LOCO corridor since the Pliocene (Howard et al., 2015; Crow et al., 2019a). Furthermore, the Bullhead Alluvium is cut near Yuma, Arizona, by the Algodones fault, a fault that was likely the major plate boundary fault for most of the last 6 m.y. (Howard et al., 2015).

Deformation within the Chocolate Mountains anticlinorium has been shown to be ≤9.45 Ma based on faulted basalt and intercalated conglomerate, and has been characterized as north-south contraction (Ricketts et al., 2011). Contraction may be responsible for uplift of the Chocolate Mountains anticlinorium (Beard et al., 2016). Uplift could have affected the depositional history of the Bouse Formation north of the divide in numerous ways: (1) isolation of the Blythe Basin into a closed basin with subsequent lake filling (e.g., Pearthree and House, 2014); (2) incursion of marine water into a closed inland sea (McDougall and Martínez, 2014), perhaps through a narrow inlet across the Chocolate Mountains (e.g., Beard et al., 2016); (3) uplift (whether local or regional) postdating deposition of the Bouse Formation (e.g., Karlstrom et al., 2017); or (4) some combination of the three.

Structures within key geologic units (Fig. 4) were geometrically and kinematically characterized from field- and map-based data. For simplification, we demarcate pre-, syn-, and post-Bouse structures, deformation, and tectonism under the following criteria:

• Pre-Bouse: Structures in units older than the Bouse Formation. These consist predominantly of faults and folds in fanglomerate, as well as faults within basaltic lava flows, late Miocene volcaniclastic units, Miocene plutonic rocks in the Mount Manchester area, and veins and faults in the Plomosa Mountains that cut the Plomosa detachment fault and related structures. For the Bill Williams River area (Fig. 1), a large set of data (n = 52) for the Bill Williams River fault zone was collected in Proterozoic basement rock and Miocene fanglomerate. Another set of Bill Williams River fault zone data (n = 9) was collected in stratigraphically lower portions of Miocene fanglomerate mapped as latest syn- to post-detachment near the Bill Williams River (Sherrod, 1988). Though a relative minimum age for pre-Bouse structures is not always certain, their post-detachment middle to late Miocene age is deemed likely by unconformable relationships in many locations in the region (see below).

• Syn-Bouse: Structures within the Bouse Formation, as well as structures in ca. 5.4 Ma basalt flows from Mohave Wash (Fig. 1) and sandy fanglomerate from the southern Whipple Mountains. For Bouse Formation and sandy fanglomerate from the Whipple Mountains, these structures generally do not cut overlying units (commonly piedmont alluvial deposits), an unconformable relationship that indicates fault formation soon after deposition.

• Post-Bouse: Structures within Bullhead Alluvium, Colorado River sediments, and Quaternary gravel deposits. These data are solely geometric (i.e., lack kinematic information) from the Blythe graben, Needles deformation zone, and scattered localities.

These criteria do not preclude the possibility of faults in older units having formed during younger deformation events. However, unconformities throughout the study area are commonly observed where undeformed piedmont gravels (likely Quaternary in age) cap faults in older units, suggesting that these relative age designations are suitable simplifications. Such unconformities were observed at the “Big fault”, the Mesquite Wash Hills fault, various faults in the Cottonwood Basin, and faults in the southern Whipple Mountains.

In order to assess the likelihood of preexisting structural weaknesses, structural reactivation, and/or predominant structural fabrics that controlled latest Miocene to recent tectonism, mapped faults within the LOCO study area were compiled. Structural features and dikes were digitized from published and unpublished maps and structural studies (Fig. S1 and Data Item S1 in the Supplemental Material¹). Map scales ranged between 1:500,000 to 1:12,000. Structures were attributed with age and fault-type criteria. Given the complexity of accurately attributing the age of deformation, digital compilation results were classified into: “all faults” (indiscriminate of age), “all folds” (indiscriminate of age), “dikes”, and “≤6 Ma” to represent syn- to post-Bouse structures. Faults flagged as possibly ≤6 Ma (i.e., did not exhibit explicit age certainty) were also included in the ≤6 Ma designation. Fault orientations for ≤6 Ma faults were checked against the entire dataset to deduce the role of structural reactivation. Faults from published maps not identified as normal, reverse, or strike-slip were not assigned a type. Geometric analysis of the compilation was conducted using OATools (https://is.muni.cz/www/175417/OATools.html) (Kociánová and Melichar, 2016). End points of digitized lines were used to average the orientation of the overall structure. Therefore, the number of lines measured represents line segments, not singular faults. Results are plotted as rose diagrams (Fig. 5).

Shear-sense criteria were determined according to Petit (1987) on minor to macroscale faults in the field (Figs. 4A–4F). Where discrete fault zones were investigated, multiple kinematic data were collected along strike (e.g., Riedel shears, rough facets, etc.). We included faults of various scales, as fault kinematics are typically scale invariant (Marrett and Allmendinger, 1990).

Geometric analysis was conducted using Stereonet 10 (Allmendinger et al., 2012; Cardozo and Allmendinger, 2013). Kamb contours were calculated from poles to planes in order to decipher statistically significant geometric trends, and are contoured at a 2σ interval with a significance level of 3σ. Paleostrain analysis was conducted from fault kinematic data using FaultKin 7 (Marrett and Allmendinger, 1990; Allmendinger et al., 2012) to determine the incremental shortening (P) and extension (T) axes of faults. Modeling assumes that subsequent deformation has not reoriented fault surfaces (Marrett and Allmendinger, 1990), a fair assumption given low-strain latest to post-Miocene deformation within the region. Some data were collected from faults in the digital compilation (Fig. 5A). For the Mesquite Wash Hills fault (Fig. 1), 14 data points for the fault were simplified into a single average plane of 028°, 67°SE.

A synthesis of all geometric and kinematic results and the age of faulting for each structural domain and designation (pre- and syn-Bouse) are given in Table 1. Faults are shown to predominantly exhibit north-south and NW-SE strikes. Paleostrain results prevalently exhibit normal and strike-slip fault plane solutions and approximately east-west principal extension axes. In contrast, Cottonwood Basin and the northern Plomosa Mountains show an ESE-WNW principal extension axis. Geometric, kinematic, and paleostrain results are described in detail below.

Results for all faults (Fig. 5B) show prevalence of north-south– to NW-SE–striking faults. Characterizing these faults into type shows that normal faults are typified by a north-south strike with lesser NW-SE to NNW-SSE strike (Fig. 5C), and dextral faults show a nearly uniform NW-SE strike (Fig. 5D).

Fold axes show an east-west trend throughout the study area (Fig. 5E). These data include folds associated with the Chocolate Mountains anticlinorium (Ricketts et al., 2011; Beard et al., 2016), folded syn- to post-detachment fanglomerate in the Monkeys Head quadrangle near Parker divide (Sherrod, 1988), and folds associated with the Needles deformation zone (post-Bouse; Pearthree et al., 2009). Fold axes associated with the Needles deformation zone (Fig. 1) are NW-SE trending (∼315° azimuth).

Dikes associated with ≤13–5.4 Ma volcanism in the vicinity of Mohave Wash and Parker, Arizona (Figs. 1 and 5A), are north-south striking with a minor NW-SE strike (Fig. 5F). Dikes observed in the field were generally high angle to vertical. In some cases within the Monkeys Head quadrangle (Sherrod, 1988), dike spacing appeared systematic.

Figure 5A shows the distribution of faults within the digital compilation that were attributed as ≤6 Ma. Rose diagram results are shown in Figure 5G (for all faults regardless of type) and show approximately equal prevalence of north-south– and NW-SE–striking faults. Distinguishing faults by type shows that normal faults predominantly strike NW-SE as well as north-south (Fig. 5H). Dextral faults strike NW-SE (Fig. 5I). One of these dextral faults is part of the Algodones fault system.

Fault data throughout the entire study area (Fig. 6A) are shown on a Kamb-contoured stereonet (Fig. 6B), and show a propensity for north-south– (azimuth 000°/180° ± 22.5°; ∼41% of data) to NW-SE–striking (azimuth 135°/315° ± 22.5°; ∼34% of data) faults. East-west–striking faults form a distinct though minor population at the 2σ level (Fig. 6B). The maximum eigenvector (e1) for poles to all fault data (trend/plunge [T/P] = 258°/01°) equates to a mean plane of 348°, 89°NE. Normal (∼66% of data), dextral (∼21% of data), sinistral (∼9% of data), and reverse (∼4% of data) faults were all observed. Dip-slip faults (60°–90° rake) compose 41%, oblique-slip faults (30°–60° rake) compose 30%, and strike-slip faults (<30° rake) compose 29% of all data. Moderate- to high-angle faults with ≥60° dip are most common at ∼73% of the data, while ∼35% of ≥60° dipping faults exhibit a subvertical dip of ≥80°.

Faults preceding deposition of the Bouse Formation show a predominant NW-SE strike (azimuth 135°/315° ± 22.5°; ∼38% of pre-Bouse data) and north-south to NNE-SSW strike (azimuth 000°/180° ± 22.5°; ∼34% of pre-Bouse data). The maximum eigenvector (e1) of poles to fault data (T/P = 255°/06°; Fig. 6C) equates to a mean plane of 345°, 84°NE. Faults display shallow-angle to high-angle dips and an array of slickenline orientations, though moderate- to high-angle faults (≥60° dip) represent 68% of pre-Bouse data, and dip-slip (60°–90° rake; 39% of pre-Bouse data) and oblique-slip (30°–60° rake; 33% of data) slickenline orientations are more common for pre-Bouse faults.

North-south–striking faults (azimuth 000°/180° ± 22.5°) are most common at ∼54% of syn-Bouse data, as well as minor NW-SE–striking faults (azimuth 135°/315° ± 22.5°) at ∼26% of syn-Bouse data (Fig. 6D). The maximum eigenvector (e1) of poles to faults (TP = 079°/06°; Fig. 6D) gives a mean plane of 169°, 84°W. Syn-Bouse faults are typically moderate to high angle (≥60° dip; 78% of all syn-Bouse data) and have predominantly normal sense of slip (60°–90° rake; 83% of syn-Bouse data). The “Big fault” (Figs. 3D, 3E, 4F) is the most prominent syn-Bouse fault with an estimated throw of ≥35 m based on proximal outcrops of Bouse Formation travertine observed in the footwall and base of Bouse Formation on golden gravel in the hanging wall.

Discrete post-Bouse fault surfaces are typically absent, and units that postdate the depositional time frame of the Bouse Formation (<6–4.8 Ma) are typically observed to unconformably overlie faulted pre- and syn-Bouse strata. Significant post-Bouse faults are most evident within the Blythe graben and Needles deformation zone (Fig. 1), though we were unable to find any kinematic indicators at these locations.

Figure 7 shows kinematic data collected from the entire study area (Fig. 7A) and paleostrain results. Slip lineations show three distinct groups (Fig. 7B). Given field observations, most north-south–striking faults were observed to be normal, while NW-SE–striking faults were observed to be oblique dextral-normal, dextral, and rarely oblique dextral-reverse. Paleostrain analysis shows an east-west principal extension axis (T/P = 085°/00°) that corresponds to an approximately north-south–striking normal fault plane solution without oblique slip (Fig. 7C).

It was noted in the field that some pre-Bouse structural zones show a propensity for oblique (Mount Manchester) and subhorizontal (Bill Williams River) slip lineations. Given these observations, we break these data into separate structural domains presented below.

Southern Cottonwood Basin, including the area near Pyramid divide–Davis Dam (Figs. 7D, 7E), shows a NE-striking normal fault plane solution and WNW-ESE principal extension axis (T/P = 118°/07°). Most faults noted in the Cottonwood Basin–Davis Dam domain are high-angle normal faults, some showing oblique slip (Fig 4C). Faults to the north are characterized by NE-SW strikes, whereas faults in the south (closer to Davis Dam) were NW-SE to north-south striking.

Faults from the Mount Manchester area show an east-west principal extension axis (T/P = 088°/01°) and north-south–striking normal fault plane solution with slight to moderate oblique slip (Figs. 7F, 7G). All fault data were collected in a ∼1-km-wide NW-SE–trending zone of minor to outcrop-scale faults and fault surfaces (Fig. 4A; Fig. S2 [^{footnote 1}]). Many faults at the northwestern end are north-south striking. Faults and fault surfaces are noticeably rare to absent outside of this zone. Possible strike-slip conjugates with a north-south–trending horizontal σ₁ bisector are also observed here.

The NW-SE–striking Bill Williams River fault zone (Sherrod, 1988) accommodates up to 1 km of dextral offset and up to 200 m of apparent down-to-the-northeast normal throw in ca. 10 Ma basalts at its southeastern segment. Slickenlines in these basalts exhibit subhorizontal lineations (Fig. 4B) and NW-SE–striking dextral and NE-SW–striking sinistral conjugate sets.

North of the Bill Williams River, the NNW-SSE–striking Little Black Mountain fault zone, informally named herein and partially mapped by Sherrod (1988), cuts similar-age basalts and associated bimodal rhyolitic units. The central portion of the Little Black Mountain fault may have up to 80 m of reverse separation, though kinematic indicators along its trace suggest dextral motion. The northwestern segment of the fault exhibits a vertical to steeply NE-dipping (presumably positive) flower structure in a cliff face of sub-basalt rhyolitic units that transitions ∼0.5 km northwest along strike into an interpreted north-south–striking normal fault with Proterozoic basement in its footwall to the west and latest Miocene basalt and rhyolite in the hanging wall to the east. North of here along Mohave Wash (Figs. 1, 7A), approximately north-south–striking faults demonstrably have normal slip sense and cut both 13–7 Ma and ca. 5.4 Ma basaltic units. Another normal fault cutting ca. 13–8 Ma basalts and late Miocene sediments is mapped (Howard et al., 1999) 7 km north of the northern terminus of the Little Black Mountain fault zone.

A collective model for all pre-Bouse fault data from Proterozoic, fanglomerate, basalt, and rhyolite units shows a NNW-SSE–striking fault plane solution and WSW-ENE principal extension axis (T/P = 075°/04°; Fig. S3 [^{footnote 1}]). However, numerous subhorizontal slip lineations and dextral shear-sense criteria were observed in 10 Ma basalt within the Bill Williams River fault zone and in volcaniclastic and rhyolitic units within the Little Black Mountain fault zone. Basaltic, volcaniclastic, and rhyolitic units are stratigraphically higher than fanglomerate in the Bill Williams River and Mohave Wash areas (Fig. 1). A paleostrain model comprising fault data for only these latest syn- to post-detachment fanglomerate shows a WNW-ESE–striking normal fault plane solution with minor oblique slip and NNE-SSW extension axis (T/P = 200°/22°; Fig. S3). Given these results and the likely syn-detachment strain history, faults from Precambrian and fanglomerate units within the Bill Williams River fault zone are excluded from further analysis and the discussion.

Slip lineations in post-detachment units from the Bill Williams River and Little Black Mountain fault zones and Mohave Wash record dominantly dextral slip with very minor obliquity. These results (Figs. 7H, 7I) show a strike-slip fault plane solution and WNW-ESE principal extension axis (T/P = 103°/08°).

Structural data within the Plomosa Mountains (Fig. 8A) demonstrate disparate extension directions between syn-detachment and post-detachment structures that formed during early to middle Miocene core complex development in the northern Plomosa Mountains (Strickland, 2017). Detachment-related deformation defines subhorizontal NE-directed extension at 040°–048° (Fig. 8A). Northeast-directed extension differs from that recorded by post-detachment barite and carbonate veins (T/P = 298°/11°; Fig. 8B) and post-detachment faults (T/P = 109°/06°; Figs. 8D, 8E) that primarily strike north-south to NE-SW and record WNW-ESE extension axes. Northeast-striking sinistral and oblique sinistral-normal faults are common, whereas NW-striking dextral and oblique dextral-normal faults that are observed elsewhere in the LOCO region are uncommon in the northern Plomosa Mountains.

Small-scale faults were measured in gently dipping, moderately consolidated alluvium in the southern Plomosa Mountains (Figs. 8F, 8G). These faults include both NW-striking dextral as well as NNE-striking sinistral faults, and normal and oblique-normal faults, recording overall east-west extension (T/P = 080°/03°; see also Wyatt et al., 2017).

The NE-striking Mesquite Wash Hills fault in the eastern Mesquite Mountains (Fig. 8A; Knapp, 1989, 1993) juxtaposes brecciated crystalline gneiss to the southeast against moderately to poorly consolidated alluvium to the northwest. Kinematic indicators along the fault consistently record sinistral slip that is compatible with WSW-ENE extension (T/P = 251°/16°; Figs. 8H, 8I). The faulted alluvium appears to be overlain by, and locally intercalated with, basal Bouse Formation carbonate. Unfaulted Bullhead Alluvium overlaps the fault in one location.

A composite paleostrain model for all pre-Bouse faults (Fig. 9A) shows an east-west principal extension axis (T/P = 085°/02°) with associated north-south–striking normal fault plane solution, and includes all faults from the structural domains described above. The composite model for syn-Bouse faults (Fig. 9B) exhibits a principal extension axis oriented east-west (T/P = 266°/08°) and an approximately north-south–striking normal fault plane solution with minor oblique slip. Syn-Bouse results are from the central (Bill Williams River and Mohave Wash areas) and southern (“Big fault”) portion of the region (Fig. 1). For post-Bouse faults, reliable kinematic indicators were not found throughout the region. Therefore, we rely on the north-south– and NW-SE–striking geometric results of ≤6 Ma faults (Figs. 5G–5I) to make inferences on the likely strain field in which post-Bouse faults formed.

Strain within the LOCO corridor is manifest by characteristic fault geometries and types. Pre- (>6 Ma), syn- (6–4.8 Ma), and post-Bouse (<4.8 Ma) structures are composed of north-south–striking normal and NW-SE–striking oblique dextral-normal to dextral faults (Figs. 5 and 6). East-west–striking contractional structures form a minor but notable proportion of the overall results, whereas minor east-west–striking and NE-SW–striking sinistral structures become more prevalent west of the study area in the eastern Transverse Ranges and Eastern California shear zone (e.g., Miller et al., 2014; Miller, 2017). Given the results of the age-independent fault compilation, the study area appears to have mainly north-south and NW-SE structural geometries (Figs. 5B, 5C, 5D). Geometries of faults in units that precede deposition of the Bouse Formation and late Miocene fanglomerate suggest that an engrained brittle structural fabric was in place prior to deposition of the Bouse Formation, and further suggests that the Bouse Formation was deposited over an intensely faulted upper crust. The brittle fabric is likely a product of numerous faulting regimes of diverse ages that formed from NE-SW–directed contraction and extension between the Cretaceous Period and the Miocene Epoch, with east-west–directed extension during the latest phases of large-magnitude middle Miocene extension (Campbell-Stone et al., 2000; Faulds et al., 2001; Singleton, 2015). A spatial change in the dominant structural trend is apparent near the Chemehuevi Mountains, where north-south strikes prevail to the north and NW-SE strikes prevail to the south (Fig. S1 [^{footnote 1}]). We speculate that the change in structural trend could be the result of changing stress directions during latest middle Miocene extension that resulted from time-transgressive deformation.

Our field observations of faulted latest Miocene (<12 Ma) to Pliocene units, as well as ample geometric and kinematic fault data and paleostrain analyses, show that modest deformation within the study area continued following middle Miocene extension. Continued deformation was likely the result of a transitioning strain field that changed from dominantly NE-SW– to east-west–directed extension at the latest phase of large-magnitude extension (Campbell-Stone et al., 2000; Singleton, 2015). Structures within the Buckskin-Rawhide Mountains constrain this changing stress field to ≤12 Ma as exemplified by dextral and oblique dextral structures with separations on a scale of tens to hundreds of meters. The geometry and kinematics documented in the Buckskin-Rawhide Mountains by Singleton (2015) are similar to our results and show a principal post-detachment extension axis oriented east-west (T/P = 267°/09°, as reported by Singleton [2015]). Post-detachment faulting in the Buckskin-Rawhide Mountains is interpreted to represent a dextral to dextral-transtensional strain field, which could have been the result of distributed shear set up by the incipient San Andreas fault system (Singleton, 2015).

Our work definitively identifies an east-west extensional strain field to the latest Miocene (<12 Ma; Fig. 9A). Latest Miocene (and younger) deformation is perhaps most convincingly shown by the orientation of ca. 13–5.4 Ma north-striking dikes within the Bill Williams River and Mohave Wash area (Fig. 5), as dikes preferentially form perpendicular to the extensional strain field. Numerous north-south–striking normal faults and east-west principal extension axes corroborate an east-west extensional strain field interpretation. Therefore, east-west extension was likely established by 13–11 Ma (start of basaltic volcanism locally) and continued through 7 Ma (latest basaltic flows and dikes dated with certainty in the Mohave Wash area). Though we find it unlikely that the sandy fanglomerates on the southern Whipple Mountains are part of the Bouse Formation, their age likely immediately predates that of the Bouse Formation. Additionally, modern age dating of young basalt flows in Mohave Wash is warranted to test young ages there, but it is likely that their age is within or just prior to the depositional time frame of the Bouse Formation. Therefore, syn-Bouse structures, such as the “Big fault”, suggest continued east-west extension from 6 to 4.8 Ma (Fig. 9B). Lastly, the Needles deformation zone and Blythe graben record post-Bouse deformation that is observed as reverse faulting, folding, and normal faulting in alluvial fan deposits, thus indicating that deformation continued locally into the Quaternary Period. Diffuse low-strain deformation from geodetic studies, showing up to 1.4 mm/yr (strain rate of 4–8 × 10⁻⁹ per year) east-west diffuse extension in the southern Cottonwood Basin (Kreemer et al., 2010), suggests that deformation, at least in the northern LOCO region, remains active. Given these observations, we posit that strain was dominantly east-west extensional before, during, and after deposition of the Bouse Formation and integration of the Colorado River system (Fig. 10).

East-west extension is viable for all or most pre- and syn-Bouse structures considered in this study. However, it is not readily consistent with the NW-SE–striking Blythe graben and Needles deformation zone. Geophysically determined faults (Beard et al., 2011) are found coincident with (Blythe graben; Miller et al., 1979) and along strike of (Needles deformation zone) these structures. In the absence of kinematic indicators, we interpret the Blythe graben as dextral oblique-normal based on the propensity of approximately NW-SE–striking faults to be oblique dextral-normal and dextral faults as shown by our study (Figs. 5G–5I, 7I, 8G). Likewise, we observe that the Needles graben occurs within a deformation zone that exhibits reverse faulting and folding that is geometrically and kinematically consistent with a NW-SE–striking zone of dextral shear (e.g., north-south–striking normal and east-west–striking reverse faults; Fig. 11). Furthermore, the northwestern terminus of the Needles deformation zone appears to transfer strain into north-south–striking normal faults (Pearthree et al., 2009). Therefore, the Blythe graben and Needles deformation zone may represent zones of dextral shear. We stress, however, that the unconsolidated nature of the units that these features formed in make it difficult to discern any direct kinematic evidence. Lastly, results from Figure 5G that show a prevalence of NW-SE–striking normal faults may not be entirely accurate, as the kinematics for many of these mapped faults in the digital compilation have not been documented, and these may in fact represent oblique-slip dextral faults, or, more simply, non-reactivated normal faults from early to middle Miocene NE-directed extension.

Our findings show many similarities to those of Singleton (2015), although our results are ambiguous as to whether inboard strain is the result of extension, transtension, or a combination of both. Observations in the northern part of the study area may better suit an extensional strain field (with exception of the central Cottonwood Basin; Fig. 7E), whereas the appearance of apparent dextral systems (Bill Williams River and Little Black Mountain fault zones, southern Plomosa Mountains) to the south would suggest a transtensional strain field. We must also consider the likelihood that our study gave preference to areas away from localized zones of deformation, as data-collection localities were typically chosen based on the appearance of young (post-detachment) units, and it is likely that much of the strain could be recorded in bedrock units with poor constraints on age of faulting. Therefore, our results may only represent areas that have undergone diffuse strain. Regardless, our results and previous work (Singleton, 2015) show that the dominant component of <12 Ma strain in the LOCO study area was east-west extension.

The largest exposed faults in the study area are the Bill Williams River fault, with up to 200 m of apparent normal separation and 1 km of dextral offset, as well as a NE-striking apparent sinistral fault and a north-striking oblique dextral-normal fault in the northern Plomosa Mountains that accommodate cumulative offset of the Plomosa detachment fault by ≥320 m (Fig. 8A; Strickland et al., 2017). Other important faults are the “Big fault” with 35 m of normal throw, the Little Black Mountain fault with ≤80 m of apparent reverse (?) separation and an unknown amount of dextral slip, Mohave Wash normal faults with ≥10 m normal separation and throw, the Blythe graben with ∼5 m scarp heights, and lastly the Needles deformation zone with ∼5 m scarp heights within the graben and two east-west–striking reverse faults with ≤10 m (?) of reverse throw and ∼30 m structural relief of the monoclinal fold there (Pearthree et al., 2009).

Most post–12 Ma faults that we measured display small-magnitude centimeter-scale to 1–5-m-scale displacements, many of which exhibited normal-sense down-to-the-basin throws. Based on the pattern and diffuse nature of the strain, we infer that one to tens of meters of vertical separation across faults accumulated to several hundred meters to facilitate post-Bouse tectonic subsidence in basin centers. Fault-controlled lowering of the Bouse Formation within basins is viable, given Bouse Formation structure contours in the southern Blythe and Mohave Basins that show ∼500 m vertical difference between basin-margin and basin-axis deposits (Fig. 1; Turak, 2000; Pearthree and House, 2014; Howard et al., 2015). In the southern Mohave Basin, a geomorphic comparison of the angles of sub–Bouse Formation piedmont slopes with those of modern fans is consistent with substantial syn- or post-Bouse vertical deformation in the basin axis (Pearthree et al., 2018). Pearthree and House (2014) suggested as much as 50 m of post-Bouse lowering of the southern Mohave Valley. In the Blythe Basin (Palo Verde Valley), the Bullhead Alluvium is as much as 75 m below sea level (Metzger et al., 1973; Howard et al., 2015). Because the Bullhead Alluvium was deposited after integration of the Colorado River and after development of a quasi-equilibrium profile (Howard et al., 2015; Crow et al., 2019a), this requires >75 m of lowering in that area since 4.5–3.5 Ma. Howard et al. (2015) suggested that compaction of the Bouse Formation was a possible explanation for subsidence; we prefer the interpretation that subsidence is tectonically and/or isostatically driven given that subsequent deposition of overlying modern riverine deposits would only accommodate an estimated ≤25–30 m of compaction (as determined per the method of Angevine et al. [1990]; Data Item S1 [^{footnote 1}]).

We present a model for inboard strain that accounts for both an extensional and a transtensional strain field (Fig. 11). The ambiguity regarding which may be dominant could be a matter of scale (i.e., extension would be prevalent in a transtensional strain field). The change in orientation of inherited structures, as determined by the fault compilation (Fig. S1 [^{footnote 1}]), from north-south striking in the Mohave and Cottonwood Basins to NW-SE striking in the Chemehuevi and Blythe Basins, is likely due to late Miocene to recent east-west extension affecting the entire region (Fig. 11A). East-west extension would reactivate north-south–striking faults as normal faults, while NW-SE–striking faults would reactivate as oblique dextral-normal and dextral faults. This model is consistent with the predominance of north-south–striking normal faults and east-west extension.

Reactivation of older structures and development of new structures in the context of a wide deforming zone inboard of the Pacific–North America plate margin is consistent with inferred transtensional strain in the study area (Fig. 11B) and explains the variety of structures that are expected within a NW-SE–striking dextral shear regime. These are north-south–striking extensional, east-west–striking contractional, NW-SE–striking synthetic dextral, and NE-SW–striking antithetic sinistral structures. In this case, shear coupling may be operating between the San Andreas fault system and the Stateline fault system to the north, or simply between the Pacific–North America plate margin and the more rigid western margin of North America. The transtension model is in agreement with northward advancement of transtensional deformation and with marine conditions filling pull-apart basins in the northern Gulf of California by ca. 6 Ma (Bennett et al., 2015; Umhoefer et al., 2018). Our results suggest that low-strain transtensional deformation has extended north into the lower Colorado River corridor since ca. 12 Ma.

Delineation of inboard zones of dextral shear is essential to understanding post-detachment LOCO region tectonism and, therefore, better characterizing observed deformation and its implications for the Bouse Formation and plate-margin deformation. In the study area, NW-SE–striking zones of dextral shear that appear to postdate Miocene detachment slip have been identified in the Bill Williams River area (Sherrod, 1988), Buckskin-Rawhide Mountains (Singleton, 2015), southern flank of the Whipple Mountains (Dickey et al., 1980; Carr, 1993), southern Plomosa Mountains (Miller and McKee, 1971), and the Chocolate Mountains (Beard et al., 2016). In addition, several zones of dextral and sinistral shear have been documented and/or suggested in this study. These include the Bill Williams River and Little Black Mountain fault zones, Needles deformation zone (suggested), Blythe graben (suggested), northern Plomosa Mountains, eastern Mesquite Mountains, and on the southern flank of the Whipple Mountains. In the southern Whipple Mountains, a possible NW-SE–striking positive flower structure was observed in this study, corroborating the potential for a dextral fault there. Our work also suggests two zones of dextral deformation, the Bill Williams River and Little Black Mountain fault zones, which are along strike of the post-detachment dextral systems in the Buckskin-Rawhide Mountains (Singleton, 2015). The Bill Williams River fault zone may be an along-strike continuation of these systems, while the Little Black Mountain fault zone could represent a splay of these systems in its southeastern segment that transitioned into a north-south–striking normal fault at its northwestern end.

Our observations also suggest that some zones of dextral shear are connected by approximately north-south–striking zones of normal faulting (northwestern Needles deformation zone, Mohave Wash, southern Whipple Mountains), where NW-SE–striking zones of likely dextral deformation transition to north-south–striking zones of extension as well as WNW-ESE–directed extension in possible transtensional stepovers (northern Plomosa Mountains, Cottonwood Basin–Davis Dam) that may be controlled by bounding NW-SE–striking dextral systems. In the northern Plomosa Mountains, WNW-ESE extension may be accommodated between the Buckskin-Rawhide dextral system to the north and a NW-SE–striking dextral system in the southern Plomosa Mountains and/or Kofa Mountains south of the Plomosa Mountains (Miller and McKee, 1971; Wyatt et al., 2017; Strickland et al., 2018b; this study). The northerly trending La Posa Plain west of the northern Plomosa Mountains metamorphic core complex may have developed as a transtensional pull-apart basin kinematically linked to these dextral systems. A drill hole in the northern La Posa Plain (Fig. 8A; La Posa Federal #1, T7N R19W sec. 24, La Paz County, Arizona) suggests that the Plomosa detachment fault is present at a depth of −450 m elevation (below sea level), which is ∼780 m lower than the exposed detachment fault at the northern end of the core complex 10 km to the east. Some of this elevation difference can be attributed to the corrugated nature of the Plomosa detachment fault, but the apparent north-south truncation of the western margin of the mylonitic footwall suggests that the detachment fault has been downdropped along a concealed west-dipping normal fault (Fig. 8A).

Bounding faults within the Cottonwood Basin have been comprehensively mapped (e.g., Faulds, 1995, 1996; Faulds et al., 1995, 2003; Hinz et al., 2012), though which of these faults may accommodate latest Miocene to Quaternary deformation remains unclear. The presence of a localized WNW-ESE extensional strain field in two distinct areas of the LOCO region (northern Plomosa Mountains and Cottonwood Basin domains) could suggest a transtensional stepover mechanism controlled by NW-SE–striking structures. Overall, these observations are compatible with inboard zones of strike-slip deformation and a diffuse Pacific–North America plate boundary, as well as minor to significant transtensional deformation inboard of the San Andreas plate-margin fault and Eastern California shear zone that clearly postdates Miocene detachment faulting.

Southeastward continuation of dextral strain associated with the Stateline fault system into the LOCO corridor has been considered (Singleton, 2015; Bennett et al., 2016; Thacker et al., 2017; Singleton et al., 2019), and is enticing in light of observations of a NW-SE–striking zone of diffuse deformation in the Mount Manchester area and the presence of the Needles deformation zone further along strike to the southeast. All or most of the Stateline fault system strain appears to partition into the SSW-striking oblique sinistral-normal Nipton fault at the western foot of the New York Mountains (Fig. 1; Mahan et al., 2009). Our results from Mount Manchester suggest east-west extension and a north-south–striking normal fault plane solution with slight oblique slip, though possible strike-slip conjugates were observed (consistent with NW-SE–striking dextral shear; e.g., Fig. 11B). Age of deformation at Mount Manchester is difficult to constrain, and at present can only be ascribed to younger than the age of the pluton (≤17–15 Ma). However, accounting for 40°–50° of westward tilting (Walker et al., 2007) shows unlikely paleostrain results of NNE-striking low-angle thrusting (approximately planar) when rotated back to horizontal (Fig. S4 [^{footnote 1}]). Given the unlikelihood of low-angle thrusting synchronous with detachment within a broad area of the pluton (Fig. S2) between 16 and 13 Ma (approximate age of detachment faulting there), we deem it likely that data collected in the Mount Manchester domain are representative of the post-detachment strain history (i.e., <12 Ma). Therefore, we speculate that some strain may have continued past the southeastern terminus of the Stateline fault in the form of an en echelon system of normal faults within an overall dextral zone of shear, similar to that in localized parts of the Walker Lane (e.g., Wesnousky et al., 2012).

Lastly, diffuse dextral shear inboard from the Pacific–North America plate margin can explain the 30–35 km dextral shear discrepancy in the LOCO region (Fig. 2; Bennett et al., 2016). We present in Figure 12 a schematic map of possible deformation zones that could have taken up highly diffuse inboard strike-slip deformation, as compiled from published sources and our new data and interpretations. These areas warrant further investigation to refine or refute this proposed inboard system of faulting.

We conclude that strain continued after 12 Ma within a broad diffuse deformation zone inboard of the San Andreas plate-margin fault, variably expressed as east-west extension on north-south–striking faults and by dextral shear on NW-SE–striking faults. Modern geodetic strains show extension (Kreemer et al., 2010), but these may reflect the extensional incremental strains generated at 45° from overall plate-driven shear. Kreemer et al. (2010, their figure 1) showed extensional strain rates of 4–8 × 10⁻ⁿ per year in the LOCO corridor that equates to extension rates of 1.1–1.4 mm/yr relative to a Colorado Plateau reference frame. Extrapolating these rates back in time for 10 m.y. results in accumulation of extensional strain of 0.04–0.08, resulting in 11–14 km of extension across the zone since 10 Ma. During simple shear, extensional strain for directions at 40°–60° from the shear zone accumulates at about half the shear strain rate (Ramsay and Huber, 1983) such that these extensions would be compatible with shear strain of 0.08–0.16. If we assume that these average shear strains acted across a ∼350-km-wide homogeneously deforming dextral shear zone (LOCO) between the Colorado Plateau and San Andreas–Algodones fault systems, they would have accumulated to 14–28 km total dextral shear since 5 Ma (approximate age of Bouse Formation deposition) and 28–56 km since 10 Ma².

Such an extrapolation is oversimplified for numerous reasons, including the assumption of homogeneous simple shear and steady strain rates through time, a need for more rigorous conversion of GPS velocities to full strain tensors, consideration of strain partitioning between extension and strike slip, and incorporation of part of the eastern Transverse Ranges domain into the calculated area. Nevertheless, given our documentation of deformation over a wide zone that postdates large-magnitude extension, we note that extrapolating modern rates back in time is sufficient to produce significant shear strain (28–56 km displacement) that could account for all or part the proposed 30–35 km of missing dextral shear (Bennett et al., 2016). We highlight, then, that more dextral systems within the LOCO region remain undocumented. Many faults and lineaments previously suggested to be Quaternary in age (e.g., Carr, 1993, their table 5) were not visited in this study and are viable candidates for further research.

Given our results, it is likely that localized deformation along systems of dextral shear progressively partitioned into a zone of diffuse deformation prior to, during, and after deposition of the ca. 6–4.8 Ma Bouse Formation. An east-west extensional strain field persisted, and considerable deformation (though relatively minor in comparison to the Walker Lane, Eastern California shear zone, and San Andreas fault) has been documented in this study. Perhaps most compelling is that known and postulated areas of major subsidence of the southern Mohave and Blythe Basins (Metzger and Loeltz, 1973; Metzger et al., 1973; Turak, 2000; Howard et al., 2015; Pearthree and House, 2014; Pearthree et al., 2016, 2018; Cassidy et al., 2018; Crow et al., 2019a) are coincident with or near zones of substantial deformation (Needles deformation zone, Blythe graben, “Big fault”; structure contours in Figs. 1 and 12). This implies that a tectonic component for thick and deeply buried accumulations of Bouse Formation and Bullhead Alluvium is likely, and that depositional elevations for Bouse Formation outcrops have been modified by post-Bouse deformation, especially in basin centers.

Thick and localized accumulations of Bouse Formation and Bullhead Alluvium in the subsurface are best explained by persistence of strain in the LOCO corridor before and throughout deposition of these units (Metzger, 1968; Metzger and Loeltz, 1973; Howard et al., 2015; Cassidy et al., 2018; Crow et al., 2019a). Strain, as characterized herein, likely resulted in north-south–striking basin deepening by tectonic subsidence. Given the magnitude of deformation, a considerable amount of the intra-basin vertical distribution (scale of tens of meters) of Bouse Formation deposits in the Mohave and Blythe Basins can be explained with a tectonic component. Syn-sedimentary draping, observed at the outcrop scale and inferred based on regional analysis (Pearthree and House, 2014), suggests a combination of mechanisms resulted in the present-day distribution of Bouse Formation outcrop elevations. Assuming a constant strain rate, the shorter time span of syn-Bouse deformation (6–4.8 Ma) would suggest that most of this strain accumulated after deposition of the Bouse Formation.

We interpret the subsurface thickness variation of the Bouse Formation in the Mohave and Blythe Basins in terms of graben-like basins related to distributed faulting and amplified by isostatic subsidence due to sediment loading (Karlstrom et al., 2017). These areas are consistent with zones of deformation (Fig. 12), and areas away from these zones (Chemehuevi and northern Cottonwood Basins) exhibit the least amount of structure contour relief (Fig. 1). We suggest, then, that it is imperative that retro-deformation be taken into account when considering models that bear upon the depositional framework of the Bouse Formation and the integration history of the lower Colorado River corridor.

Diffuse low-strain deformation within the LOCO region continued after large-magnitude Miocene extension had ceased (ca. 12 Ma). Deformation occurred before (>6 Ma), during (6–4.8 Ma), and after (<4.8 Ma) deposition of the Bouse Formation, and may be ongoing as suggested by Quaternary deformation (Needles deformation zone and Blythe graben) and modern geodetics (e.g., Kreemer et al., 2010). Post–12 Ma deformation is typified by characteristic fault geometries and kinematics: north-south–striking normal faults, NW-SE–striking oblique dextral faults, NE-SW–striking oblique sinistral faults, and east-west–striking contractional structures. In light of these results, we propose that the Needles deformation zone and Blythe graben are dextral features.

Pre-, syn-, and post-Bouse deformation is low strain throughout the region and typically at scales of centimeters to meters, though deformation at scales of meters to tens of meters is evident (e.g., “Big fault”). Locally, 100-m-scale deformation is observed (e.g., the northern Plomosa Mountains and the Bill Williams River fault zone). Paleostrain analyses suggest dominant east-west extension throughout the region. We interpret strain to be the result of dextral transtension related to the Pacific–North America plate boundary that forms a wide distributed zone of inboard deformation. Furthermore, we have proposed inboard systems of <12 Ma transtensional deformation (Fig. 12). Transtensional stepovers are characterized by WNW-ESE extension, and in some areas north-south–striking zones of extension connect with NW-SE–striking dextral zones. Dextral strain may extend along strike of the Stateline fault system past its southeastern terminus into the LOCO region via an en echelon system of normal faults. Lastly, geodetically derived modern strain rates are adequate to account for a 30–35 km discrepancy in dextral shear within the LOCO region as compared to surrounding domains when extrapolated to 10 Ma.

Post–12 Ma faulting likely amplified basin accommodation within Bouse Formation depocenters at the scale of meters to tens of meters (namely in the southern Mohave and Blythe Basins). Continuation of an east-west extensional strain field and accumulated strain within LOCO basins suggest that some amount of the vertical distribution between highest-elevation basin-margin and lowest-elevation basin-axis Bouse Formation deposits (a maximum of ∼500 m) is tectonically controlled. Post-Bouse tectonic adjustments suggest that Bouse Formation deposits have not been static since deposition, therefore requiring retro-deformation in order to properly account for their depositional framework and the integration history of the lower Colorado River.

We wish to thank Phil Pearthree, Keith Howard, Kyle House, Becky Dorsey, Brian Gootee, Jason Ricketts, Anna Buising, and Brandon Schmandt for their insight, reviews of previous abstracts and papers, and interesting conversation. Kris McDougall, Tracey Felger, and Debra Block were of great help with logistical support in the office and field. Katie Holleron was invaluable for her assistance in the field. Wanda Taylor, Jim Faulds, Phil Pearthree, and the editors (Andrea Hampel and Todd LaMaskin) provided thorough reviews and constructive suggestions that enhanced this manuscript greatly.

Material in this publication is based upon work supported by the National Science Foundation (NSF grant number EAR-1545986) and an internship provided through the Graduate Student Preparedness Program (GSP) with the U.S. Geological Survey (USGS) in Flagstaff, Arizona. Any findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. Coordination of GSP at USGS was through the Youth and Education in Science programs within the Office of Science Quality and Integrity. We especially wish to acknowledge Lina Patino and Eleanour Snow for their help with the application process and support during the internship. Fieldwork in the northern Plomosa Mountains was supported by USGS EDMAP grant G16AC00142 to Singleton.