The 12–6 Ma Hualapai Limestone was deposited in a series of basins that lie in the path of the Colorado River directly west of the Colorado Plateau and has been deformed by an en-echelon normal fault pair (Wheeler and Lost Basin Range faults). Therefore, this rock unit represents an opportunity to study the sedimentological and structural setting over which the Colorado River first flowed after integration through western Grand Canyon and Lake Mead. In this study, we quantify the structural geometry of the Hualapai Limestone and separate the deformation into syn- and postdepositional episodes. Both the Wheeler and Lost Basin Range faults were active during Hualapai Limestone deposition, as shown by thickening of strata and fanning of time lines toward half-graben faults that bound the Hualapai subbasins. The structure is characterized by a prominent reverse-drag fold and broad, shallow syncline adjacent to the Lost Basin Range fault, and a small-magnitude reverse-drag fold and short-wavelength normal-drag fold adjacent to the Wheeler fault. We find ∼450 m of throw between the footwall and hanging-wall Hualapai Limestone sections, suggesting faulting was ongoing after Hualapai Limestone deposition ceased and during Colorado River incision. To investigate a range of possible fault geometries that may have been responsible for Hualapai Limestone deformation, we compared our structural results against surface deflections calculated by a two-dimensional (2-D) geomechanical model. While nonunique, our results are consistent with a scenario in which the Wheeler fault was surface rupturing, or nearly surface rupturing throughout deposition of the Hualapai Limestone, but was inundated at ca. 6 Ma by coalescing paleolakes in Gregg and Grand Wash Basins as sedimentation kept pace with deformation. In contrast, we find evidence suggesting the Lost Basin Range fault was deeply buried by the Hualapai Limestone and likely propagated upward and laterally to break the surface sometime after 6 Ma. Therefore, we interpret the landscape over which the Colorado River first flowed to be of low relief within the terrain bounded by the Grand Wash Cliffs, the Hiller Mountains, and subtle topographic highs to the north and south of our field area. This original low-relief depositional surface was deflected into the structure exposed today by continuing deformation by the Wheeler and Lost Basin Range faults, allowing for calculation of apparent incision rates of the modern Colorado River drainage system that spatially vary between 33 and 42 m/m.y. in the hanging wall and between 108 and 115 m/m.y. in the footwall. Hanging-wall incision rate values are similar to, but faster than, a previously published point measurement, and footwall values are similar to measured incision rates in the western Grand Canyon, suggesting the Wheeler fault system may resolve as much as ∼410 m of Colorado Plateau uplift in the last 6 m.y.
The timing and tempo of uplift of the Colorado Plateau and incision of Grand Canyon have fascinated scientists for decades. Much of this uplift may be resolvable onto finite faults that break the surface of Earth; these faults, if active during river incision, are expected to impart patterns of differential incision that locally modulate the rate of incision into bedrock (Pederson et al., 2002; Karlstrom et al., 2007). Therefore, quantifying the reach-scale structural deformation over which the Colorado River flowed may provide spatial context that complements single-point measurements of incision. At the mouth of Grand Canyon in eastern Lake Mead, the 12–6 Ma Hualapai Limestone (Longwell, 1936; Bohannon, 1984; Crossey et al., 2015) records the sedimentological and structural setting immediately prior to the arrival of the modern Colorado River and may be used to understand both the nature of the pre–Colorado River drainage in this region and the long-term differential incision rates of the Colorado River as it carved the modern canyon through eastern Lake Mead. Furthermore, the tectonic setting of the Hualapai Limestone within the transition between the Basin and Range and the Colorado Plateau, clear exposures, and temporal constraints make this unit an excellent candidate to study normal fault processes and basin development.
For example, extensional “drag” (concave-upward) and “reverse-drag” (concave-downward) folds are commonly thought to form due to frictional drag during normal fault slip and rollover of hanging-wall stratigraphy due to the space problem created in the hanging wall of a curved fault, respectively. Both types of extension-forced folds have expected geometries and assumed formational processes based on widely cited research (e.g., Hamblin, 1965; Poblet and Bulnes, 2005). For example, a concave-upward (listric) fault geometry has long been assumed to be a necessary and sufficient condition to form reverse-drag folds in the hanging wall of normal faults (Hamblin, 1965). However, there is a growing body of evidence that suggests reverse drag may be caused by slip on planar or antilistric faults as well (Reches and Eidelman, 1995; Grasemann et al., 2005; Resor, 2008), and footwall fold structure may be more diagnostic of specific fault geometries (Resor and Pollard, 2012); similarly, normal-drag folds are now thought to result from fault propagation below and past the deforming strata (Willsey et al., 2002; White and Crider, 2006). However, the full range of possible fold geometries associated with normal faulting is not clear, nor is the specific way in which fault geometry may be responsible for folds visible in outcrop.
The Hualapai Limestone is folded into a reverse-drag monocline and nearby syncline and offset variably along strike by ∼300–500 m by the Wheeler fault system, suggesting this unit may be used to study normal fault and fold processes. Furthermore, at least as far back as Lucchitta (1979), workers have recognized that the Hualapai Limestone may be used to help constrain the magnitude and timing of Colorado Plateau uplift due to its spatial and temporal setting and clear structural relationships. More recently, the incision of the Colorado River has become the focus for studies attempting to discern between models of Colorado Plateau uplift (Pederson et al., 2002; Karlstrom et al., 2007, 2008; Crow et al., 2014). Incision rates on the Colorado Plateau are lower in hanging walls of major W-dipping normal faults relative to footwall blocks, and, within hanging-wall blocks, incision rates are lowest adjacent to faults and increase at distances of ∼10 km due to hanging-wall flexure (Pederson et al., 2002; Karlstrom et al., 2007; Crow et al., 2014). This indicates, not surprisingly, that normal fault deformation in the western Grand Canyon and Lake Mead regions modulates incision of the Colorado River, suggesting that rates of denudation at least partly reflect local fault and fold processes and must be placed in an appropriate structural context (Pederson and Karlstrom, 2001; Willis and Biek, 2001; Karlstrom et al., 2007, 2008; Crow et al., 2014). Similar patterns in incision rates may be found at much larger scales. For example, incision rates of rivers draining the Colorado Plateau vary from east to west (Darling et al., 2012; Pederson and Tressler, 2012). Incision rates peak in the eastern Grand Canyon, while incision in the western Grand Canyon occurs more slowly (Karlstrom et al., 2007). Farther east, in the central Colorado Plateau, major tributaries of the Colorado River appear to incise at lower rates than those for the western Grand Canyon (Wolkowinsky and Granger, 2004), although not as low as those in the lower Colorado River (Karlstrom et al., 2007) to the west and south of the Colorado Plateau. These larger patterns of incision have been linked to mantle buoyancy through studies of basaltic volcanism and seismic structure beneath the plateau (Crow et al., 2014), further indicating the importance of understanding the potentially complex interplay among large-scale tectonic drivers of uplift, local fault processes, and river incision. Thus, quantifying the deformation of the Hualapai Limestone has the potential to contribute to understanding basin evolution and normal fault and fold processes, provide information on Colorado Plateau uplift, and lend insight into the way in which the Colorado River has incised through evolving structures.
In this paper, we examine the structural setting over which the Colorado River first flowed after integration through western Grand Canyon and into the Lake Mead region at 5–6 Ma. The field area lies directly west of the Grand Wash Cliffs (Fig. 1) and therefore is minimally affected by dynamic topography of the Colorado Plateau. We claim that the Colorado River, directly postintegration, flowed over low-relief, sediment-filled structural basins in eastern Lake Mead; postintegration, the surface was faulted and warped into a reverse-drag monocline and hanging-wall syncline and continued to accumulate displacement as the river incised over the last 6 m.y. Furthermore, these along-strike variations in fold geometry likely required complex patterns in apparent incision rate as the river flowed across the active structures. To substantiate our claim, we construct a three-dimensional surface model and structural contour map of an idealized horizon within the upper Hualapai Limestone, a unit uniquely situated in time and space to provide clues to the geomorphic response of the Colorado River to fault-related deformation. Further, we explore physically meaningful estimates of fault geometry that may be responsible for creating the observed surface deflections using a forward-sense two-dimensional (2-D) geomechanical model, thereby investigating a range of possible fault geometries at depth. We conclude by estimating apparent incision rates of the Colorado River across the active structure over the last 6 m.y.
In contrast to the relative stability of the Colorado Plateau interior, the Lake Mead region of the central Basin and Range exhibits excellent exposures of complex extensional structures and volcanic features, and it has been used extensively to study processes of crustal deformation (e.g., Lucchitta, 1979; Duebendorfer and Simpson, 1994; Anderson et al., 1994; Beard, 1996; Brady et al., 2000; Fitzgerald et al., 2009; Umhoefer et al., 2010). This study focuses on the area directly west of the Grand Wash fault, the structural boundary between the Colorado Plateau and the Basin and Range Province (Fig. 1). Accordingly, the study area presents a high strain gradient from the relatively unextended Colorado Plateau to the highly deformed Basin and Range.
The eastern Lake Mead region consists of a series of W-dipping normal faults and associated half grabens that are responsible for ∼15 km of Miocene extension (Brady et al., 2000). The largest of these normal faults, the Grand Wash fault, is thought to have accommodated greater than 3.5 km of normal offset (Brady et al., 2000) along a concave-up geometry (Faulds et al., 1997). Flat-lying Paleozoic strata are exposed in the footwall Grand Wash Cliffs, and the Grand Wash Trough is the N-NE–trending structural depression formed in the hanging wall of the Grand Wash fault. The Grand Wash fault is buried beneath late Miocene basin fill in the Grand Wash Trough. Extension related to the Grand Wash fault was most active from 16.5 to 13 Ma (Faulds et al., 2001a), and deformation waned or ceased by 11–8 Ma (Faulds et al., 2001a).
Our study area (Fig. 2), just west of the mouth of the Grand Canyon, encompasses an area extending from Sandy Point in the north to Hualapai Wash in the south and from Temple Mesa in the west to Grapevine Mesa in the east. The Wheeler fault system is the next ridge-bounding normal fault west of the Grand Wash fault and trends roughly N-S through the center of our study area. Longwell (1936) measured a 60°W-dipping polished surface along the fault trace south of the Colorado River. Wallace et al. (2005) also assigned a moderate westward dip to the Wheeler fault (55°) based on three-point calculations using their mapped fault trace. The Wheeler fault system extends northward from the map area for >50 km, where it merges with the Grand Wash fault (Umhoefer et al., 2010; Brady et al., 2000). Gregg Basin is the half graben formed in the hanging wall of the Wheeler fault south of South Cove (Fig. 2). The Lost Basin Range fault, a westward-stepping, en-echelon southern continuation of the Wheeler fault (Wallace et al., 2005; Fitzgerald et al., 2009; Umhoefer et al., 2010), divides the footwall Lost Basin Range from the hanging-wall Gold Basin (Umhoefer et al., 2010; Swaney et al., 2010). For simplicity, we will hereafter refer to the hanging wall of both the Lost Basin Range and Wheeler faults as Gregg Basin. The relay zone between the two segments is characterized by a northward-dipping ramp (Umhoefer et al., 2010; Wallace et al., 2005). The E-dipping Meadview fault (Fig. 2) is poorly exposed on the eastern flank of the Lost Basin Range. This fault is thought to have accommodated mid-Miocene extension in the area, with a cessation of uplift at ca. 14 Ma (Umhoefer et al., 2010). Therefore, the Meadview fault is one of the oldest in the field area (Swaney et al., 2010).
Longwell (1936) deduced >1.5 km offset for the Wheeler fault where it crosses the Colorado River based on offset of Paleozoic strata. Moreover, the Wheeler fault has a minimum of 275 m of post–late Miocene, down-to-the-west offset, as indicated by the elevations of the uppermost surfaces of the Hualapai Limestone in the footwall and hanging-wall blocks (Wallace et al., 2005). Based on a similar line of reasoning, Fitzgerald et al. (2009) assigned 500 m of slip to the Lost Basin Range fault, at least some of which must have been post–late Miocene. Apatite fission-track thermochronology integrated with structural and stratigraphic observations indicates extension along the Wheeler fault system was active by ca. 15.3 Ma (Fitzgerald et al., 2009; Umhoefer et al., 2010). Growth strata in the 12–6 Ma Hualapai Limestone (see later discussion) and underlying conglomerate (Wallace et al., 2005; Umhoefer et al., 2010) indicate continued faulting through 6 Ma, while offset Hualapai Limestone strata and minor tilting point to a possible lower rate of deformation after 6 Ma (Umhoefer et al., 2010). Quaternary slip is indicated by alluvium displaced 10–15 m across the Wheeler fault trace north of the Colorado River, although the scarp is gentle, and activity is not thought to be recent (Pearthree, 1997).
The Hualapai Limestone (Fig. 2) was deposited in a series of lakes that filled extensional half grabens westward from the Grand Wash Trough to Temple Bar (Lucchitta, 1979; Bohannon, 1984; Faulds et al., 2001b). The limestone has been variably interpreted as marine, lacustrine, and brackish in origin. If marine (Blair and Armstrong, 1979), the limestone provides evidence for as much as 880 m of uplift since ca. 6 Ma (Lucchitta, 1979). However, Faulds et al. (1997) and subsequent workers have hypothesized that the Hualapai Limestone is lacustrine in origin, fed by waters debauched from springs at the base of the Colorado Plateau. Early researchers noticed the importance of the end of Hualapai Limestone deposition as heralding the arrival of Colorado River waters; original radiometric dates for the cessation of limestone deposition and thus this important hydrological event were 5.97 ± 0.07 Ma (Spencer et al., 2001). The commencement of Hualapai Limestone deposition was found to be 11 Ma (Faulds et al., 2001b). More recently, radiometric dating of an ash-fall tuff indicates initial deposition of the Hualapai Limestone began as early as 12 Ma (Crossey et al., 2015), while the roughly 6 Ma end of limestone deposition in the western (Temple) basin is still thought to be robust. The limestone is the uppermost member of a series of sedimentary rocks that fill the Grand Wash Trough. These informally named rocks of the Grand Wash Trough (Bohannon, 1984) consist of various evaporite, calcareous sandstone, and conglomerate deposits. Geological and topographic relations suggest that at least the Pierce Ferry Hualapai limestone section was deposited in a closed basin at the time of deposition and received no exogenous sediment (Longwell, 1936; Lucchitta, 1979), a result more recently extended by geochemical evidence to include the Hualapai Limestone in the Grand Wash Trough and Gregg Basin (Crossey et al., 2015). The Hualapai Limestone consists of several facies types, including limited evaporites and reddish siltstones, sandy limestones, the predominant wavy limestone, and travertine (for a thorough discussion of the sedimentology, stratigraphy, and geochemistry, see Crossey et al., 2015). Interfingering of the basal conglomerate and lower Hualapai Limestone and evaporite facies and increasingly robust limestone deposition up section indicate an environment in which ephemeral lakes periodically received clastic input from surrounding highlands, were desiccated regularly, and eventually coalesced to form paleo–Lake Hualapai (Crossey et al., 2015). In fact, the upper horizons within both the Gregg Basin and Grand Wash Trough sections are thought to represent a time when separate basins were connected by a lake highstand in late Miocene time (Crossey et al., 2015). In Gregg Basin, the Hualapai Limestone is greater than 120 m thick, and it achieves thicknesses of 210 m in the Grand Wash Trough to the east (Lopez-Pearce, 2010; this paper, section titled Syndepositional Deformation [ca. 12–6 Ma]).
QUANTIFYING POST-HUALAPAI DEFORMATION (CA. 6 MA–PRESENT)
The uppermost sediments of the Hualapai Limestone consist of planar peloidal wackestone and/or vuggy travertine facies, indicative of shallow lacustrine and marsh deposition, respectively, across the entire study area (Crossey et al., 2015; section titled Syndepositional Deformation [ca. 12–6 Ma] herein). These sediments are inferred to represent a widespread lake highstand between ca. 7 and 6 Ma. The present-day elevation of these beds thus provides a means by which to measure relative vertical deformation across the region over the last ∼6 m.y. In this section, we describe our efforts to characterize faults that cut this horizon as well as the folding of the horizon between major faults.
We employed digital photogrammetric mapping techniques to create a three-dimensional (3-D) surface model, cross sections, and a structure contour map of an idealized layer in the uppermost Hualapai Limestone of Gregg Basin and the Grand Wash Trough. We used a stereo model of aerial photographs created by MDA Geospatial Services. The model was originally constructed to create a series of statewide orthophotos from 2005 imagery and thus meets 1:12,000 National Map Accuracy Standards, implying a horizontal accuracy better than 10.16 m at 90% confidence (http://sco.azland.gov/flight-data/mda-all.htm). We then imported the stereo model into the LPS software package (INTERGRAPH), where several missing aerial images were replaced by scans of the original film images. We used the LPS software for visualizing the stereo model and mapping in 3-D.
Cliff-forming beds in the upper Hualapai Limestone are exposed in tributary washes across the entire study area, which permitted mapping of 3-D vertices every 2 to 3 m near the top surface of the Hualapai Limestone. Cliff bands in the Hualapai Limestone are generally unvegetated and continuous. However, interpretations had to be made across areas of shadow in the photographs. Furthermore, in some regions, the primary bed chosen for mapping was eroded, necessitating a vertical adjustment up or down several layers in order to continue mapping. These adjustments were noted and corrected for by a simple vertical shift of the affected data. We estimate an additional elevation error in our data of ±2 m (the approximate distance between cliff-forming beds) to account for these stratigraphic uncertainties. Although this error may lead to a local upward or downward bias, we believe that it is unlikely that misinterpretations in this regard would accumulate across the map area. Finally, photogrammetric mapping was complemented by field work both to corroborate the remote mapping as well as to better constrain details of the structure, including fault locations and orientations and bedding attitudes (Fig. 2).
We imported the structural data points into GOCAD (Mallet, 1992) and interpolated them to form the structure contour map (idealized horizon, or idealized bed; Fig. 3). The GOCAD software uses a proprietary smooth interpolation algorithm (Mallet, 1989) and is ideal for such a manipulation because it incorporates faults and satisfactorily estimates the deformation between the faults in the fault array of Gregg Basin. Thus, the idealized bed represents a continuous layer within the Hualapai Limestone that is near the top of the section across the entire map area. In places, the idealized bed may rise above the elevation of current exposure due to erosion. Across much of the study area, however, the idealized bed resides 5–10 m below the elevation of the current Hualapai Limestone exposure top. We extracted the structural cross sections (Fig. 4) through the structural surface along lines shown in Figure 3B using the 3D Analyst tool in ArcGIS 10. We then overlaid the structural profiles on topographic data from the National Elevation Data Set 10 m digital elevation model.
The Wheeler and Lost Basin Range faults are the largest-offset faults that cut the upper Hualapai Limestone. Near South Cove, the idealized bed is at an elevation of ∼900 m in the footwall of the Wheeler fault and ∼450 m in the hanging wall, indicating a throw of ∼450 m (Fig. 3; Fig. 4, sections A–A′ and B–B′). The idealized bed reaches its minimum elevation of ∼300 m near the northern end of the Lost Basin Range fault, suggesting that the maximum post–ca. 6 Ma throw on the fault system may be greater than the estimate from near South Cove when fold magnitude variation is taken into account.
Our mapping of these faults is largely consistent with that of previous workers (Wallace et al., 2005; Beard et al., 2007; Umhoefer et al., 2010; Felger and Beard, 2010). All faults appear to have normal-sense offset and trend roughly N-S. The Lost Basin Range and Wheeler faults exhibit a right-stepping relay near the center of the study area that is not breached by any known fault of significant offset. Our mapping reveals additional, previously unrecognized, small-offset fault strands that form a horsetail-type structure near the northward termination of the Lost Basin Range fault (Fig. 2). Another set of small-offset faults is located in the footwall of the Wheeler fault near South Cove (Wallace et al., 2005).
A newly discovered exposure along a strand of the Lost Basin Range fault (Fig. 5A; red star on Fig. 2; UTM coordinates: 759916E, 3990966N, zone 11N, WGS 1984), juxtaposing fanglomerate near the base of Hualapai Limestone and crystalline rock, strikes roughly SSW (202°) and dips 53°W at the surface. This moderate to high dip is in agreement with that calculated for the Wheeler fault using three-point analysis (Wallace et al., 2005). Kinematic striations developed in thin fault gouge on the fault surface rake 76°SW, indicating that at least the most recent movement was dominantly dip slip, but may have included a small component of left-lateral slip as well. Alternatively, the strike-slip component may represent an effect of the northern tip of the fault. Short-wavelength (3–5 m) normal-drag folds (Fig. 5B) along the Lost Basin Range fault provide additional evidence for moderate to steep fault dip. Because drag folds drape and conform to the shape of the fault, they stand as minimum approximations of the dip of the fault. The mean dip of Hualapai Limestone bedding thus warped upward toward the Wheeler fault system is 44°W.
Grand Wash Trough
The upper Hualapai surface is warped into a subtle syncline that trends parallel to the axis of the Grand Wash Trough (Fig. 3). Absolute elevations of the idealized bed range from 995 m (far eastern edge of the map area, below the Grand Wash Cliffs) to 775 m (middle of Grapevine Wash). Structural relief along a section perpendicular to this syncline (Fig. 4, section A–A′) is 130 m measured over a horizontal distance of ∼3000 m with a maximum structural dip of 0.07 m/m (4°). This dip is on the low end of the range of the structural attitude measurements collected in the field nearby in the Grand Wash Trough (Fig. 2), likely due to the fact that the structure contour map does not resolve highly localized structures. The amplitude of this syncline increases southward toward the limit of the Hualapai outcrops. The upper Hualapai Limestone is not cut by any fault associated with this syncline; however, small faults and folds are present in lower Hualapai Limestone strata (Wallace et al., 2005). To the west of this syncline, the idealized horizon is offset by a series of small faults in the immediate footwall of the Wheeler fault. These faults are associated with local folding of the strata (Wallace et al., 2005), but the idealized horizon does not reveal any broader warping associated with proximity to the Wheeler fault. The idealized horizon is not exposed in the immediate footwall of the Lost Basin Range fault.
The elevation and folding of Hualapai Limestone strata in Gregg Basin vary from north to south. Hualapai Limestone strata in the hanging wall of the Wheeler fault (Fig. 4, sections B–B′ and C–C′) are warped down to the east from 450 m to 435 m elevation over a distance of 760 m, forming a subtle reverse-drag fold. The maximum eastward structural dips are 0.13 m/m (7.4°) and 0.005 m/m (<1°) along cross-sections B and C, respectively. This compares well with dips of 2°–12°E measured in the field (Fig. 2). Within ∼50–100 m of the fault, the strata are deflected upward again, creating a classic short-wavelength normal-drag fold.
In the region between the Wheeler and Lost Basin Range faults, the idealized horizon rises slightly to the southeast, with elevations increasing from 546 m to 567 m over a distance of ∼750 m, forming a classic relay ramp with northwest dips of 0.028 m/m (1.6°). The horsetail faults within this region locally deflect Hualapai Limestone strata (Fig. 5C), but their offsets do not significantly impact deformation of the Hualapai Limestone at the scale of the structure contour map. Immediately to the south of the horsetail faults, a transverse fold (Schlische, 1995) with southwest dips up to 28° (Figs. 2, 4, and 5D) is associated with an ∼40 m decrease in elevation of the idealized horizon in the hanging wall of the Lost Basin Range fault.
The structure contour map reveals a broad, shallow syncline near the northern tip of the Lost Basin Range fault that continues south. This is the Gregg Basin syncline of Wallace et al. (2005). Where the idealized bed is best constrained by the data within the Gregg Basin syncline (Fig. 4, section D–D′), the marker horizon rises from 430 m near the syncline axis to 490 m near the Lost Basin Range fault over a distance of 650 m. Along this line, structural dip varies from 0.17 m/m (9.6°) west to 0.072 m/m east (4.1°), in good agreement with structural attitudes of ∼12°W and ∼4°E measured in the area (Fig. 2). The lowest elevation of the idealized bed is found near Hualapai Cove between sections D–D′ and E–E′ (Fig. 4). In section E–E′, the Gregg Basin syncline is still present; however, the dominant structural feature is a westward up-warping of the bedding to the west of the syncline axis. The maximum structural dip westward at E–E′ is 0.034 m/m (1.9°). The maximum eastward dip is 0.068 m/m (3.9°), lower than the measured dips of 4°–8°E. The highest elevations of the marker horizon in Gregg Basin are found in the southwestern corner of the map area. Along section F–F′, the marker horizon drops from an elevation of 660 m at the western end of the section to 525 m over a horizontal distance of 2100 m. Maximum eastward structural dip is 0.15 m/m (8.5°), again lower than but similar to measured structural dips (Fig. 2).
SYNDEPOSITIONAL DEFORMATION (CA. 12–6 MA)
Several lines of evidence suggest that the major structures described in the previous section, which deform the youngest strata in the Grand Wash Trough and Gregg Basin (<6 Ma), were also active during deposition of the Hualapai Limestone within these subbasins (7–12 Ma). In this section, we document the syntectonic nature of Hualapai Limestone strata at three scales: the basin scale, the scale of individual structures (folds), and the outcrop scale.
Methods: Facies Descriptions and Bed Thickness Measurements
At the basin scale, syntectonic deformation is evident in lateral variation in the thickness and facies of Hualapai Limestone strata. These changes are documented in a series of 13 measured sections that extend from the Grand Wash Cliffs in the east to Detrital Wash in the west (Fig. 6). The basal contact of the Hualapai Limestone is transitional with underlying siliciclastic units and is defined in this study as the first appearance of a limestone bed >0.5 m in thickness. The Hualapai Limestone section is subdivided into eight facies listed here in order of generally increasing water depth: (1) a sandy limestone facies, interpreted as a fluvial deposit, consisting of calcite-cemented sand with local carbonate clasts deposited in massive beds that locally contain trough cross-beds; (2) a red siltstone facies and (3) an evaporate facies consisting primarily of gypsum beds; facies 2 and 3 interfinger and are interpreted as evidence of fluvial and playa lake environments; (4) a vuggy travertine facies, interpreted as a vegetated marsh deposit; (5) a wavy laminated limestone facies with abundant evidence of biological activity including stromatolitic and biohermal mats of likely microbial or algal origin, interpreted as a shallow lacustrine deposit and locally containing limestone breccias consistent with wave activity; (6) a banded sparite facies, interpreted as a spring deposit that is found within and between beds of facies 4 and 5; (7) a planar peloidal wackestone facies, indicative of a low-energy lacustrine environment; and (8) ash-fall tuff beds that may be found interlayered with any of the other facies and that provide time lines in the section. In several of the sections, fanning of these time lines also provides evidence for thickening of strata toward active faults that bound the east side of subbasins.
At the scale of individual folds, the syntectonic nature of the Hualapai Limestone is evident in the thickening and thinning of stratigraphic packages (e.g., bed sets) across these structures (Fig. 7). In order to quantify changes in stratal thickness across the Gregg Basin syncline and the monocline of Hualapai Wash, we constructed photographic panoramas of the upper ∼40–60 m of Hualapai Limestone section exposed in crosscutting canyon walls. At select points along each photo traverse location (Fig. 2), we made thickness measurements in the field of two to three bed sets or series of bed sets with a laser range finder. Each bed set’s thickness was measured a minimum of seven times to find the average bed-set thickness at each location and the precision of the measurement (estimated as twice the standard deviation).
Finally, observations of outcrop-scale soft sediment deformation features provide further corroborating evidence for the syntectonic nature of the Hualapai Limestone (Fig. 8). Tepee structures and load features up to 1.5 m high are common, particularly where interbedded siltstone is abundant, such as the lower section of the Grand Wash Trough and in Gregg Basin. We interpret these structures as evidence of tectonically activated dewatering and fluid escape in an environment of high groundwater pressure and gradient (Assereto and Kendall, 1977; Kendall and Warren, 1987; Ferguson et al., 1982; Wheeler, 2002; Onasch and Kahle, 2002; Wallace, 1999). Wavy and contorted bedding is also present in the lower part of the section in the Grand Wash Trough and appears to be more common near small intradeposit faults. We interpret these features as seismites (Onasch and Kahle, 2002; Wheeler, 2002), again suggesting syntectonic deposition.
The Grand Wash Trough contains a wedge of Hualapai Limestone that thickens from the Wheeler fault trace to the Grapevine Wash section. A composite section extending from the base of the wash up Grapevine Canyon has the thickest accumulation, with 212 m of Hualapai Limestone. This section is dominated by beds of the planar peloidal, wavy laminated, and vuggy travertine limestone facies consistent with a shallow lacustrine environment. Thin (1–2 m) deposits of red siltstone layers suggest intermittent input of clastic material during discreet lake lowstands (Lopez-Pearce, 2010). Banded sparite is fairly common throughout the section. Three sections across Airport Point (Fig. 2) show progressively thinner Hualapai Limestone sections measuring 86, 44, and 21 m from east to west. Interbedded sandy limestone beds appear in the easternmost of these sections, and both the eastern and middle sections contain thicker (up to 13 m) and more abundant siltstone layers than the Grapevine section. The westernmost sections, located at west Grand Wash Trough and North Tower, are 10 and 11 m thick, respectively, and dominated by the planar peloidal wackestone with minor interbedded siltstone. All sections are capped by planar peloidal wackestone or vuggy travertine facies. The lateral facies changes in the Grand Wash Trough sections suggest that a long-standing, spring-fed marsh/lake system existed at least as far east as Grapevine Wash, while the more western portions of the basin contained ephemeral lakes until the deposition of the youngest Hualapai Limestone sediments.
In Gregg Basin, the easternmost two sections, at Smith Bay and Hualapai Wash, are both incomplete, indicating a total Hualapai Limestone thickness >97 m and >56 m in proximity to the Wheeler and Lost Basin Range faults, respectively. Both sections contain abundant red siltstone interlayered with thin to medium beds of the planar peliodal wackestone and vuggy travertine limestone facies. The complete Hualapai section was measured in two locations farther west, a 120-m-thick composite section east of Spring Wash and an 88-m-thick section north of Little Burro Bay. At these locations, the Hualapai Limestone also contains abundant red siltstone interlayered with medium beds of planar peliodal wackestone, vuggy travertine, and wavy laminated limestone facies. All sections in Gregg Basin are once again capped by planar peloidal wackestone or vuggy travertine facies. A 12.5-m-thick section on the eastern edge of the Temple Basin is dominated by vuggy limestone with lesser interlayered siltstone and planar peloidal wackestone and may represent the spillover between Gregg Basin and Temple Basin.
In Gregg Basin, we collected two sets of panoramic measurements to quantify thickness changes across the major structures: the Gregg Basin syncline and the monocline of Hualapai Wash. In the Gregg Basin syncline section, two stacked bed sets were surveyed, each capped by prominent limestone ledges. The bed sets thicken by 30% and 31% over a distance of ∼40 m from the eastern limb of the syncline into the synclinal axis. In the section from the monocline of Hualapai Wash, the canyon wall is cut by a fault of indeterminate offset, but strata on each side of the fault reveal eastward thickening. To the east of the small fault, the upper two bed sets capped by prominent limestone ledges thicken ∼30% over a distance of 110 m, while a lower bed set thickens ∼9%. To the west of the small fault, an upper bed set thickens ∼28%, while a lower bed set thickens 13%. Both sections thus reveal significant growth of the upper Hualapai section clearly associated with the monocline. Such thickness variations are interpreted to represent differential compaction and broad folding of lacustrine and marsh facies due to mild syndepositional tectonism. We observed features similar to progressive unconformities in the footwall of the Wheeler fault (Fig. 8D); we did not, however, attempt to quantify thickness changes in the footwall.
MODELING THE DEFORMATION
As structural mapping demonstrates, fault-parallel (longitudinal) folds adjacent to the Wheeler fault are significantly different in amplitude and geometry than those adjacent to the Lost Basin Range fault. Longitudinal folds in extensional settings are generally inferred to form in response to subsurface fault geometry (Schlische, 1995). In this section, we use numerical models to explore how changes in a few basic fault parameters (lower tip depth, upper tip depth, and slip magnitude) might lead to the observed differences in structure between these two faults. Footwall stratigraphy is not exposed over most of the map area, ruling out a formal inversion for fault geometry (Resor and Pollard, 2012). Instead, we take an illustrative approach. The results are nonunique, but they provide plausible explanations for the observed patterns of deformation.
We employed a linear elastic 2-D boundary element model based on the code TwoDD by Crouch and Starfield (1983). Linear elasticity theory has enjoyed success over a range of problems, including fault-related folding (Gupta and Scholz, 1998; Resor, 2008), and, more recently, it has been useful in examining fault propagation (Martel and Langley, 2006). Our code has been translated into MATLAB (Martel and Langley, 2006) and modified to accommodate displacement discontinuity (relative displacements across the fault) boundary conditions (Resor and Pollard, 2012). We drive the models by applying constant slip across the dipping portions of the fault. Detachments, where present, are modeled as horizontal surfaces that are free to slip and end at a vertical free surface 10 km from the surface trace of the fault to allow uninhibited extension. Elastic properties for the surrounding rock volume (Poisson’s ratio of 0.25 and Young’s modulus of 48 GPa) are based on average values for limestone (Pollard and Fletcher, 2005).
A first set of experiments was designed to find the best-fitting lower tip depth and to compare isolated planar faults with detached faults, approximating listric geometry (Fig. 9A). For these models, we superimposed data from cross-section A (footwall) onto cross-section F, which represents our most laterally continuous hanging-wall section, in order to estimate a plausible lower tip depth. We iterated through fault lengths varying between 1000 m and 10,000 m in increments of 1000 m and slip magnitudes ranging from 450 m to 900 m in increments of 50 m, and we used a simple grid search algorithm to minimize root mean square error (RMSE) between the model and the data, thereby solving for a best-fitting fault length and slip magnitude for a surface-breaking fault. This process was repeated for isolated and detached geometries. Next, we compared simulated surface deflections to structural cross-sections B and D, representing the Wheeler fault and the Lost Basin Range fault, respectively, using the best-fit fault length (4000 m) and geometry (detached) from the previous experiment (Figs. 9B and 9C) and iterating over a range of fault slips (450–900 by 50 m increments). We chose these cross sections because they are of similar lengths, are equally close to the lateral tip line of their respective fault traces, and demonstrate interesting variation in fold geometry. Finally, we explored the effect of blind slip through the superposition of the surface-breaking fault model used in the previous experiments with a model where the fault tip ended at some distance below the surface. Models were optimized by searching over a range of values of upper tip depth (0–1200 m) for the buried fault and slip for both faults (0–900 m by 50 m increments).
The first experiments reveal that deformation associated with the Wheeler and Lost Basin Range faults is best explained by models with faults that extend to a relatively shallow depth of 3.3 km (detached; Fig. 9A, model 2) to 4.1 km (isolated; Fig. 9A, model 1). The isolated fault model significantly overpredicts footwall flexure and thus fault slip (750 m). The detached fault model appears to improve the fit to the footwall data with a slip of 650 m; however, the RMSE of both models is nearly the same (24.6 m for isolated and 25.2 m for detached). We thus prefer a model that includes a detachment at ∼3.3 km depth. Neither model explains the westward tilt at the far eastern end of the section. Models of cross-sections B and D using the detached fault geometry also result in a best-fit slip of 650 m (Figs. 9B and 9C). Cross-section B, across the Wheeler fault, appears to be reasonably well fit by this model, although the residual is slightly higher (RMSE = 37.5 m). The model, however, fails to capture the flexure associated with the Gregg Basin syncline in the hanging wall of the Lost Basin Range fault (section D, RMSE = 47.3). The addition of a component of buried slip (250 m) with the upper tip located 800 m below the surface to a model with surface-breaking slip of 550 m yields a significantly better fit to section D (Fig. 9C, model 2; RMSE = 11.7 m). Deformation associated with the Lost Basin Range fault thus appears to be best explained by a significant component of blind slip, while deformation associated with the Wheeler fault is reasonably modeled by a surface-breaking fault.
Variation in Fault and Fold Geometry Along Strike
Previous authors (Wallace et al., 2005; Umhoefer et al., 2010) mapped the Gregg Basin syncline as a continuous feature in the hanging wall of both the Wheeler and Lost Basin Range faults from north of South Cove in the north to near Hualapai Cove in the south. Our structural data add to this interpretation by demonstrating that the geometry of the syncline is distinctly different in the hanging wall of each fault, and that the syncline may continue much farther south than previously recognized (Figs. 3 and 4). While short-wavelength normal-drag folds are common across our map (indicated by steep westward dips over short wavelengths recorded adjacent to the Lost Basin Range and Wheeler faults; Fig. 5B), field observations and the structure contour map indicate that the shallow, long-wavelength, W-dipping fold exists exclusively adjacent to the Lost Basin Range fault. This relationship suggests that drag folding is an important contributor to kilometer-scale deflection of Hualapai Limestone strata along the Lost Basin Range fault, but only at much smaller scales along the Wheeler fault. Figure 10 shows photos highlighting the variation in fold geometry adjacent to the Wheeler and Lost Basin Range faults, as well as a schematic diagram demonstrating our interpretation of the shape of the drag and reverse-drag folds that crop out south of the fault relay. Both folds demonstrate variation from classical fold geometry (Hamblin, 1965; Schlische, 1995). The wavelength of the reverse-drag fold is an order of magnitude less than expected for continental-scale normal faults (e.g., Stein and Barrientos, 1985), while the normal-drag fold wavelength is 1–2 orders of magnitude larger than that of typical drag folds. We suggest that these apparently anomalous fold widths can provide insights into the geometry of the fault system during and after deposition of the Hualapai Limestone.
Furthermore, cross-section E, which is influenced by data closer to the Lost Basin Range fault than cross-section F, demonstrates a distinct eastward rise in structural elevation with proximity to the fault. The geometry of cross-section E is similar to that seen closer to the fault relay (cross-section D) and mapped by Wallace et al. (2005) as the Gregg Basin syncline. At cross-section F, alluvium shed from the Lost Basin Range covers the Hualapai Limestone stratigraphy to ∼1500 m from the fault, limiting our ability to constrain the fold geometry within that range. Thus, we interpret the Gregg Basin syncline as existing at least as far south as cross-section E, and perhaps the entire length of the Lost Basin Range fault.
To investigate possible reasons for the along-strike variation in fold geometry, we modeled surface deflections related to slip on faults with a range of upper and lower tip-line depths. Reverse- and normal-drag fold geometry is a function of fault geometrical properties (Resor and Pollard, 2012; Resor, 2008; Willsey et al., 2002; Withjack et al., 1990), mechanical properties of the deforming strata (Wilson et al., 2009), mechanical effects due to fault linkage (Crider and Pollard, 1998; Willemse, 1997), and/or slip distribution along the fault (White and Crider, 2006; Grasemann et al., 2005). Furthermore, modeling efforts using a wide range of techniques have found a positive correlation between fault upper tip depth and width of the hanging-wall syncline created by blind faulting (e.g., Withjack et al., 1990; Erslev, 1991; Johnson and Johnson, 2002; Willsey et al., 2002). Because the wavelength of the Gregg Basin syncline varies along strike in our field area, our models are consistent with different tip-line depths for the Wheeler and Lost Basin Range faults. Specifically, our illustrative modeling effort suggests that much of the folding seen today at the surface in the hanging wall of the Lost Basin Range fault formed when that fault was significantly buried (Fig. 9C, model 2; we reiterate that the modeling results are suggestive of this, but they should not be taken as quantitative calculations of fault geometry or slip magnitude). In contrast, deformation related to the Wheeler fault shows very little normal-drag flexure and is thus best modeled by a surface-breaking (or nearly surface-breaking) fault (Fig. 9B).
Therefore, our results are consistent with a surface-breaking Wheeler fault that divided the Hualapai Limestone into two subbasins until the deposition of the youngest Hualapai strata (ca. 6 Ma). This is consistent with geochemical evidence for distinct Hualapai Limestone basins prior to 6 Ma (Crossey et al., 2015). At ca. 6 Ma, these youngest strata overtopped the fault (Crossey et al., 2015). Therefore, at ca. 6 Ma, a low-relief lake/marsh depositional surface formed by the upper Hualapai Limestone bedding in Gregg and Grand Wash Basins likely existed between the topographic highs formed by the Hiller Mountains to the west, the Grand Wash Cliffs to the east, and subtler yet important topographic highs to the north and south. After integration, carbonate deposition ceased, and the Hualapai Limestone in the hanging wall of the Wheeler fault was subsequently folded into a short-wavelength drag fold. In contrast, an upward-propagating Lost Basin Range fault tip deformed currently exposed Hualapai Limestone strata while far beneath the idealized bed and propagated to the surface sometime after deposition of the latest Hualapai Limestone. This different history may reflect the location of the northern end of the Lost Basin Range fault in the hanging wall of the Wheeler fault or a change in the regional fault system, as discussed in the following.
Apatite fission-track dating suggests Wheeler Ridge and the Lost Basin Range had been significantly exhumed by ca. 15 Ma (Fitzgerald et al., 2009; Umhoefer et al., 2010). Our results are consistent with these data. Specifically, the top of the footwall Hualapai Limestone exposure lies at a similar elevation to the crest of the present-day Wheeler Ridge; therefore, the coalesced paleo–Lake Hualapai at 6 Ma (Crossey et al., 2015) may have come close to inundating Wheeler Ridge. Given the relatively thin nature of the Hualapai Limestone and the fact that apatite fission-track methods employed by Fitzgerald et al. (2009) measure cooling through the upper 3–5 km of Earth’s crust, the resulting inundation and low-relief surface (bounded by the topographic highs mentioned earlier) are still consistent with significant exhumation of Wheeler Ridge and the Lost Basin Range by 15.3 Ma (Fitzgerald et al., 2009; Umhoefer et al., 2010), yet not with significant relief formed between the Hualapai Limestone and Wheeler Ridge or the Lost Basin Range. Specifically, our results are consistent with a hypothesis in which exhumation prior to 12 Ma (or perhaps 14 Ma; Umhoefer et al., 2010) of the Lost Basin Range occurred due to faulting on the Meadview fault. The Wheeler fault was also active during this time. At ca. 12 Ma, Hualapai Limestone deposition commenced in a basin created in the distal footwall of the Meadview fault (e.g., west of the Lost Basin Range but prior to expression at the surface of the Lost Basin Range fault) and in the hanging wall of the Wheeler fault. Sometime after 6 Ma, the Lost Basin Range fault propagated toward the surface and unified the Wheeler–Lost Basin Range fault step-over system. This is a modification of the interpretation presented in Umhoefer et al. (2010), in which the Lost Basin Range fault is thought to have been active synchronously with the Wheeler fault due to its close spatial association. Umhoefer et al. (2010) further presented evidence that slip on the Meadview fault had ceased by ca. 14 Ma. Our results do not conflict with this interpretation, but point to an intriguing gap of as much as 6 m.y. during which the Wheeler fault was the only active surface-breaking fault west of the Grand Wash Trough and east of the Hiller Mountains. After the integration of the Colorado River, carbonate sedimentation ceased, yet the faults continued to accrue displacement, leading to the landscape we see today.
The idea that Hualapai Limestone strata south of Hualapai Bay were deflected into their current geometry by a propagating fault tip line below the depositional surface stands in contrast with the observation that the Lost Basin Range fault breaches Earth’s surface in the present day, offsetting the upper Hualapai Limestone by ∼450 m. We speculate that the propagation to slip ratio increased at some point after Hualapai Limestone deposition, extending the tip line to the surface. The present-day structures are thus the result of overprinting episodes of deformation, beginning when the Lost Basin Range fault was buried deeply beneath Gregg Basin and propagating toward the currently exposed horizon, and ending at a time in which the fault had broken through the surface and offset all of the preserved strata. We have attempted to capture the essence of this evolution by summing model results using surface-breaking and buried faults. However, our modeling approach is independent of time. It is clear that to fully reproduce the complex strain field observed in eastern Lake Mead, we would need to use models that more easily incorporate fault propagation. However, this is beyond the scope of this study, as we wish only to highlight possible reasons for variation in fault geometry along strike.
The lower fault tip model results are revealing as well (Fig. 9A). Few exposures exist that document the full thickness of the basin fill beneath the Hualapai Limestone, yet Wallace et al. (2005) estimated there are 120 m of clastic sedimentary rock beneath the 120 m of Hualapai Limestone filling Gregg Basin. This implies that the basin is far thinner than the lower fault tip depths calculated in our modeling experiments. Gregg Basin is underlain by Proterozoic gneiss and volcanic rocks (Wallace et al., 2005), rocks with different mechanical properties than the Hualapai Limestone. The models we employ only permit one set of elastic constants and therefore do not satisfactorily model the variable material properties the faults likely experience at depth. However, the depth to detachment implied by our best-fit length (3.3 km) is of the same order of magnitude as depth to detachment values using a range of kinematic techniques (5.2 km; Lopez-Pearce, 2010).
Given the caveats and limitations involved in our data and modeling approach, we focus primarily on the broad patterns of Hualapai Limestone deformation: Our two main conclusions regarding fault and fold geometry for our field area are (1) the Gregg Basin syncline likely represents deformation over a fault tip line buried beneath the deforming horizon, and (2) the Wheeler fault upper tip likely was close to the surface for the entirety of Hualapai limestone deposition and deformation.
It is possible that our structural data are limited by the extent of Hualapai Limestone exposure and thus do not represent the full deformation field related to slip on the Wheeler and Lost Basin Range faults. However, there are several pieces of evidence suggesting we have quantified most, if not all, of the deformation. First, the wavelength of reverse-drag folds increases with increasing fault throw (Schlische, 1995), with large-offset faults building 30–40 km of reverse drag (e.g., Barnett et al., 1987; Stein and Barrientos, 1985). However, we regard the smaller wavelengths of the hanging-wall folds in our field area as plausible for moderate-offset faults. Second, hanging-wall structural cross-sections B, C, and F all achieve clearly horizontal slopes with distance westward from the fault trace within the extent of our data (Fig. 4). Based on previous documentation of reverse-drag geometry (Hamblin, 1965; Barnett et al., 1987; Stein and Barrientos, 1985; Resor, 2008), we would not expect the structure of the Hualapai Limestone to flatten for several kilometers and then continue to rise westward, unless another, unrecognized fault exists somewhere in the vicinity of the current Lake Mead. Corroborating this expectation is the elevation of the upper Little Burro Bay Hualapai Limestone section (Howard et al., 2003; Beard et al., 2007), which is ∼20 m below the elevations achieved by our structural data at cross-section F (Fig. 4). Cross-sections D and E both continue to gain elevation to the west, indicating that the classic reverse-drag fold imaged at cross-section F is likely present along the entire length of the Lost Basin Range fault. While we cannot definitively tie our idealized horizon to this exact level, we note that the upper Little Burro Bay section and the idealized bed mapped in Gregg Basin are composed of carbonate facies indicative of upper Hualapai Limestone deposition (Fig. 6), indicating a maximum possible offset of ∼30 m. In the extreme case that the idealized bed projects to 30 m above the Little Burro Bay section, the slightly altered geometry would not significantly change our modeling results or interpretations.
Implications for Differential Incision Rates in Gregg Basin and the Grand Wash Trough
Figures 3 and 4 indicate that extension in the area took different forms along fault strike: vertical displacement between Grapevine Mesa and Gregg Basin was largely accommodated by vertical block movement in the northern map area; in contrast, deformation in the southern map area caused development of much more pronounced folds. This variation necessarily caused spatial gradients in the incision rate as the Colorado River cut down through actively deforming strata. We therefore refer to this variation as apparent differential incision because it reflects both river incision rates and any structural adjustments of the block being incised (Pederson et al., 2002; Karlstrom et al., 2007). The following analysis hinges on three principal assumptions: (1) There was a lake highstand at ca. 6 Ma that united the Gregg and Grand Wash Hualapai Limestone basins, (2) this surface had very little relief due to the syntectonic nature of deposition of the lacustrine/marsh Hualapai Limestone, and (3) there was a close temporal coincidence between the end of Hualapai Limestone deposition and the arrival of the Colorado River in eastern Lake Mead. The first assumption is well supported by analysis of facies presented here and detailed in Crossey et al. (2015) and further corroborated by geochemical evidence that demonstrates close chemical signatures between the two basins by ca. 6 Ma (Crossey et al., 2015). The second assumption is reasonable given the evidence for syndeformational sedimentation and the observation that tilt-fanning of Hualapai Limestone strata rarely exceeds 5°–10° (Wallace et al., 2005). The third assumption is widely believed to be correct (e.g., Howard and Bohannon, 2001; Spencer et al., 2001; Faulds et al., 2001b; Crossey et al., 2015) due to the hydrological changes implied by a nascent through-going Colorado River. Independent evidence is suggested by Matmon et al. (2012), who used the cosmogenic nuclide burial method of Wolkowinsky and Granger (2004) to measure the age of a deposit near Hualapai Wash that is interpreted to be one of the oldest known Colorado River gravel deposits. The age of this deposit, which sits above Hualapai Limestone bedding, was found to be 5.35 +1.65/–0.97 Ma. Therefore, it seems likely the Colorado River established a presence in the region soon after, if not coincident with, the end of Hualapai Limestone deposition at 6 Ma.
To provide an approximation of the apparent differential incision rate of the Colorado River since the time of integration, we first projected the structure contour map (Fig. 3) over the course of the pre–Hoover Dam historic river in swaths that parallel the river course in the hanging wall and footwall of the Wheeler–Lost Basin Range fault system (Fig. 3). For the section of the river profile influenced by the Wheeler fault (Wheeler fault influence, Fig. 3), the elevations of the idealized bed range between 470 m and 499 m. Likewise, elevations for the profile in the hanging wall of the Lost Basin Range fault range between 473 m and 521 m. The higher elevations in the hanging wall of the Lost Basin Range fault are due to the larger reverse-drag fold there. Footwall elevations range between 921 m and 960 m. Next, we extracted the vertical incision distance by subtracting the historic river elevation (268 m; Beard et al., 2007) from these structural elevations. Finally, we divided the incision distances by 6 m.y. to estimate an apparent incision rate that has been modified by the post–6 Ma folding and faulting quantified by the idealized Hualapai Limestone horizon. In the hanging wall, the apparent incision rate spatially varies between 33 and 42 m/m.y. as the river incised through the evolving structure. While more conjectural due to the larger distance between the idealized bed in the footwall and the Colorado River, the footwall apparent incision rate spatially varies between 108 and 115 m/m.y.
The map pattern of the idealized bed indicates that the hanging-wall incision rates calculated here may be minima when the full deformation field is taken into account. Apparent incision rate estimates taken in the hanging wall along the course of the pre–Hoover Dam Colorado River do not account for the highest elevations achieved by the reverse-drag fold in the hanging wall of the Lost Basin Range fault (Fig. 3). These higher elevations may be seen at the west end of cross-section F and at the Hualapai Limestone section near Little Burro Bay, assuming that this section may be tied to the idealized bed (discussed in section on “Variation in Fault and Fold Geometry Along Strike”). Using 690 m as the elevation of the Little Burro Bay section top, we estimate an apparent incision rate of ∼70 m/m.y. where the Colorado River flowed over the full flexure in the hanging wall of the Lost Basin Range fault.
These apparent incision rates may also be compared to an independent estimate at Sandy Point, where a 4.4 Ma basalt flow (Faulds et al., 2001b) overlies Colorado River gravels, indicating the presence of the Colorado River through eastern Lake Mead by at least that time. Sandy Point lies ∼6.5 km north of the extent of the structural data along the river course and yields an apparent incision rate of 27 m/m.y. (Faulds et al., 2001b), slightly lower than our hanging-wall estimates for the Wheeler fault. We suspect that the hanging-wall structural elevations continue to lessen to the north, and the discrepancy may therefore reflect true structure-dampened incision, although we cannot rule out the possibility that the average incision rate slowed after 4.4 Ma.
The Hualapai Limestone–based apparent incision rates, although not vastly different from the point measurement at Sandy Point, are interesting in the context of regional uplift and incision patterns (e.g., Darling et al., 2012; Crow et al., 2014). In particular, Crow et al. (2014) found a semisteady incision rate of 101.0 ± 4.6 m/m.y. for the western Grand Canyon over the last 4 m.y. The samples used for this estimate were collected ∼25 km west of the Hurricane fault, outside of the expected influence of hanging-wall flexure (Karlstrom et al., 2007). This measurement is similar to our estimate for the footwall incision rate of 108–115 m/m.y. Therefore, if apparent incision rate can be linked to uplift through an assumption of topographic steady state over the chosen time interval, the similarity between eastern Lake Mead and western Grand Canyon incision rates suggests the Wheeler fault system may have been a significant player in regional uplift of the Colorado Plateau, at least over the past 6 m.y. Furthermore, the elevations of the structural contour map may provide estimates for how regional uplift (fault throw minus flexure) may have varied along strike. A simple way to conduct this analysis is to subtract elevations at the far westward end of each hanging-wall structural cross section from a standard footwall value of ∼900 m, therefore avoiding the complication of flexure and fault throw measured directly at the fault. Doing this, we find estimates of regional uplift of 404 m, 419 m, 411 m, 303 m, and 191 m for structural cross-sections B, C, D, E, and F, respectively. Although our structural data are limited to a small portion of the Wheeler fault, we suspect the deformation style adjacent to the Wheeler fault may be more representative of regional patterns than the Lost Basin Range fault, which has a significantly shorter map-view length. Therefore, we prefer using ∼410 m as an estimate for regional uplift accommodated by the Wheeler–Lost Basin Range fault system.
By quantifying the structural geometry of the Hualapai Limestone in Gregg Basin and the Grand Wash Trough, we have demonstrated variations along fault strike that bear on the structural evolution of eastern Lake Mead. In particular, we have shown that reverse-drag folds in the northern part of Gregg Basin are in fact quite subtle, and vertical block lowering has dominated the deformation style there. In contrast, deformation along the Lost Basin Range fault is accommodated by a classic reverse-drag fold, and a normal-drag fold that is similar in amplitude to the reverse-drag fold previously recognized (Lucchitta, 1979; Howard and Bohannon, 2001; Karlstrom et al., 2007). Much of the deformation was coeval with sedimentation; therefore, the environment first experienced by the nascent Colorado River in eastern Lake Mead was primarily a low-relief lake/marsh surface filling preexisting internally drained basins. Extending our structural framework to depth, the structural data are consistent with prior relief on the Wheeler fault that was overtopped to form paleo–Lake Hualapai (Crossey et al., 2015). Further, our results suggest that Hualapai Limestone deposition occurred at first in the hanging wall of the Wheeler fault and west of the Lost Basin Range in the distal footwall of the Meadview fault. Sometime after 6 Ma, the Lost Basin Range fault propagated toward the surface, eventually offsetting the hanging-wall Hualapai Limestone from footwall Lost Basin Range metamorphic rocks. This is consistent with apatite fission-track exhumation dates presented by Fitzgerald et al. (2009). Postintegration, the Colorado River continued to incise as the Hualapai Limestone was offset by faulting and folding that modulated incision rates in areas over which the river flowed. Our results also emphasize the importance of constraining the full strain field, including fault-related folding, when modeling incision in tectonically active areas. Variations along strike may lead to differences in fault and fold relationships that may influence landscape evolution.
Seixas would like to acknowledge support for field work provided by the Gutmann Scholarship at Wesleyan University, where much of this work was accomplished as part of a senior thesis. Seixas would also like to thank Max Gardner for valuable insight and assistance in the field and M. Gilmore for constructive reviews of an early draft. Thanks go to E. Duebendorfer, P. Umhoefer, and S. Beard for detailed reviews that greatly helped to clarify the ideas in the manuscript. Funding for Resor came in part from the Smith Fund and a Project Grant at Wesleyan University. For Karlstrom and Crossey, funding was in part from National Science Foundation (NSF) grant EAR-0838575 from the Hydrologic Sciences Program, EAR-1119629 and EAR-1242028 from the Tectonics Program, and DGE-0538396 from the NSF GK–12 program (in support of Lopez-Pearce). Research permits from Grand Canyon National Park and Lake Mead Recreation also helped enable the work.