Abstract

New geologic mapping and tephrochronologic assessment of strata in extensional basins surrounding Knoll Mountain (Nevada, USA) reveal a geologic history linked to tectonic development of the Yellowstone hotspot and Snake River Plain to the north, and to the Ruby–East Humboldt–Wood Hills metamorphic core complex to the south. Data from these areas are utilized to present a paleogeographic reconstruction of northeastern Nevada–south-central Idaho depicting the architecture of extensional faulting and basin development during collapse of the Nevadaplano over the past 17 m.y.

Knoll Mountain is a northeast-trending horst along the southern margin of the Snake River Plain and track of the Yellowstone hotspot. The horst is bounded on the east by the Thousand Springs fault system and basin, and on the west by the Knoll Mountain fault and basin, where streams currently drain north into the Snake River Plain. The Knoll and Thousand Springs basins form half-grabens that are filled with the ca. 16 Ma to ca. 8–5 Ma Humboldt Formation, which was deposited in alluvial, eolian, and lacustrine environments during slip along range-bounding faults and a series of late-stage synthetic intrabasin faults. Structural, chronologic, and sedimentologic assessment of the Humboldt Formation in the Knoll basin indicates that it records overall southward fluvial drainage with slip along the Knoll Mountain fault beginning ca. 16 Ma and continuing to at least 8 Ma, and that between 8 and ca. 5 Ma, a west-dipping intrabasin fault system had developed. Between ca. 8–5 Ma to ca. 3 Ma, several fundamental changes took place, beginning with the cessation of faulting followed by widespread erosion that in turn was followed by deposition of older alluvium. The reversal of drainage direction from south to north flowing in the Knoll basin also took place during this time period, but its age relative to the widespread erosion or older alluvium is unknown.

An integration of our work with previous studies north of Knoll Mountain reveal that the Knoll Mountain and intrabasin faults terminate to the north in the vicinity of the Jurassic Contact pluton, and that this area forms an accommodation zone separating broadly coeval and colinear faults bounding the ca. 10–8 Ma north-trending Rogerson graben, the northern end of which merges with the Snake River Plain. Furthermore, an integration of our work with previous work south of Knoll Mountain reveals that the Knoll Mountain fault formed part of a >190-km long, west-dipping fault zone that included the Ruby–East Humboldt detachment. This fault zone, which we refer to as the Knoll-Ruby fault system, had an extensive hanging-wall basin, the Knoll-Ruby basin. The Knoll-Ruby fault system was a prominent structure facilitating collapse of the Nevadaplano in northeastern Nevada between ca. 16 and ca. 8–5 Ma, and its central part produced partial exhumation of high-grade, mid-crustal metamorphic rocks in the Ruby–East Humboldt–Wood Hills metamorphic core complex. By 8–5 Ma, during the waning stages of extension along the Knoll-Ruby fault system, a series of intrabasin faults developed at about the same time as the integration of streams to form the incipient eastern reaches of the Humboldt River system. Profound changes in tectonics and paleogeography took place between ca. 8–5 Ma and ca. 3 Ma, that included the extinction of the Knoll-Ruby and intrabasin basin fault systems followed by southward migration of significant tectonism away from the Snake River Plain, resulting in development of a set of modern normal faults responsible for uplift of the southern Snake Mountains, Ruby Mountains, East Humboldt Range, and Pequop Mountains. These new faults cut and dismembered the central and southern part of the Knoll-Ruby fault system and basin, effectively ending any fluvial connection between the northern and southern parts of the Knoll-Ruby basin. Since ca. 8–5 Ma to the present, the Knoll Mountain region has remained relatively tectonically quiescent, and continued subsidence in the Snake River Plain to the north induced capture of the drainage system in the Knoll basin and reversed the drainage direction from south to north flowing.

Our new findings indicate that (1) the Knoll-Ruby fault system and associated intrabasin faults were active until ca. 8–5 Ma, which is younger than the 12–10 Ma age generally recognized for cessation of major extension elsewhere in the northern Nevada region; (2) although this fault system was responsible for partial exhumation of core-complex metamorphic rocks, it extended well beyond the confines of the core complex proper; and (3) slip along faults in the Knoll Mountain region occurred before, during, and after passage of the hotspot at the longitude of Knoll Mountain. With the exception of significant faulting postdating passage of the hotspot, the timing of faulting in the Knoll Mountain area is consistent in a general way with the space-time pattern of extension recognized elsewhere along the southern margin of the Snake River Plain. However, it is unknown if the rate of fault slip increased during passage of the hotspot as it did in other areas.

INTRODUCTION

The northern Basin and Range Province constitutes part of a large area of middle Miocene to present continental rifting characterized by north-trending, normal fault–bounded basins and ranges (Fig. 1A). The volcanic track of the migrating Yellowstone hotspot, and the subsiding eastern Snake River Plain that developed along its track, are superimposed on the Basin and Range structure (Figs. 1A, 1B). Extension associated with development of the Basin and Range Province began at about the same time that the Yellowstone hotspot formed ca. 17–16 Ma, and both are still operative today (e.g., Camp et al., 2015). Crustal extension between ca. 17–16 and 12–10 Ma was relatively rapid and locally large magnitude (e.g., Colgan and Henry, 2009), and was partially manifest by the exhumation of mid-crustal metamorphic rocks in the Ruby–East Humboldt–Wood Hills and the Albion–Raft River–Grouse Creek metamorphic core complexes (Fig. 1B; e.g., Mueller and Snoke, 1993a, 1993b; Camilleri and Chamberlain, 1997; Wells et al., 2000; Egger et al., 2003; Colgan and Henry, 2009; Colgan et al., 2010; Henry et al., 2011; Konstantinou et al., 2012). Since ca. 10–12 Ma, extension was of lesser magnitude, and mostly focused on the eastern and western margins of the Basin and Range Province (Colgan et al., 2004; Wallace et al., 2008; Camp et al., 2015).

Although much is known about the general timing of normal faulting responsible for development of the northern Basin and Range Province, the details of its paleogeographic evolution are just beginning to emerge, in particular from two recent but geographically separated regional paleogeographic studies. In one study, Beranek et al. (2006) presented a paleogeographic reconstruction showing the evolution of topography and drainage divides along the track of the Yellowstone hotspot and associated Snake River Plain. Wallace et al. (2008) presented a paleogeographic reconstruction farther south, focused on faulting, basin, and stream development of the area west of, and including, the Ruby–East Humboldt–Wood Hills metamorphic core complex (Fig. 1B). These two reconstructions highlight different but interrelated processes affecting the paleogeographic evolution of the Basin and Range Province, but no reconstruction exists between these areas linking the paleogeography of the Snake River Plain area with that of the metamorphic core complex region. This paper presents and utilizes new data bearing on the timing of faulting and basin development between the Snake River Plain and the core complex area to present a more regional paleogeographic reconstruction of northeastern Nevada and south-central Idaho. Our reconstruction, from a paleogeographic perspective, broadly links processes and timing of events along the hotspot with those in the metamorphic core complex while illustrating the evolution and extent of fault systems operative between these areas since ca. 17 Ma.

The migration of the Yellowstone hotspot in northeastern Nevada and south-central Idaho had a profound influence on paleogeography in terms of the evolution of topography, drainage divides, and the timing of faulting. Hotspot rhyolitic volcanism and development of an associated tumescent topographic bulge began ca. 17–16 Ma in north-central Nevada and then migrated to its present position in Wyoming and Montana (Fig. 1B; e.g., Pierce and Morgan, 1992, and references therein; Beranek et al., 2006). The topographically elevated hotspot formed a regional drainage divide (Fig. 1B) with Miocene drainage systems in northern Nevada flowing to the south until subsidence in the eastern Snake River Plain was sufficient to cause southward migration of the drainage divide concomitant with reversal of drainage direction from south to north flowing in northernmost Nevada (e.g., Beranek et al., 2006; see Fig. 1B for the locations of the Miocene divide and the migrated, modern divide). Because young volcanic rock in the eastern Snake River Plain conceals older hotspot-related volcanic rock and structures, what is known about the effect of the hotspot on the space-time pattern of extensional faulting has been largely gleaned from late Cenozoic normal fault–bounded basins and their footwalls that project into the margin of the eastern Snake River Plain (e.g., Rodgers et al., 1990; Pierce and Morgan, 1992, 2009; Anders, 1994; Rodgers et al., 2002). Previous studies of these normal faults and basins that are east of ∼114°W longitude collectively indicate that a minor, but regional phase of Basin and Range extension between 16 and 11 Ma preceded a later time-transgressive phase of major extension (rapid slip) that migrated northeastward with the hotspot from ca. 13.5 Ma in the Albion–Raft River–Grouse Creek metamorphic core complex to 4–0 Ma in the vicinity of the modern hotspot (Fig. 1B; Rodgers et al., 1990; Pierce and Morgan, 1992, 2009; Anders, 1994; Wells et al., 2000; Rodgers et al., 2002; Egger et al., 2003; Konstantinou et al., 2012). Whether this space-time pattern of extension continues along the southern margin of the Snake River Plain west of ∼114°W longitude is uncertain due to the general paucity of regional data on the distribution and timing of extensional faulting in this area. Limited data from this region indicate that the Rogerson graben developed at the western end of the eastern Snake River Plain between ca. 10 and 8 Ma during volcanism in the Bruneau-Jarbidge and Twin Falls eruptive centers from 12.7 to 8 Ma (Fig. 1B; Bonnichsen and Godchaux, 2002; Andrews et al., 2008; Knott et al., 2016b). Although the Rogerson graben may be part of the late phase of relatively rapid extension that accompanied the migrating hotspot, an older phase cannot be ruled out because the lack of exposure of rocks older than ca. 12 Ma precludes assessment of earlier extension (Andrews et al., 2008; Knott et al., 2016b). However, south of the Rogerson graben, albeit very distal to the track of the hotspot, significant extension between 16 and 10 Ma predating the Rogerson graben has been documented in the Ruby–East Humboldt metamorphic core complex and the adjacent Windermere Hills (e.g., Mueller et al., 1999; Colgan et al., 2010). The area between the Rogerson graben and the Ruby–East Humboldt–Wood Hills metamorphic core complex, which we refer to as the Knoll Mountain area (Fig. 1B), is relatively unstudied, but contains middle Miocene basins and fault systems (Deibert and Camilleri, 2006) that are an important link for connecting the paleogeography of the Snake River Plain to that of the core complex region and for assessing whether the space-time pattern of extension along the margin of the eastern Snake River Plain persists west of ∼114°W longitude.

This paper presents new geologic mapping, stratigraphic, structural, and tephrochronologic data from the Knoll Mountain area that help to refine the paleogeographic evolution of northeastern Nevada–south-central Idaho. The purpose of this paper is threefold. First, we utilize our data to develop a tectonic and paleogeographic reconstruction of the Knoll Mountain region. Second, we assess the lateral and temporal relationships between faults in the Knoll Mountain, Rogerson graben, and Ruby–East Humboldt–Wood Hills core complex areas in order to establish the areal distribution of faulting and whether this faulting formed part of the eastward-sweep of extension that accompanied migration of the hotspot. Third, we present the first Miocene to Holocene tectonic and paleogeographic reconstruction along the margin of the Snake River Plain in northeastern Nevada–south-central Idaho by integrating the fundamental elements of our reconstruction of the Knoll Mountain region with published paleogeographic assessments along the track of the Yellowstone hotspot to the north and the Ruby–East Humboldt–Wood Hills metamorphic core complex area to the south.

GEOLOGIC FRAMEWORK OF NORTHEAST NEVADA

Modern Drainage Basins and Erosion: Windows into the Development of Late Cenozoic Normal Faults and Basins

Northeastern Nevada forms part of the modern topographic backbone of the Basin and Range Province and contains the headwaters of the Humboldt River, Snake River, and Salt Lake drainage systems (Fig. 1C). These stream systems have significantly incised middle Miocene and younger basin fill and associated normal faults, thereby revealing a remarkable record of sedimentation and extension and their effect on the evolution of drainage divides during development of the Basin and Range Province (Fig. 1C).

Stratigraphy

Pliocene and older rocks in northeastern Nevada can be divided into three informal general sequences (Fig. 2A). From oldest to youngest, they are the basement sequence, the Clover Creek sequence, and the middle Miocene to Pliocene sequence. The basement sequence includes Precambrian to Triassic continental margin clastic, carbonate, and igneous rocks that are overlain in places by 46–29 Ma Eocene to Oligocene volcanic and sedimentary rocks (Brooks et al., 1995; Mueller et al., 1999; Haynes, 2003; Henry, 2008; Lund Snee, 2013; McGrew and Snoke, 2015; Lund Snee et al., 2016). In most places, the basement sequence is overlain by the middle Miocene to Pliocene sequence, which is composed largely of the ca. 16 to ca. 8 Ma Humboldt Formation, but locally also contains the ca. 16–15 Ma Jarbidge Rhyolite (Brueseke et al., 2014) and Pliocene lacustrine strata (Fig. 2A; Reheis et al., 2003, 2014; Wallace et al., 2008). However, locally in the northern Pequop Mountains, East Humboldt Range, and possibly in the Wood Hills, there is a sequence of strata between the middle Miocene to Pliocene and basement sequences. In the East Humboldt Range, this intervening sequence has been referred to as the sedimentary sequence of Clover Creek (McGrew and Snoke, 2015), and we use this name for units in nearby areas that occupy the same stratigraphic position. The age of this sequence is bracketed between 38 and 16 Ma in the Windermere Hills and between 29 and 16 Ma in the East Humboldt Range (Mueller et al., 1999; McGrew and Snoke, 2015; no geochronologic data exist for the section in the Wood Hills).

Tectonics

Just prior to development of the Basin and Range Province, northeastern Nevada was part of the Nevadaplano, which was a regionally elevated orogenic plateau with a north-northeast–trending drainage divide in eastern Nevada (Fig. 1A; Henry, 2008). The Nevadaplano formed as a consequence of Mesozoic to Paleogene tectonic convergence and crustal thickening along the western margin of North America (e.g., DeCelles, 2004; Henry, 2008; Cassel et al., 2014). During this period of convergence, northeastern Nevada formed part of the hinterland of the Sevier fold and thrust belt and underwent Jurassic and Cretaceous thrust faulting, plutonism, metamorphism, and as much as ∼30 km of tectonic burial of Paleozoic basement (e.g., Hodges et al., 1992; Camilleri and Chamberlain, 1997; McGrew et al., 2000; Hallett and Spear, 2014). Convergence was followed by protracted, multiphase extension beginning in the Late Cretaceous and continuing to the present (e.g., Snoke and Miller, 1988; Hodges et al., 1992; Mueller and Snoke, 1993a, 1993b; McGrew and Snee, 1994; Camilleri and Chamberlain, 1997; McGrew et al., 2000; Howard, 2003; Sullivan and Snoke, 2007; Colgan et al., 2010; Henry et al., 2011; Pape et al., 2016; Lund Snee et al., 2016). Extension ultimately culminated in the exhumation of once structurally buried, mid-crustal metamorphic rocks in the Ruby Mountains–East Humboldt–Wood Hills metamorphic core complex and the associated Pequop Mountains (Fig. 2). Within this core complex, maximum metamorphic grade and pressure are recorded in rocks in the northernmost East Humboldt Range (Hodges et al., 1992; McGrew et al., 2000; Hallett and Spear, 2014), and from there, the grade decreases to the south and east, where metamorphism dies out and gradually transitions into unmetamorphosed rock (Fig. 2; Camilleri and Chamberlain, 1997; Camilleri, 1998). Exposure of the metamorphic rocks at the surface occurred during three successive phases of extension that postdate the basement sequence. Remnants of these phases of extension include successive generations of normal faults and their associated hanging-wall basin fills (Fig. 2A).

The first phase of extension resulting in initial exposure of the metamorphic rocks occurred between 38 and 16 Ma and is associated with localized north- to north-northeast–directed extension along the Holborn fault or faults (Mueller and Snoke, 1993a; Mueller et al., 1999) and deposition of the Clover Creek sequence (Fig. 2). Because of overprinting by the younger phases of extension, only scarce remnants of this earliest phase of extension are present. McGrew and Snoke (2015) reported the presence of clasts of metamorphic rock in the Clover Creek sequence in the East Humboldt Range that are likely sourced from a Wood Hills–like part of the core complex, indicating that initial exhumation of the metamorphic complex occurred during this phase of extension.

The second phase of extension triggered regional collapse of the Nevadaplano and the formation of the Basin and Range Province (Dickinson, 2006). It also was responsible for widespread exhumation of the metamorphic rocks in the Ruby–East Humboldt–Wood Hills core complex (e.g., Mueller and Snoke, 1993a; Camilleri and Chamberlain, 1997; Wallace et al., 2008; Colgan and Henry, 2009; Colgan et al., 2010; Lund Snee et al., 2016). In northeastern Nevada, this phase of extension took place from ca. 16 to at least 7 Ma and involved the development of three major half-graben basins filled with the synextensional middle to late Miocene Humboldt Formation (Table 1). These basins and their bounding normal faults include: the Thousand Springs basin and fault system; the Knoll basin and Knoll Mountain fault; and the Ruby basin, which was bounded by the Ruby–East Humboldt detachment (also known locally as the Mary’s River fault of Mueller and Snoke, 1993a) and the northern Snake Mountains fault (Fig. 2; Mueller and Snoke, 1993a; Deibert and Camilleri, 2006; Wallace et al., 2008; Colgan et al., 2010; Lund Snee, 2013; Lund Snee et al., 2016). Please note that the Ruby basin is spatially and temporally equivalent to the Elko basin of Wallace et al. (2008); however, we have chosen to use the name Ruby basin herein to avoid any possible confusion with other studies that use the term Elko basin to refer strictly to the region in which sediment of the Eocene Elko Formation was deposited (e.g., Smith and Ketner, 1976; Solomon et al., 1979; Haynes, 2003; Cline et al., 2005; Lund Snee et al., 2016; and many others). Today, of these three basins, only the Thousand Springs and Knoll basins and their bounding faults remain relatively intact. In contrast, the Ruby–East Humboldt detachment and the Humboldt Formation along the eastern margin of the Ruby basin have been dismembered by normal faults related to the third late Miocene (?) to Holocene phase of extension, which we informally refer to as the modern phase of extension.

The third, modern phase of extension is localized and produced the set of normal faults (red faults in Fig. 2) responsible for uplift of the modern Ruby Mountains, East Humboldt Range, southern Snake Mountains, and Pequop Mountains (e.g., Mueller and Snoke, 1993a; Camilleri and Chamberlain, 1997; Colgan et al., 2010; McGrew and Snoke, 2015), some of which have Quaternary offset as well as recent earthquake activity (Wesnousky and Willoughby, 2003; Wesnousky et al., 2005; Ramelli and dePolo, 2011; Henry and Colgan, 2011). Today, remnants of the Ruby Mountains–East Humboldt detachment and early Ruby basin (i.e., sediment of the Humboldt Formation) are extensively exposed along the western margins of the Snake Mountains, Ruby Mountains, and East Humboldt Range, as well as in a topographic and structural embayment between the East Humboldt Range, Wood Hills, and southern Snake Mountains (Fig. 2; Thorman, 1970; Coats, 1987; Mueller and Snoke, 1993a; Camilleri and Chamberlain, 1997; Howard, 2003; Thorman et al., 2003; Wallace et al., 2008; Camilleri, 2010a; Colgan et al., 2010; Henry and Thorman, 2011; McGrew and Snoke, 2015).

STRATIGRAPHIC AND STRUCTURAL ARCHITECTURE OF THE KNOLL MOUNTAIN AREA

The Knoll Mountain area contains the Granite Range, the northern Snake Mountains, Knoll Mountain (the H D range of Riva, 1970), and the Knoll and Thousand Springs basins (Fig. 2A). The mountain ranges expose Paleozoic strata, and locally, in the Granite Range, these strata are intruded by the Jurassic Contact pluton (Gardner, 1968; Riva, 1970; Coats, 1987; McFarlane, 2001). In contrast, the Knoll and Thousand Springs basins primarily expose the Miocene Humboldt Formation (Riva, 1970; Coats, 1987; Deibert and Camilleri, 2006), which is locally capped by a veneer of older alluvium (Fig. 3A).

Our mapping indicates that Knoll Mountain is a horst bounded on the west by the Knoll Mountain fault and on the east by the Bell Canyon and Black Rock faults (Fig. 3A), and that it is also a structural divide that regionally separates a series of west-dipping faults on the west from a system of east-dipping faults to the east (Fig. 2A). The horst terminates to the north in the vicinity of the Contact pluton where extension is distributed into a network of northwest- to north-northeast–striking conjugate faults (Fig. 3A).

The Knoll Mountain, Bell Canyon, and Black Rock faults are brittle faults that mostly have low to moderate dips of ∼14°–40° (dip angles are derived trigonometrically from contouring fault planes). Adjacent to these faults, Paleozoic rocks in the footwall are characterized by brecciation or cataclasis, whereas the poorly lithified Humboldt Formation in the hanging wall generally exhibits normal or reverse drag (Fig. 3).

Knoll Basin

Three primary tectonic structures controlled the development and architecture of the Knoll basin. These include the Knoll Mountain fault, a west-dipping intrabasin fault zone that consists of the Henry and Summer Camp faults that transect the middle of the basin, and the Hice syncline in the northeastern corner of the basin (Fig. 3A). The sediment filling the Knoll basin consists of two general sequences: the Miocene Humboldt Formation (Riva, 1970; Coats, 1987; Deibert and Camilleri, 2006) and alluvium (Fig. 3A). In the northeastern corner of the basin, Deibert and Camilleri (2006) subdivided the Humboldt Formation into several informal members as well as assessed their depositional environments and history with a focus on the origin and evolution of an incised-valley system. Here we expand upon the work of Deibert and Camilleri (2006) to establish the larger scale stratigraphic and structural architecture and evolution of the Knoll basin.

Humboldt Formation

The Humboldt Formation is composed primarily of poorly lithified sandstone and conglomerate with minor amounts of non-welded and welded tuff (Deibert and Camilleri, 2006). Beds of shale, breccia, and limestone are sparse. Framework grains of sandstone are mostly volcanic glass shards that were reworked from air-fall and pyroclastic flow deposits. Conglomerate units contain clasts derived from Paleozoic strata, the Jurassic Contact pluton, and Tertiary extrusive and intrusive igneous rocks in the surrounding mountain ranges (see the Supplemental Item1 for additional information about granitic clasts in the Humboldt Formation). Paleocurrents and the presence of clasts of the Contact pluton indicate that drainage and sediment transport overall was to the south during deposition of the Humboldt Formation (Deibert and Camilleri, 2006; Appendix 1). This indicates that the Contact pluton and its wall rocks formed a west-trending topographic high that formed the northern margin of the Knoll basin during deposition of the Humboldt Formation.

The Humboldt Formation is divided into five informal members: from oldest to youngest, they are the Blanchard, Knoll, Cave, Bloody Gulch, and Eagle Flat members (Fig. 3). Here we summarize the stratigraphy and depotectonic environments of the four oldest members, previously described in Deibert and Camilleri (2006), and the Eagle Flat member, which is a newly recognized unit described in more detail in Appendix 1. The Blanchard member forms the base of the Humboldt Formation and unconformably overlies pre-Miocene rocks. The Blanchard member was deposited in an alluvial plain with minor braided streams during initial subsidence of the basin related to extension along the Knoll Mountain fault. The overlying Knoll member reflects a profound nearly basin-wide change in environments to a shallow, freshwater lacustrine environment. With a local exception, the lacustrine Knoll member is conformably overlain by the Bloody Gulch member, which represents a return to an alluvial plain with minor braided streams. The aforementioned exception is in the northeastern corner of the basin where parts of the Blanchard and Knoll members were eroded during development of a southwestward-draining incised valley that was subsequently filled with eolian and fluvial sediment of the Cave member (Fig. 4). The Cave member is only present locally between the Bloody Gulch member and either the Knoll or Blanchard members in the northeastern part of the basin (Fig. 4).

The youngest member, the Eagle Flat member, is a widespread unit that is restricted to the hanging wall of the Henry, and possibly Summer Camp, faults (Fig. 3). It is composed of sandstone and minor amounts of conglomerate and siltstone deposited in an alluvial plain environment (Fig. 5). In addition, the Eagle Flat member also contains minor amounts of debris-flow deposits. Because steep slopes are required to initiate debris flows (Costa, 1984), the presence of debris flows in this member suggest that parts of the alluvial plain were in proximity of steep highlands (Fig. 5; see Appendix 1 for details).

The lower four members of the Humboldt Formation represent syntectonic basin fill deposited during slip along the basin-bounding Knoll Mountain fault system (Deibert and Camilleri, 2006), and we infer that the Eagle Flat member is syntectonic basin fill deposited during slip along the Henry fault. We make this inference because the Eagle Flat member is restricted to the hanging wall of the Henry fault and it appears to be very thick, probably as much as, or greater than, two times the total aggregate thickness of the older members of the Humboldt Formation (e.g., see section A-A′ in Fig. 3B).

Alluvium

Older Alluvium. A relatively thin layer (0 to ∼36 m) of older alluvium locally unconformably overlies, and is typically in angular discordance with, the generally eastward dipping Humboldt Formation (see Appendix 2 for a description of the older alluvium). We infer that the older alluvium probably largely represents post-tectonic sedimentation because it has not been tilted and it overlaps nearly all tectonic features in or bounding the Knoll basin, such as the Knoll Mountain fault, intrabasin faults, and the Hice syncline (Fig. 3A).

Modern Alluvium and Stream Systems. The youngest unit in the Knoll basin comprises Quaternary alluvium along the streams of two modern drainage systems. The drainage systems consist of the north-flowing Snake River drainage and the east-flowing Great Salt Lake drainage, with a drainage divide transecting the southernmost part of the Knoll basin (Fig. 3A). Streams of these drainage systems have deeply incised the Humboldt Formation and older alluvium.

Intrabasin Faulting, Tilting, and Folding of the Humboldt Formation

Stratigraphic and crosscutting relationships indicate that the intrabasin fault zone and the Hice syncline formed after initiation of extension and subsidence in the basin, and that they are responsible for intrabasin folding and tilting of strata.

Intrabasin Fault Zone. The intrabasin fault zone is broadly composed of three west-dipping, left-stepping normal fault segments. Two of the segments compose the Henry fault and are linked by a transfer zone in Eagle Draw, and the third segment is the Summer Camp fault in the southern part of the basin (Fig. 3). The intrabasin fault zone coincides approximately with the eastern margin of a localized increase in thickness of basin fill (Fig. 2B). We suggest that this localized maximum probably reflects differential subsidence and consequent preservation of a greater thickness of the Humboldt Formation in the hanging wall of the intrabasin fault system.

Strata in proximity of the intrabasin faults, both in the hanging wall and footwall, are typically rotated to the east, suggesting that eastward tilting of the Humboldt Formation in the center of the basin is related to slip along these faults (see cross-sections A-A′ and B-B′ in Figs. 3B and 3C). The Henry and Summer Camp faults are overlapped by the older alluvium, indicating that they formed before the older alluvium (Fig. 3). However, a small fault scarp cutting the older alluvium is present adjacent to the northernmost extent of the Henry fault, suggesting that some localized fault slip along this zone postdates or was synchronous with deposition of the older alluvium (Fig. 3A). In Eagle Draw, faults in the intrabasin fault zone offset and rotate most of the Humboldt Formation, including strata as young as the Bloody Gulch member (Fig. 3B). Therefore this fault zone appears to have formed late in the history of the basin, mostly before deposition of the older alluvium but after deposition of strata in the Bloody Gulch member.

Hice Syncline. The simple conformable sequence of the lower members of the Humboldt Formation in the central and southern part of the Knoll basin is in sharp contrast to the Humboldt Formation in the northeastern corner of the basin, which defines the Hice syncline and contains three intraformational unconformities related to synextensional faulting, folding, and valley incision (cf. cross sections in Figs. 3, 4, and 6). The Hice syncline is a fault-propagation fold related to slip along, and emergence of, the Hice-Valder normal fault, which is a synthetic fault adjacent to the Knoll Mountain fault (Deibert and Camilleri, 2006; Fig. 4). The lowest four members of the Humboldt Formation are folded, including an incised valley (demarcated by unconformity 3) and its fill (Cave member) located along the fold hinge region (Fig. 6D). Structural analysis of the strata in the syncline (Deibert and Camilleri, 2006) indicates that (1) prior to deposition of the Knoll member, the Blanchard member was rotated to the east as much as ∼17° and then eroded, resulting in development of angular unconformity 1 between the Knoll and Blanchard members (Figs. 6B and 6C); (2) the Hice syncline began to develop during the deposition of the Knoll member and resulted in development of angular unconformity 2 (Fig. 6B); and (3) cessation of folding occurred sometime after deposition of the Bloody Gulch member, which is folded (Fig. 6B). Furthermore, stream valley incision took place during fold development with localization of incision being influenced by the developing syncline’s structurally low hinge region (Deibert and Camilleri, 2006; Fig. 4A and 6A).

Thousand Springs Basin

The Thousand Springs basin contains two primary tectonic structures that controlled its development. These are the east-dipping faults bounding the Thousand Springs basin, which include the Black Rock and Bell Canyon faults, and an east-dipping intrabasin fault zone in the middle of the basin (Figs. 2 and 3). In addition, a relatively young fault at the southern end of Knoll Mountain offsets older alluvium (Fig. 3A).

The basin fill in the Thousand Springs basin is similar to that in the Knoll basin in terms of stratigraphy, lithology, and depositional environments of the Humboldt Formation, but it is less well exposed. Nonetheless, assessment of moderately well exposed Humboldt Formation adjacent to the Bell Canyon fault in Bell Canyon (Figs. 3A and 7) indicates that it was deposited during slip along the faults bounding Knoll Mountain and therefore is syntectonic. The sequence in Bell Canyon consists mostly of primary and fluvial-reworked tuff punctuated by sparse rock-fall avalanche deposits (Fig. 7). The rock-fall avalanche deposits are beds of breccia 2–10.5 m thick with pebble- to boulder-sized clasts of highly fractured Paleozoic chert and siltstone (e.g., Fig. 7). The clasts in the rock-fall avalanche deposits are identical to Paleozoic rock units exposed in the nearby footwall of the Bell Canyon fault. Consequently, we interpret the avalanche deposits to have been derived from footwall rocks that were exposed along an active fault scarp generated by slip on the Bell Canyon fault.

The age of the intrabasin faults in the Thousand Springs basin relative to the Bell Canyon and Black Rock faults is poorly constrained. However, the intrabasin faults cut and rotate the Humboldt Formation in their hanging walls and footwalls (Fig. 3A), suggesting that this fault zone formed late in the history of the basin.

AGE OF THE HUMBOLDT FORMATION IN THE KNOLL MOUNTAIN AREA

We utilized tephrochronology and biostratigraphy to determine the age of the Humboldt Formation in the Knoll and Thousand Springs basins in order to constrain the timing of basin evolution and associated faulting and folding. Tephra correlations involved petrographic and geochemical assessment of volcanic glass shards conducted at the University of Utah using the methods detailed in Perkins et al. (1995, 1998) and Perkins (2014). We collected 33 tephra samples; their correlations, ages, and sample locations are presented in Table 2 and are shown on geologic maps and cross sections in Figures 3, 4, 79. The geochemical data from tephra samples are shown in Table 3. In addition, to further refine the age of the Humboldt Formation, we integrated our tephrochronologic data with age constraints from North American Land Mammal stages and fossil data from the Knoll basin reported by Macdonald (1949).

Knoll Basin

An integration of tephrochronologic and mammalian fossil data indicates that the age of the Humboldt Formation in the Knoll basin is ca. 16 Ma to at least 8.3 Ma and possibly as young as ca. 5 Ma (Barstovian Stage to at least Hemphillian Stage; Table 2; Figs. 10 and 11). These data also provide important age constraints on the members of the Humboldt Formation.

Blanchard, Knoll, Cave, and Bloody Gulch Members

Our tephrochronologic data in the Knoll basin (Table 2) are from the Blanchard, Knoll, Cave, and Bloody Gulch members in three areas. Knoll Creek in the northeastern corner of the basin (Fig. 4A), Eagle Draw in the center of the basin (Fig. 8), and Red Point in the southern part of the basin (Fig. 9). Collectively, these data indicate that the Blanchard member is ca. 16 Ma to 9.8 Ma; the Knoll member is ca. 9.8 Ma; the Cave member (incised valley fill) is 9.8 to ca. 9.6 Ma; and the Bloody Gulch member is 9.6 to at least 8.3 Ma (Table 2). Note that both the Knoll and Cave members contain the same 9.8 Ma Hazen tephra (Table 2), indicating that the incised valley fill (Cave member; Figs. 4 and 6) was derived mostly from the underlying and adjacent Knoll member. Our tephrochronologic data also indicate that a distinctive welded tuff, which can be traced nearly across the basin in the Blanchard member, is correlative with the 11.03 Ma Cougar Point Tuff XIII (Figs. 3A and 3B). This welded tuff is important in correlating rock and time units because it an isochronous datum recognized in many areas within and outside of the Knoll basin. This tuff has been correlated with the Jackpot Member of the Rogerson Formation and the Big Bluff Member of the Cassia Formation, which are present north of the Knoll basin along the southern margin of the Snake River Plain (Ellis et al., 2012; Knott et al., 2016a, 2016b; Supplemental Item [see footnote 1]). An undated welded unit of the Cougar Point Tuff in the Blanchard member crops out intermittently just to the east of the Henry fault (Fig. 3A), and although its age is unknown, it indicates that the ages of strata in this area are in the range of ca. 12.7–10.5 Ma (Perkins et al., 1995, 1998; Perkins and Nash, 2002; Bonnichsen et al., 2008), the age span of the Cougar Point Tuff.

Within the Humboldt Formation in the Eagle Draw area, Macdonald (1949) reported the occurrence of mammalian fauna whose ranges of genera overlap in the Clarendonian Stage (Fig. 11) and he assigned a Clarendonian age to the strata in this area. To better constrain the age of the Humboldt Formation we have taken Macdonald’s (1949) data with some minor updates and combined it with our tephrochronologic age constraints and Tedford et al. (2004) numerical ages of the North American Land Mammal stage boundaries (Fig. 11). Our tephrochronology corroborates Macdonald’s (1949) findings, but also indicates that some of the Humboldt Formation is Hemphillian in age. Macdonald’s (1949) fossil samples are stratigraphically from the upper part of the Blanchard member, the lower part of the Bloody Gulch member, and the Eagle Flat member (Figs. 3A, 3B, and 8). The fossils recovered from the Blanchard member are stratigraphically near the undated welded unit of the 12.7–10.5 Ma Cougar Point Tuff (see locations F6–F8 in Fig. 3A), consistent with a Clarendonian age (Fig. 11). Fossil samples in the Bloody Gulch member are in a stratigraphic interval that contains 9.8–9.3 Ma tephra, which also indicate a Clarendonian age (Figs. 8 and 11). However, our tephrochronologic data indicate that at least some of the Bloody Gulch member is ca. 8 Ma in age and therefore the Bloody Gulch member extends into the Hemphillian.

Eagle Flat Member

We do not have tephrochronologic data from the Eagle Flat member; however, mammalian fossils coupled with our tephrochronology from the Bloody Gulch member suggest that it is Hemphillian in age (Fig. 11). The youngest tephra from the Bloody Gulch member is 8.3 Ma (Hemphillian), and it provides a maximum age limit for the Eagle Flat member. A specimen of Peromyscus antiquus collected from near the top of the Eagle Flat member further indicates that most of this member cannot be younger than 4.8 Ma because the range of this fossil does not extend beyond the ca. 4.8 Ma Hemphillian-Blancan stage boundary (fossil location F1 in Fig. 11). Collectively, these observations indicate that the age of Eagle Flat member is younger than 8.3 Ma and is mostly older than 4.8 Ma (Fig. 11), but could be younger (Blancan) because chronologic data from the top of the member are not available.

Thousand Springs Basin

Three tephra samples from the Thousand Springs basin were collected from strata in the hanging wall of the Bell Canyon fault in Bell Canyon (Fig. 7). The ages of these samples indicate that the Humboldt Formation in this area ranges from 11.7 to 10.7 Ma. This tephra includes the 10.7 ± 0.1 Ma Ibex Peak 19, 11.4 ± 0.1 Ma Cougar Point Tuff IX, and 11.7 ± 0.1 Ma Cougar Point Tuff XI tephras (Table 2).

AGE CONSTRAINTS ON TECTONIC FEATURES IN THE KNOLL MOUNTAIN AREA

Knoll Mountain Fault

The age of the Knoll Mountain fault can be inferred from the collective ca. 16 to 8.3 Ma ages of the Blanchard, Knoll, Cave, and Bloody Gulch members (Fig. 10), which were deposited during slip along the Knoll Mountain fault (Deibert and Camilleri, 2006). We therefore infer that extension along the Knoll Mountain fault broadly began ca. 16 Ma and ceased after 8.3 Ma. In addition, our data also indicate that one or more episodes of rotation and erosion of the Blanchard member adjacent to the Knoll Mountain fault in the northeastern corner of the basin took place between ca. 11 and 9.8 Ma (Fig. 10). In this area, strata of the Blanchard member below angular unconformity 1 underwent eastward rotation prior to deposition of the Cave member above the unconformity (Figs. 6B and 6C; Deibert and Camilleri, 2006). We attribute rotation of the Blanchard member to reverse drag along the Knoll Mountain fault, and so it reflects a period, or periods, of slip. Our data indicate the presence of the 9.8 Ma Hazen tephra in the Cave member above the unconformity, implying that rotation of the Blanchard member must predate 9.8 Ma. Furthermore, the local preservation of the distinctive 11.03 Ma welded Cougar Point Tuff XIII immediately below unconformity 1 provides the youngest age of the Blanchard member beneath the unconformity and therefore a maximum age limit on rotation (Fig. 4). Consequently, the rotation of the Blanchard member, reflecting an episode or episodes of slip along the Knoll Mountain fault, took no longer than ∼1 m.y. between ca. 11.03 and 9.8 Ma. We note, however, that because ca. 11–9.8 Ma strata are not exposed in proximity of the Knoll Mountain fault south of the northeastern corner of the Knoll basin, it is unclear if this slip event, or events, produced rotation of hanging-wall strata elsewhere.

Intrabasin Fault Zone

Our tephrochronologic data coupled with geologic mapping suggest that movement along the intrabasin fault zone, and filling of its hanging-wall basin with the Eagle Flat member, largely took place between 8.3 and 4.8 Ma (Fig. 10). The best age constraints for the intrabasin fault system come from the center of the fault zone in the Eagle Draw area. In this area, 8.3 Ma tephra of the Bloody Gulch member is cut and rotated by the Henry fault, indicating that slip postdates 8.3 Ma (see cross-section A-A′ in Fig. 3B), and the upper age limit of 4.8 Ma of the upper part of the Eagle Flat member suggests that much of the slip predates 4.8 Ma. Furthermore, because the syntectonic Eagle Flat member records southward drainage (Appendix 1), the intrabasin fault zone can also be inferred to predate the reversal of drainage to a north-draining system in the basin; on the basis of detrital zircon studies, Beranek et al. (2006) documented this reversal of drainage to have occurred prior to ca. 3 Ma. In summary, available data indicate that the intrabasin fault zone was largely active between at least ca. 8 and 5 Ma, but was likely mostly inactive by ca. 3 Ma.

Hice Fault-Propagation Syncline and Valley Incision

The tephrochronologic data from strata in the Hice syncline (Fig. 4A) provide some constraints on the timing of development of unconformities in the syncline and the Hice-Valder fault. Inception of the development of the Hice-Valder fault and related Hice syncline, as well as angular unconformity 2, took place during deposition of the Knoll member, which is constrained to be ca. 9.8 Ma (Table 2). Stream incision along the axis of the syncline and subsequent filling of the valley with the Cave member must postdate the 9.8 Ma Knoll member, but predate a 9.6 Ma tuff near the base of the Bloody Gulch member that overlaps the incised valley and its fill (sample location 1 in Fig. 4A; sketch and photo in Fig. 6D). These data indicate that incision and filling of the valley took place in ∼200 k.y. or less between 9.8 and 9.6 Ma. Furthermore, the 9.6 Ma tephra also helps constrain the timing of folding in the Hice syncline because it, and the remainder of the overlying Bloody Gulch member, is folded (Fig. 4). In summary, synclinal folding and slip along the Hice-Valder fault began ca. 9.8 Ma and continued to at least 9.6 Ma (Fig. 10).

Bell Canyon Fault and Thousand Springs Basin

Ages of the tephra samples collected from the synorogenic deposits of the Thousand Springs basin imply that slip along the Bell Canyon fault and basin subsidence began before 11.7 Ma and ceased sometime after 10.7 Ma (Fig. 7).

PALEOGEOGRAPHY OF THE KNOLL MOUNTAIN REGION

We synthesize our geologic mapping, stratigraphic, structural, and tephrochronologic data to present a paleogeographic reconstruction of the Knoll Mountain region from ca. 16 Ma to the present (Fig. 12). The reconstruction utilizes the paleogeography of the northeastern corner of the Knoll basin (from Deibert and Camilleri, 2006) and extends it to include new data from the southern and western parts of the basin as well as from the northwestern part of the Thousand Springs basin. Because the history of the Thousand Springs basin is less well known, our reconstruction focuses on the Knoll basin. In addition, although it is clear that slip along the Knoll Mountain fault and subsidence in the Knoll basin began ca. 16 Ma, the age of inception of faulting on the east side of Knoll Mountain can only be constrained to predate 11.7 Ma. However, south of Knoll Mountain, Mueller et al. (1999) documented that slip on the faults bounding the western margin of the southern part of the Thousand Springs basin began by ca. 15 Ma, and probably as early as 16 Ma (Table 1), and hence we have assumed that faulting began at approximately the same time to the north. In addition, we infer that Knoll Mountain and its drainage divide had not formed prior to ca. 16 Ma, in concert with the Henry (2008) and Henry et al. (2012) observation that this area was situated to the east of a regional north-trending paleodrainage divide (Fig. 1A) and was transected by east-trending and east-flowing paleodrainages. It is therefore unlikely that a drainage divide existed in this area until the emergence of Knoll Mountain ca. 16 Ma, indicated by the development of the Knoll basin at that time.

Paleogeography from ca. 16 to 11 Ma: Inception of Extension and Relative Rise of the Knoll Mountain Horst

Our reconstruction begins with the emergence of the Knoll Mountain horst along the Knoll Mountain and Bell Canyon–Black Rock faults concomitant with inception of subsidence in the Knoll and Thousand Spring basins, which were characterized by vegetated alluvial plains with braided streams (Fig. 12A; Deibert and Camilleri, 2006). The uplift of Knoll Mountain resulted in the establishment of a regional north-northeast–trending drainage divide with the Knoll basin overall draining to the south. The Contact pluton was a topographic high that formed the northern margin of the basin. Both the Contact pluton and Knoll Mountain, and its drainage divide, were south of a regional east-northeast–trending drainage divide that developed on top of the tumescent topographic bulge of the migrating Yellowstone hotspot (Fig. 1B; Beranek et al., 2006).

Paleogeography from ca. 11 to 9.8 Ma: Development of a Tectonically Generated Unconformity and a Perennial Lake

Between ca. 11 and 9.8 Ma, erosion and as much as 17° of eastward stratal rotation of the Blanchard member took place locally in the northeastern corner of the Knoll basin, probably related to slip along the proximal Knoll Mountain fault (Deibert and Camilleri, 2006; Fig. 12B). Then, ca. 9.8 Ma, a large perennial freshwater lake was established and the Hice syncline began to form in advance of the blind Hice-Valder fault (Fig. 12C). It was during this time period that deposition of lacustrine sediment of the Knoll member on top of the eroded and rotated Blanchard member in the northeastern corner of the basin produced unconformity 1 (Figs. 6B and 6C).

Paleogeography from ca. 9.8 to 9.6 Ma: Lake Shoreline Regression and Valley Incision

Between ca. 9.8 and 9.6 Ma the lake shoreline in the northeastern part of the basin regressed with a southward-draining incised valley developing in its wake along the topographically low hinge of the Hice syncline (Fig. 12D). This was followed by transgression of the lake concomitant with filling of the valley with fluvial and eolian deposits composed largely of reworked lacustrine sediment (Fig. 12E). The stream incision and filling of the valley produced unconformity 3 (Fig. 6D).

Paleogeography from ca. 9.6 to 8.3 Ma: Return to an Alluvial Plain Environment

Between ca. 9.6 and 8.3 Ma the Knoll basin was again characterized by an alluvial plain with braided streams, and slip along the Knoll Mountain fault, as well as folding of sediment along the Hice syncline, continued (Fig. 12F).

Paleogeography from after 8.3 Ma to before ca. 4.8 Ma: Development of the Intrabasin Fault System

The period between 8.3 and 4.8 Ma is characterized by the full development of the intrabasin fault system. This resulted in the migration of the basin depocenter to the hanging wall of the Henry fault, and with it deposition of sediment of the Eagle Flat member (Fig. 12G). Slip along the intrabasin fault system resulted in rotation of strata in its footwall and hanging wall (see cross sections in Figs. 3B and 3C). It is uncertain if the faults bounding Knoll Mountain were still active during this time.

We infer that the Snake River drainage divide had not yet migrated into the Knoll basin because paleocurrent data from the Eagle Flat member (see Appendix 1) suggest that drainage in the basin was still to the south during this time, and detrital zircons from the Contact pluton and environs had not yet reached the Snake River Plain (Link et al., 2002, 2005; Beranek et al., 2006).

Paleogeography from ca. 4.8 Ma to before 3 Ma: Cessation of Faulting, Deposition of Older Alluvium(?), and Drainage Reversal

The period of time following deposition of the Eagle Flat member and ca. 3 Ma marks a fundamental time of transition that involved cessation of faulting, and ultimately drainage reversal in the Knoll basin (Fig. 10). During this time, tectonism waned and was accompanied and/or followed by widespread basin erosion. The drainage direction and outlets for the drainage system that produced this period of basin-wide erosion are unknown. Erosion was followed by deposition of the older alluvium (Fig. 12H) and migration of the Snake River drainage divide into the Knoll basin, which transformed the long-lived south-flowing drainage system into a north-flowing drainage system. We note, however, that the timing of drainage reversal relative to deposition of the older alluvium is uncertain, but reversal must have occurred within or between the time frames represented by the paleogeographic maps in Figures 12G and 12H. That is, drainage reversal (1) may be associated with the basin-wide erosion event and so predates the older alluvium, (2) may have begun during deposition of the older alluvium, or (3) may entirely postdate deposition of the older alluvium (Fig. 12H). Although the relative timing of drainage reversal is uncertain with regard to the older alluvium, a numerical age estimate of the timing of reversal can be inferred from detrital zircon data from the Snake River Plain. Link et al. (2002, 2005) and Beranek et al. (2006) indicated that drainage reversal along Salmon Falls Creek at the northern end of the Knoll basin (Fig. 3A) must have occurred prior to 3 Ma because detrital zircons from the Contact pluton and Eocene–Oligocene volcanic rocks, which are exposed on the western margin of the Knoll basin (McFarlane, 2001), make their first appearance in sediment deposited in the Snake River Plain in the Glenns Ferry Formation by 3 Ma. Consequently, we have shown migration of the Snake River drainage divide to the southern end of Knoll basin by 3 Ma (Fig. 12H), and with it, a reversal of drainage direction during or following deposition of the older alluvium. The time period represented by Figure 12H therefore represents a transition between cessation of deposition of the older alluvium and the inception of widespread incision of the older alluvium and parts of the Humboldt Formation by the modern Snake River drainage system.

Paleogeography after ca. 3 Ma to the Holocene: Incision of the Older Alluvium and Stream Capture in the Southern Part of the Knoll Basin

Sometime following drainage reversal, and continuing to the present, basin-wide erosion and incision has resulted in northward transport of sediment out of the basin via the Snake River drainage system and reexposure of the Humboldt Formation (Fig. 12I). This time period is also characterized by a lack of faulting, with the exception of a fault scarp that cuts the older alluvium on the southeast side of Knoll Mountain that is inferred to be younger than 15 ka (dePolo, 2008) and a fault scarp of uncertain age that displaces the older alluvium along the northern end of the Henry fault (Fig. 3A). The last event to occur during this time period was the capture of north-draining streams in the southernmost part of the Knoll basin by the east-flowing Salt Lake drainage system and consequent westward migration of the drainage divide (Fig. 12J; additional data regarding stream capture are in the Supplemental Item [footnote 1]).

REGIONAL CORRELATION OF MIDDLE MIOCENE TO HOLOCENE BASINS AND FAULTS IN NORTHEAST NEVADA–SOUTH CENTRAL IDAHO

In this section we establish correlations between basins and fault systems in the Knoll Mountain area with similar, adjacent features to the north and south. These correlations are fundamental to developing a regional middle Miocene to Holocene tectonic and paleographic reconstruction of northeastern Nevada–south-central Idaho.

Relation of Knoll Mountain Faults to the Faults Bounding the Rogerson Graben on the Southern Margin of the Snake River Plain

Knoll Mountain faults terminate in and around the Jurassic Contact pluton; between the pluton and the Snake River Plain, a different set of Miocene faults are present (Fig. 13A). These faults define the ∼30-km-long, north-trending Rogerson graben, the northern end of which merges with the Snake River Plain (Fig. 13A). The Rogerson graben developed between ca. 10 and 8 Ma during a time of maximum hotspot volcanism and subsidence of the Snake River Plain (Bonnichsen and Godchaux, 2002; Andrews et al., 2008; Knott et al., 2016b). The graben is bounded on the west by the master fault, the Browns Bench fault, and on the east by the Shoshone Hills fault (Fig. 13A; Bonnichsen and Godchaux, 2002; Andrews et al., 2008; Knott et al., 2016b). The graben has a protracted history with the Shoshone Hills fault, forming at 10.3–10.1 Ma, followed by the development of the Browns Bench fault between 10.1 and 8 Ma, with intermittent activity until the Pliocene (Andrews et al., 2008; Knott et al., 2016b). Subsidence in the Snake River Plain in the Rogerson graben area initiated between 10.59 and 10.34 Ma and is reflected by the development of a regional east- to east-northeast–trending flexural monocline that involved downflexing of its northern limb concurrently with deposition of volcanic units that onlapped and thinned on the growing monocline until at least 8 Ma (Fig. 13A; Knott et al., 2016a, 2016b).

The faults in the Rogerson graben are similar to those in the Knoll Mountain area in terms of their orientation and timing. The Browns Bench and Shoshone Hills faults are temporally correlative with some of the youngest faulting in the Knoll basin, including the latter stages of slip along the Knoll Mountain fault (Fig. 13B). Perhaps the most striking structural similarity is between the Browns Bench fault and the intrabasin fault system in the Knoll basin. Both are broadly along strike, were active between ca. 8 and 5 Ma, and although opposite in polarity, they both terminate in proximity of the Contact pluton (Fig. 13A). It appears that the area of the pluton may have served as, and influenced the location of, an accommodation zone from at least 10 to 8 Ma. It is unknown if such an accommodation zone existed prior to 11–10 Ma because the Rogerson graben region is covered primarily by 12 Ma and younger volcanic rocks that conceal evidence of any older extension (Andrews et al., 2008; Knott et al., 2016b).

Correlation of Knoll Mountain’s Faults with Fault Systems in and Near the Ruby Mountains–East Humboldt–Wood Hills Metamorphic Core Complex Area

The faults and basins adjacent to Knoll Mountain can be correlated with Miocene basins and faults that are to the south of, and are along strike with, those bounding Knoll Mountain. Specifically, the Knoll Mountain faults and basins appear to be correlative with coeval structures bounding the Windermere Hills and the Ruby Mountains–East Humboldt–Wood Hills metamorphic core complex to the south (Fig. 2). We propose that the now-separated Knoll and Ruby basins (Fig. 2) were once part of a large middle to late Miocene basin system bounded by a north-northeast–trending, >190-km-long fault system consisting of the Ruby–East Humboldt detachment and the Knoll Mountain fault and that, at the same time, the Thousand Springs basin was bounded by a separate regional east-dipping fault system.

Thousand Springs Basin and the Thousand Springs Fault System

The east-dipping faults that define the eastern margin of Knoll Mountain are along strike and in proximity of similar east-dipping faults that bound the eastern margin of the Windermere Hills (Fig. 2). Mueller et al. (1999) documented that deposition of the Humboldt Formation in the hanging walls of the east-dipping faults bounding the Windermere Hills was coeval with faulting from ca. 16 to 7 Ma (Table 1), consistent with our age constraints on fault slip beginning before 11.7 Ma and continuing after 10.7 Ma to the north. Consequently, because the east-dipping faults along the eastern flanks of the Windermere Hills and Knoll Mountain are geometrically similar, along strike and in proximity of each other, and share similar age constraints, we interpret them as being segments of the same fault system, which we refer to as the Thousand Springs fault system.

In summary, the Thousand Springs fault system forms the western margin of the middle Miocene Thousand Springs basin, which we infer was active from ca. 16 to 7 Ma. Collectively, our mapping and that of Mueller et al. (1999) indicates that this fault system terminates to the north in the vicinity of the Contact pluton and extends southward between the Pequop Mountains and northernmost Wood Hills (Fig. 2). However, it is possible that this fault zone may have extended farther southward toward Spruce Mountain, but is concealed beneath Quaternary valley fill to the east of the Wood Hills (shown queried in Fig. 2A). We note that there are two lobes of relatively thick basin fill (see Fig. 2B) that are on strike with the Thousand Springs basin that could partly reflect the Humboldt Formation in the hanging wall of a concealed part of the Thousand Springs fault system and partly reflect basin fill in the hanging wall of the modern west-dipping fault bounding the Pequop Mountains. Unfortunately, whether the Thousand Springs fault system extended farther southward is unknown because seismic reflection and well data for the valley between the Wood Hills and Pequop Mountains are not available.

Knoll-Ruby Basin and the Knoll-Ruby Fault System

The Miocene Knoll and Ruby basins project into one another, are both bounded by west-dipping faults (Fig. 2) in proximity of each other, and they share similar histories. Here we establish a correlation between these basins and their associated fault systems. However, in order to establish this correlation, we first define the extent and history of the Ruby basin.

Former Extent of the Ruby Basin and Bounding Faults. Although extensive exposures of the Humboldt Formation representing the Ruby basin are present west of the Ruby Mountains, East Humboldt Range, and northern Snake Mountains (e.g., Wallace et al., 2008), these exposures do not reflect the entire extent of the Ruby basin and its bounding faults. This is because most of the eastern margin of the Ruby basin is dismembered by modern normal faults (e.g., Colgan et al., 2010), leaving only remnants of the basin and its detachment faults preserved in the hanging walls of these younger faults. Furthermore, the detachment and basin fragments that are east of the modern Ruby Mountains and East Humboldt Range are mostly concealed by Quaternary sediment filling the modern basins. However, on the basis of seismic reflection and well data, a fragment of the Ruby–East Humboldt detachment and Humboldt Formation is present in the subsurface due east of the Ruby Mountains (Fig. 2A; Satarugsa and Johnson, 2000), and there is a small area of surface exposure of Humboldt Formation between the East Humboldt Range and Spruce Mountain (Fig. 2A; Hope, 1972). A rough approximation of the former eastern margin of the fault-bounded Ruby basin can be inferred by assessing the areal distribution of the Humboldt Formation and remnants of the Ruby–East Humboldt detachment. We have placed the inferred basin margin just east of the easternmost extent of the present distribution of basin and detachment remnants (Fig. 2). The position of the fault-bounded basin margin is a relative position that assumes that if the footwall rocks are held fixed, the original breakaways must have been relatively east of the easternmost exposures of the fault and or basin fill. We also infer that the basin margin probably extended to the Windermere Hills (as envisioned by Mueller and Snoke, 1993a), although very little basin fill is preserved in that area (Fig. 2A).

We note that the inferred margin of the Ruby basin makes an abrupt east-trending bend in the southern Ruby Mountains region (Figs. 2A and 13A). South of the bend, the placement of the basin margin was based on the Colgan et al. (2010) cross-section reconstructions of the Ruby–East Humboldt detachment. The eastward jog in the basin margin is required by the presence of the detachment and basin fill in the subsurface east of the Ruby Mountains (e.g., Satarugsa and Johnson, 2000; Fig. 2A). This bend coincides with an east-trending lineament defined by a line of exhumed Mesozoic plutons and mountain range terminations (Fig. 13A). Consequently, this bend may represent an accommodation zone possibly influenced by pre-Cenozoic crustal anisotropy or structure. In summary, on the basis of cross sections and reconstructions of the Ruby–East Humboldt fault system by Colgan et al. (2010) and Mueller and Snoke (1993a), coupled with the areal distribution of the remnants of the Humboldt Formation once forming its basin, it appears that the trace of the Ruby–East Humboldt fault system (and the eastern margin of the Ruby basin) had an overall north-northeast trend and was mostly located east of the modern Ruby–East Humboldt Range (Fig. 2).

History of the Ruby Basin. The Ruby basin contains two major stratigraphic units separated by an unconformity (Wallace et al., 2008; Fig. 13B). These units include the ca. 16 to 8 Ma Humboldt Formation (Table 1) that was deposited during slip along the Ruby–East Humboldt detachment and northern Snake Mountains faults, and an overlying 3–2 Ma fluvial-lacustrine unit representing deposits of Pliocene Lake Elko (Fig. 2; Reheis et al., 2003, 2014; Wallace et al., 2008). Wallace et al. (2008) reported that the unconformity separating these two units reflects widespread erosion beginning ca. 9.8 Ma. The ca. 9.8 Ma age of inception was based on the oldest known Humboldt Formation in the Ruby basin at the time; however, on the basis of additional and more recent age information, we extend this to sometime between ca. 8 and 3 Ma (Table 1; Fig. 13B).

From ca. 16 to 8 Ma the Ruby basin retained sediment and was bounded on the east by the Ruby–East Humboldt detachment and northern Snake Mountains faults (Wallace et al., 2008; Fig. 2), and late in its history it appears to have developed intrabasin faults. Local preservation of the basin fill and the detachment in the northern East Humboldt Range indicates the presence of synthetic intrabasin faults that sole into the Ruby–East Humboldt detachment and cut and rotate basin fill as young as 8 Ma (Mueller, 1992; Mueller and Snoke, 1993a, 1993b; McGrew and Snoke, 2015). One of these faults is shown schematically in Figure 2A and is labeled as an intrabasin fault (e.g., see cross-section B-B′ of McGrew and Snoke, 2015). Because these synthetic intrabasin faults cut strata as young as 8 Ma they probably formed late in the history of the Ruby basin.

Beginning sometime after ca. 8 Ma the basin underwent widespread erosion. Wallace et al. (2008) suggested that erosion was a result of integration of streams to form the southwest-flowing Humboldt River system; they inferred that the Humboldt River system developed in response to a reduction in elevation, and therefore base level, along the western margin of the Basin and Range Province due to active extension.

The next major event affecting the Ruby basin was uplift, exhumation, and dismemberment of the Ruby–East Humboldt detachment and the Humboldt Formation in the Ruby basin by a younger, modern generation of faults (Fig. 2). These faults postdate the youngest rotated Humboldt Formation, which is ca. 8 Ma (Mueller, 1992; McGrew and Snoke, 2015). We suggest that most of these faults formed after 8 Ma, but before deposition of the 3–2 Ma fluvial-lacustrine sequence. This inference is based on work of Wallace et al. (2008) that shows that the fluvial lacustrine sequence is restricted to the hanging wall of the younger faults (see Fig. 2), which suggests that the topographically uplifted footwall, i.e., the modern East Humboldt Range and southern Snake Mountains, formed the eastern margin of the basin during lacustrine deposition. Furthermore, McGrew and Snoke (2015) noted that the younger generation of faults responsible for uplift of the northern end of the East Humboldt Range formed before 2.5 Ma, prior to deposition of locally derived Quaternary older alluvium. In summary, we infer that slip along the modern faults bounding the Ruby Mountains and East Humboldt Range began prior to deposition of the ca. 3–2 Ma fluvial lacustrine sequence. Consequently, we infer that between ca. 8 and 3 Ma, during a time when the basin was undergoing predominantly erosion, the Ruby–East Humboldt fault system went extinct followed by formation of a younger generation of faults (Fig. 13B).

Correlation of the Knoll and Ruby Basins and Their Bounding Faults. Because the eastern margin of the Knoll basin and the inferred eastern margin of the Ruby basin project into one another (Fig. 2A), and their west-dipping basin-bounding faults and associated basins are coeval and in proximity of each other, we infer that the Ruby–East Humboldt detachment and Knoll Mountain faults were part of the same fault system and that their basins were likely integrated. Furthermore, on the basis of the geographic extent of the west-dipping faults that bound the Knoll and Ruby basins (Fig. 2), this fault system extended for at least 190 km. The dip angles of faults within this fault system are predominantly low angle. For example, the dip angle of most faults bounding Knoll Mountain range from ∼24° to 40° and remnants of the Ruby–East Humboldt detachment in the Wood Hills and East Humboldt Range dip only a few degrees to the west, but were likely rotated to a low-angle during extension (e.g., Mueller and Snoke, 1993a; Camilleri and Chamberlain, 1997; McGrew and Snoke, 2015). The Ruby–East Humboldt detachment in the northern Ruby Mountains is inferred to have an original dip between ∼24° and 45° (Haines and van der Pluijm, 2010), and in contrast, Colgan et al. (2010) inferred that the detachment initially formed with a dip of ∼57°, but was rotated to a low angle due to isostatically induced footwall rebound. Collectively, these observations suggest that fault dip may have varied from low to high angle. In addition, because only spatially isolated remnants of these faults are preserved, it is unclear whether this fault system may have been segmented; however, most major modern normal fault systems are segmented (e.g., Machette et al., 1991; dePolo et al., 1991; Crone and Haller, 1991) and therefore it is reasonable to assume that the Ruby–East Humboldt detachment was as well. In fact, low- and high-angle fault geometries appear to be characteristic of some modern segmented normal-fault systems (e.g., Smith and Bruhn, 1984; Morley, 2014).

Although the Knoll and Ruby–East Humboldt faults probably formed a segmented fault system, we infer that the Knoll and Ruby basins were likely connected. However, this is tentative because only one small remnant of the Humboldt Formation is preserved between the Knoll and Ruby basins in the area between the Windermere Hills and southern Snake Mountains (Coats, 1987). The lack of preservation of basin fill in this area may be related to the dismemberment of the Ruby basin by the modern normal faults coupled with erosion on their relatively uplifted footwalls. Consequently, we cannot decisively establish that Knoll and Ruby basins were the same continuous basin during the span of deposition of the Humboldt Formation (ca. 16 to before ca. 5 Ma). However, our sedimentologic analysis of the Humboldt Formation in the Knoll basin indicates that it was part of a hydrographically open freshwater system that flowed to the south. This, coupled with the occurrence of the Humboldt Formation between the southern Snake Mountains and the Windermere Hills (Fig. 2A), is suggestive that, for at least part of the time, the Knoll and Ruby basins were connected as a continuous depositional basin. It is possible that for some of the time the Knoll basin was a separate coeval subbasin that retained most of the sediment it received, but was connected to the Ruby basin by a stream.

In summary, we infer that the Knoll and Ruby basins were integrated and their basin-bounding faults were part of the same, probably segmented, fault system, which we refer to as the Knoll-Ruby fault system and Knoll-Ruby basin (Figs. 14A and 14B). The evolution of the Knoll-Ruby basin also involved the development of intrabasin faults (e.g., Henry, Summer Camp, and other faults labeled intrabasin faults in Fig. 2A) that can be constrained to have formed or been active late in the history of the basin, after ca. 8 Ma. Overall, slip along the Knoll-Ruby fault system, and subsidence and retention of sediment in the Knoll-Ruby basin in its hanging wall, took place from 16 to at least 8 Ma, but ceased sometime between ca. 8 and 3 Ma, prior to development of the younger generation of faults (Fig. 13B). In addition, our establishment of the Knoll-Ruby fault system being the master bounding fault system for the Knoll-Ruby basin indicates that the northern Snake Mountains fault (Fig. 2A) is a large synthetic fault rather than part of the master fault.

SYNTHESIS OF MIDDLE MIOCENE TO HOLOCENE TECTONICS AND PALEOGEOGRAPHY OF NORTHEASTERN NEVADA–SOUTH CENTRAL IDAHO

We present a simplified regional picture of tectonism and paleogeography that begins before 17 Ma and spans the ca. 17 Ma to Holocene collapse of the Nevadaplano and contemporaneous migration of the Yellowstone hotspot. This synthesis incorporates the fundamental elements of our paleogeographic reconstruction of the Knoll Mountain region (Fig. 12) and our regional correlations of faults and basins (Fig. 13) and integrates it with (1) data from the Snake River Plain to the north, and (2) the Wallace et al. (2008) paleogeographic reconstruction of the Ruby basin (also known as the Elko basin) to the south.

Paleogeography from Eocene (38 Ma) to pre–17 Ma

Few details are known about the late Eocene (38 Ma) to pre–17 Ma paleogeography; however, some generalizations can be made. Henry (2008) inferred that a major north-trending drainage divide was present in northern Nevada during that time (Fig. 14A), and there is evidence for local surface-breaking normal faults and development of associated sedimentary basin fill (e.g., Mueller et al., 1999; Rahl et al., 2002; Haynes, 2003; McGrew and Foland, 2004; Lund Snee, 2013; McGrew and Snoke, 2015; Lund Snee et al., 2016). With respect to the area of our paleogeographic reconstruction, the most significant pre–17 Ma feature is the Holborn fault (Mueller et al., 1999) and its basin fill, the Clover Creek sequence (Fig. 2). Although only small remnants of this basin or basins(?) are preserved, and the regional extent of the Holborn fault or system of basin-bounding faults is not known, we have schematically shown the probable extent of the basin in Figure 14A. This fault or fault system was responsible for uplift and erosion of the northernmost Pequop Mountains and the parts of the metamorphic core complex (Mueller et al., 1999). In addition, McGrew and Snoke (2015) reported that clasts of Wood Hills–like metamorphic rock make their first appearance in the Clover Creek sequence (near location 5 in Fig. 2A) between 29 and 16 Ma, indicating that initial exhumation of part of the metamorphic core complex took place during this time. This phase of extension may be related to, or perhaps be the updip surface expression of, a 29–23 Ma top-to-the-west-northwest normal-sense mid-crustal mylonitic shear zone exposed in the Wood Hills and East Humboldt Range (e.g., Wright and Snoke, 1993; Mueller et al., 1999; Camilleri et al., 2013).

Paleogeography from ca. 17 to 15 Ma: Initial Collapse of the Nevadaplano and Establishment of the Knoll-Ruby and Thousand Springs Fault Systems

The period from ca. 17 to 15 Ma marks the initiation of widespread extensional collapse of the Nevadaplano and establishment of Basin and Range physiography accommodated by the development of normal faults and facilitated in part by the formation of the Yellowstone hotspot in northwestern Nevada (e.g., Camp et al., 2015). General extension directions established during this time included northeast-southwest extension in southeast Idaho and a broad zone of northwest-southeast extension in Nevada, the center of which contained a narrow zone of northeast-southwest extension along the northern Nevada rift (Fig. 14B; e.g., Rodgers et al., 2002; Colgan, 2013). These general extension directions would persist to the present (Figs. 14B–14I) with the exception of the northern Nevada rift, which was only active from 17 to 15 Ma and is thought to be related to the initiation of the hotspot (e.g., Rodgers et al., 2002; Payne et al., 2012; Colgan, 2013).

From ca. 16 to 15 Ma, the north-northeast–trending Knoll-Ruby and Thousand Springs fault systems were active, and the Knoll-Ruby and Thousand Springs basins subsided concomitant with deposition of sediment of the Humboldt Formation (Fig. 14B). This time was also characterized by pervasive rhyolitic volcanism (Brueseke et al., 2014) with the Jarbidge Rhyolite lava flows and domes covering much of the northwestern corner of the Knoll-Ruby basin (Fig. 14B). In addition, because bedrock mapping by Coats (1987) and Gardner (1968) suggests that the Jarbidge Rhyolite is offset and rotated toward the east in the footwall of the northern Snake Mountains fault, we have inferred that the northern Snake Mountains fault was either blind or had not yet formed at this time.

Uplift of the footwall of the Knoll-Ruby fault system established local north- to north-northeast–trending drainage divides, and we infer that the drainage divide on the Knoll Mountain segment of the uplifted footwall merged with the east-northeast–trending divide associated with the Yellowstone hotspot (Beranek et al., 2006; Fig. 14B). Axial drainage in the southern (Wallace et al., 2008) and northern (Deibert and Camilleri, 2006) parts of the Knoll-Ruby basin is inferred to be southward.

Paleogeography from ca. 15 to 10 Ma: Continued Slip on the Knoll-Ruby and Thousand Springs Fault Systems and Development of the Snake River Plain

The period between ca. 15 and 10 Ma is characterized by continued slip on the Knoll-Ruby fault system and the northeastward migration of the Yellowstone hotspot (Fig. 14C). However, toward the end of this time period, during a widespread flare-up of rhyolitic volcanism between 11.7 and 10.2 Ma (Bonnichsen et al., 2008), parts of the landscape along the track of the hotspot profoundly changed with the development of the structurally distinct western and eastern parts of the Snake River Plain (Fig. 14C). We first summarize the general paleogeography of the area south of the Contact pluton, and then focus on the events that transpired between 11 and 10 Ma along the track of the hotspot with special emphasis on the areas north and south of the confluence of the western and eastern parts of the Snake River Plain.

Area South of the Contact Pluton

By ca. 10 Ma, we infer that the northern Snake Mountains fault had formed and substantial basin fill had accumulated in the Knoll-Ruby basin, with much of it consisting of primary and reworked pyroclastic deposits derived from rhyolitic volcanism. This time period also involved further emergence of metamorphic rocks of the Ruby–East Humboldt–Wood Hills core complex, as manifest by the appearance of metamorphic clasts in the Ruby basin by ca. 15–14 Ma (location 10 in Fig. 2A; Sharp, 1939a, 1993b; Smith and Ketner, 1976; Wallace et al., 2008; Lund Snee, 2013; Lund Snee et al., 2016) and by 12 Ma in the Wood Hills (Camilleri, 2010a; location 4 in Fig. 2A). The depocenter of the Knoll-Ruby basin may have been localized adjacent to the Knoll-Ruby fault system just south of the northern Snake Mountains (Fig. 14C); this is suggested by the ∼6-km-thick lobe of basin fill present between the modern Snake Mountains and East Humboldt Range (Fig. 2B). Although this thickness includes Tertiary strata above and below the Humboldt Formation, Robison (1983) indicated that at least ∼3 km of the fill is Miocene in age. This inferred depocenter is near the center of the fault system, which is where maximum subsidence and fault slip would be expected (e.g., Morley, 1999), and accordingly, the footwall of the fault system in this area (northernmost East Humboldt Range and Wood Hills) contains some of the structurally deepest exhumed rocks in the Ruby–East Humboldt–Wood Hills metamorphic core complex (e.g., Hodges et al., 1992; Camilleri and Chamberlain, 1997; McGrew et al., 2000; Sicard et al., 2011; Wills et al., 2013; Hallett and Spear, 2014; McGrew and Snoke, 2015).

The period of time between 15 and 10 Ma also involved the beginning of regional changes in the patterns of major streams. Increased fault activity along the western margin of the Basin and Range Province resulted in lowering of regional base level and triggered the beginning of a progressive eastward integration of streams to form the incipient western reaches of the Humboldt River system (Fig. 14C; Wallace et al., 2008).

Profound Changes in Paleogeography along the Track of the Hotspot 11–10 Ma

From 11 to 10 Ma, during voluminous magmatism in the Bruneau-Jarbidge, Twin Falls, and Picabo eruptive centers, significant changes took place that resulted in development of the western and eastern parts of the Snake River Plain. The injection of large mafic intrusions beneath the eastern Snake River Plain are inferred to have produced a localized crustal load that triggered subsidence and formation of the eastern part of the Snake River Plain ca. 11–10 Ma along with downflexing of rocks along the margin of the plain (Fig. 14C; e.g., McQuarrie and Rodgers, 1998; Hough, 2001; Rodgers et al., 2002; Michalek, 2009; Vogl et al., 2014; Knott 2016a, 2016b). The Rogerson graben formed shortly following the initiation of subsidence at the southwestern end of the eastern Snake River Plain, ca. 10 Ma (Knott et al., 2016b). At about the same time that the eastern Snake River Plain formed, ca. 11 Ma, the western part of the Snake River Plain formed as a consequence of the development of the ∼70-km-wide western Snake River graben (Fig. 14C; Wood and Clemens, 2002). Wood and Clemens (2002) suggested that the majority of fault slip associated with development of the western Snake River Plain graben took place between 11 and 9.5 Ma and was followed by low slip rates continuing to the present.

In the area north of the Contact pluton, we infer that the initiation of subsidence in the eastern Snake River Plain caused the drainage divide, which was centered on top of the hotspot (Beranek et al., 2006), to begin migrating southward. However, during this time, the drainage divide must have been located north of the Contact pluton, because detrital zircons from this pluton do not appear in the Snake River Plain until ca. 3 Ma (Link et al., 2002, 2005; Beranek et al., 2006). It is possible that the divide could have been in proximity of the hinge of the flexural monocline recognized by Knott et al. (2016a, 2016b) (cf. Figs. 13A and 14C), which may represent a peripheral bulge, or area of relative vertical uplift, adjacent to the flexurally subsiding Snake River Plain. Nonetheless, and regardless of what subsidence or uplift processes were operating near the hotspot, the provenance of sediment north and south of the Contact pluton indicates that there was an east-trending drainage divide north of the Contact pluton (Fig. 14C).

Coeval Events due North and South of the Confluence of the Western and Eastern Parts of the Snake River Plain

Between 11 and 10 Ma, the confluence of the eastern and western parts of the Snake River Plain was characterized by downflexing of rocks into the plain. North of the confluence, downflexing gave way to regional uplift and erosion (Fig. 14C; Vogl et al., 2014). To the south, downflexing gave way to incipient development of the Rogerson graben (Andrews et al., 2008; Knott et al., 2016b) and significant activity along the Knoll-Ruby and Thousand Springs fault systems (Figs. 14C and 12B).

Paleogeography from ca. 10 to 8 Ma: Establishment of the Hice-Valder Fault and the Humboldt River System in the Knoll-Ruby Basin

Between ca. 10 and 8 Ma, rhyolitic volcanism in the Snake River Plain was accompanied by continued faulting along the Rogerson graben and subsidence and downflexing of rocks flanking the margin of the Snake River Plain (Andrews et al., 2008; Knott et al., 2016a, 2016b). The Thousand Springs and Knoll-Ruby fault systems were still active but were probably beginning to wane (Fig. 14D). At the northern end of the Knoll-Ruby fault system, the Hice-Valder fault and the associated Hice syncline developed in the hanging wall of the Knoll Mountain fault ca. 9.8 Ma (Fig. 12C). This was followed by stream incision, valley formation, and back-filling of the valley in the Hice syncline area between 9.8 and 9.6 Ma (Figs. 12D, 12E). It is uncertain what caused the relative fall and rise of base level that produced the valley and its infilling, but one possibility is that the base-level changes may be a consequence of vertical changes in elevation related to peripheral bulge activity associated with the flexurally subsiding Snake River Plain. By the end of this time period, the streams in the southern part of the Knoll-Ruby basin (Wallace et al., 2008) and, we infer, in the northern part, integrated into the Humboldt River system (Fig. 14D).

Paleogeography after 8 Ma to 3 Ma: Extinction and the Breakup of the Knoll-Ruby and Thousand Springs Fault Systems and Transition to the Modern Extensional Regime and Physiography

Between 8 and 3 Ma, rhyolitic volcanism associated with the Yellowstone hotspot migrated northeastward to the eastern margin of the eastern Snake River Plain (Fig. 14E) and was accompanied by subsidence and local basaltic volcanism (Bonnichsen and Godchaux, 2002). South of the eastern Snake River Plain, two successive and profound changes in tectonics and paleogeography took place during this time. The first change was a basinward shift in faulting in the Knoll-Ruby and Thousand Springs basins that produced the intrabasin faults followed by the extinction of the Knoll-Ruby and Thousand Springs fault systems. The second change was the development of a younger, modern set of normal faults resulting in the termination of the Knoll-Ruby fault system and the Knoll-Ruby basin. We elaborate on these changes by dividing this time of transition between ca. 8 and 3 Ma into early and late phases, represented in Figures 14E–14G.

Early Phase

The early phase began with waning of slip on the Knoll-Ruby fault system coupled with a basinward shift in faulting reflected by the development of intrabasin faults, including the Summer Camp and Henry faults in the northern part of the Knoll-Ruby basin and an unnamed fault, or faults, to the south (cf. Figs. 14D, 14E). Drainage in the northern part of the Knoll-Ruby basin was still to the south and was possibly connected to the Humboldt River system (Fig. 14E). Subsidence and accumulation of sediment occurred in the hanging wall of the Henry and Summer Camp intrabasin faults to the north. Similarly, it is possible that sediment accumulated in the hanging wall of the unnamed intrabasin fault, or faults, to the south (see Fig. 2A), but poor preservation and subsequent faulting of the Humboldt Formation in this area preclude this assessment. Toward the end of this phase, waning slip on faults and subsidence along the margins of the Knoll-Ruby basin resulted in a transition to erosion outpacing sedimentation with net transport of sediment out of the Knoll-Ruby basin by the Humboldt River system (e.g., Wallace et al., 2008). The end of the early phase is marked by the extinction of the intrabasin faults and widespread erosion of the Humboldt Formation.

Late Phase

Following the extinction of the Knoll-Ruby fault system and its intrabasin faults, and the migration of the hotspot farther eastward to the Heise eruptive center, there was a southward shift of the locus of significant tectonism away from the subsiding Snake River Plain to the East Humboldt Range and environs (cf. Figs. 14E, 14F). This phase of tectonism culminated in the development of the younger generation of faults that effectively (1) dismembered the eastern margin of the Knoll-Ruby basin and Knoll-Ruby fault system and (2) facilitated emergence of the modern Ruby Mountains, East Humboldt Range, southern Snake Mountains, and Pequop Mountains (Fig. 14F). The emergence of these new mountain ranges profoundly altered drainage systems, resulting in migration of drainage divides and establishment of a series of hydrographically closed basins east of the Ruby Mountains and East Humboldt Range (Fig. 14F). The rate of fault-induced uplift of the southern Snake Mountains appears to have kept pace with the rate of incision by the Humboldt River resulting in the production of antecedent segments of the Humboldt River system such as Bishop Creek and Town Creek (cf. Figs. 14E, 14F). It is possible that, during the initial phases of uplift of the southern Snake Mountains, the northern part of the Knoll-Ruby basin may have still drained southward (Fig. 14F). However, migration of the Snake River drainage divide into the northern part of the basin, which produced a reversal from a south- to a north-draining stream system by ca. 3 Ma (e.g., Beranek et al., 2006), would have effectively ended any connection with the Humboldt River system (Fig. 14G), and hence would establish the Knoll basin as an separate entity. In addition, although the older alluvium in the Knoll basin is not shown on our paleographic reconstruction, we note that it may have been deposited either prior to, during, or after drainage reversal (i.e., it may be either older or younger than ca. 3 Ma; Figs. 10 and 13B).

Paleogeography from ca. 3 Ma to Holocene

Since 3 Ma, the Yellowstone hotspot has migrated to its present position and waning subsidence in the eastern Snake River Plain has been accompanied by basaltic volcanism (e.g., Kuntz et al., 1992; Hughes et al., 1999; Figs. 14H, 14I). Today, global positioning system data from Payne et al. (2012) suggest that the eastern part of the Snake River Plain is behaving as a relatively coherent, unextending block with right-lateral shear along its margins that transitions to west-northwest– to east-northeast–directed extension away from the plain (Fig. 14I). Furthermore, modern extension directions mimic those spanning the development of the Basin and Range (Figs. 14B–14I), suggesting the persistence of extension directions from the onset of Basin and Range extension to the present (e.g., Colgan et al., 2004; Puskas et al., 2007; Payne et al., 2012).

South of the Snake River Plain, the Knoll Mountain region has been relatively tectonically quiescent, and pervasive fluvial erosion by the modern Snake River drainage system has further exposed the Humboldt Formation and older alluvium (Figs. 12J, 14H, and 14I). In addition, we infer that at some point during this time, part of the Snake River drainage system in the southern end of the Knoll basin was captured by the Salt Lake drainage system (for additional information see Supplemental Item 1 [footnote 1]).

In contrast to the Knoll Mountain region, the East Humboldt Range and environs remained significantly tectonically active, and the Ruby basin underwent a brief renewed period of fluvial-lacustrine deposition from ca. 3 to 2 Ma in Lake Elko (Fig. 14H). Development of Lake Elko is inferred to be related to downstream incision into a bedrock sill, which slowed the rate of erosion upstream (Wallace et al., 2008), or to downstream damming of the river by a landslide (Reheis et al., 2014). Alternatively, it is possible it could be related to subsidence exceeding the rate of erosion of the basin due to slip along faults bounding the Ruby–East Humboldt Range. Nonetheless, after ca. 2 Ma, a throughgoing Humboldt River was reestablished and the river began incising into the lacustrine units and the underlying Humboldt Formation (Wallace et al., 2008; Fig. 14I). Although the precise timing of slip along the modern normal faults in the East Humboldt Range region is poorly constrained, evidence for Quaternary fault slip is plentiful. For example, fault scarps preserved in the Ruby Mountains and East Humboldt Range indicate latest slip at 7.6–4.8 ka (Wesnousky and Willoughby, 2003), and a fault scarp in the Pequop Mountains indicates the latest slip ca. 42 ka (Wesnousky et al., 2005). The most recent slip event is related to a 2008 M 6.0 earthquake along an east-dipping fault bordering the east side of the northernmost part of the East Humboldt Range and southern Snake Mountains (Henry and Colgan, 2011; Ponce et al., 2011).

DISCUSSION

Our data from the Knoll Mountain region provide a new perspective on (1) the paleogeography of northeastern Nevada–south-central Idaho and the relationship between tectonism and the passage of the Yellowstone hotspot, and (2) the tectonic history of the Ruby–East Humboldt–Wood Hills metamorphic core complex that we highlight in the following.

Comparison of the Extensional History of the Knoll Mountain Region with the Regional Space-Time Pattern of Faulting along the Southern Margin of the Snake River Plain and Yellowstone Hotspot

The Knoll-Ruby and Thousand Springs fault systems formed at ca. 16 Ma at the onset of Basin and Range extension and approximately the same time as initiation of volcanism associated with the Yellowstone hotspot, and both were active at the same time for a significant period of time. We interpret the evolution of these fault systems to be strongly influenced by regional Basin and Range extension because of three factors. (1) The north to north-northeast strike of the fault systems is consistent with west to west-northwest regional extension directions that have persisted in northeast Nevada since the initiation of Basin and Range extension (e.g., Colgan, 2013). (2) The Knoll-Ruby fault system overall appears to have a relatively consistent style and history along its >190 km length extending from very distal to proximal to the Bruneau-Jarbidge and Twin Falls eruptive centers along the track of the hotspot (Fig. 13B). (3) These fault systems were active before, during, and after the migration of the front of hotspot volcanism past the longitude of Knoll Mountain, and as such, the influence of passage of the hotspot on faulting is not readily apparent. However, to explore this issue, we compare the history of the Knoll Mountain segment of the Knoll-Ruby fault system to the space-time pattern of extension recognized by others along the southern margin of the Snake River Plain east of Knoll Mountain.

Elements of the history of faulting in the Knoll Mountain area fit the space-time pattern of extension along the southern margin of the Snake River Plain in some regards and not in others. East of Knoll Mountain, a minor, but regional, phase of extensional faulting (16–11 Ma) preceded the passage of the hotspot and development of the Snake River Plain (Fig. 15; Rodgers et al., 2002). This was followed by a generally time-transgressive major (rapid slip) phase of extension from ca. 13.5 to 4 Ma that broadly coincided and ended with northeastward migration of hotspot volcanism (Fig. 15; Rodgers et al., 1990, 2002; Pierce and Morgan, 1992, 2009; Anders, 1994; Wells et al., 2000; Egger et al., 2003; Konstantinou et al., 2012).

Our new data indicate that the chronology of faulting along Knoll Mountain is consistent in a very general way with the space-time pattern of extension along the southern margin of the Snake River Plain east of Knoll Mountain. Similar to other areas to the east, extension in the Knoll Mountain region began long before the ca. 12.7 and 10.5 Ma arrival of proximal hotspot volcanism of the Bruneau-Jarbidge and Twin Falls eruptive centers (Fig. 15), and there clearly was faulting in the Knoll Mountain area, as well as in the Rogerson graben, during proximal volcanism (Figs. 10 and 15), but our data do not allow the rate of fault slip to be determined. Therefore, it is unknown if extension during passage of the hotspot can be characterized as a period of rapid slip as it has in areas to the east (e.g., Rodgers et al., 2002). In addition, unlike other areas to the east where significant extension typically did not persist following passage of hotspot volcanism (Fig. 15), it appears to have persisted in the Knoll Mountain and Rogerson graben area. In these areas, extension persisted following proximal hotspot volcanism between 8 and 5 Ma in the Knoll basin, and as late as Pliocene in the Rogerson graben (Fig. 15; Andrews et al., 2008; Knott et al., 2016b). Whether the timing and duration of Basin and Range faulting recognized in the Knoll Mountain area extend along the southern margin of the track of the hotspot west of Knoll Mountain is uncertain, but at some point it must not persist because faulting near the beginning of the track of the hotspot, in the vicinity of the McDermitt volcanic center, significantly postdates, and is therefore unrelated to, passage of the hotspot (SR in Fig. 15; Colgan et al., 2004).

In summary, while it is apparent that early extensional faulting significantly predated passage of the hotspot along the southern margin of the Snake River Plain, the overall character of the later faulting history appears to be variable. Most notable is that at least to the east of Albion–Raft River metamorphic core complex, the cessation of significant faulting appears to have a strong link in space and time with the cessation of hotspot volcanism (Fig. 15; Rodgers et al., 2002), whereas to the west, extension appears to have persisted and initiated after the passage of the hotspot (Fig. 15).

New Perspective on the Extensional History of the Ruby–East Humboldt–Wood Hills Metamorphic Core Complex Area

The Ruby Mountains–East Humboldt–Wood Hills metamorphic core complex has a multiphase extensional exhumation history with episodes in the Late Cretaceous, Oligocene to early Miocene, middle to late Miocene (via the Ruby–East Humboldt detachment), and late Miocene to Holocene (e.g., Snoke and Miller, 1988; Hodges et al., 1992; Mueller and Snoke, 1993a, 1993b; McGrew and Snee, 1994; Camilleri and Chamberlain, 1997; McGrew et al., 2000; Howard, 2003; Sullivan and Snoke, 2007; Colgan et al., 2010; Henry et al., 2011, and references therein). This region is inferred to have undergone localized, large-magnitude crustal thickening, with pre-Miocene phases of extension probably localized and representing collapse of the locally thickened crust (e.g., Miller et al., 1991; Miller and Hoisch, 1992; Camilleri et al., 1992; Mueller and Snoke, 1993a, 1993b; Camilleri et al., 1997). The middle to late Miocene phase of extension involving the Ruby–East Humboldt detachment was thought to have largely taken place between ca. 16 and 12–10 Ma (Colgan et al., 2010) or ca. 16 to 8 Ma (Mueller, 1992; Mueller and Snoke, 1993a; Lund Snee, 2013; McGrew and Snoke, 2015; Lund Snee et al., 2016). Our study indicates that the Ruby–East Humboldt detachment was not localized and was part of the >190-km-long Knoll-Ruby fault system that extended well beyond the core complex area. Furthermore, overall, this fault system was active for a longer time than previously recognized. Specifically, our data and synthesis indicates that the Knoll-Ruby fault system and intrabasin faults, at least north of the Ruby Mountains, ceased to be active sometime after 8 Ma but before ca. 3 Ma (Fig. 13).

One interesting aspect of our study is the recognition that the style and history of Basin and Range faulting in both the Ruby–East Humboldt and Albion–Raft River metamorphic core complexes are nearly identical. Like the Knoll-Ruby fault system, significant core complex extension along the Albion–Raft River fault system predates passage of the hotspot and was followed by a phase of minor extension involving development of intrabasin faults that cut and rotated basin fill beginning after ca. 8 Ma (Fig. 15; e.g., Wells et al., 2000; Egger et al., 2003; Konstantinou et al., 2012) and inferred to end by ca. 5 Ma (Konstantinou, 2015). We also note that the intervening Thousand Springs fault system and basin are also similar in that extension began ca. 16 Ma, and had associated intrabasin faults that cut and rotated basin fill (Figs. 2 and 3). Although no age constraints are available for the intrabasin faults in the Thousand Springs basin, in Figures 2 and 15 we have inferred that, like the Knoll-Ruby and Albion–Raft River fault systems, they postdate ca. 8 Ma.

CONCLUSIONS

Knoll Mountain is a horst bounded by the Knoll Mountain and Thousand Springs fault systems and associated basins, which form half-grabens filled with the synextensional ca. 16 to ca. 8–5 Ma Humboldt Formation. The Humboldt Formation was deposited in alluvial, eolian, and lacustrine environments and, in the Knoll basin, it records overall southward fluvial drainage. Slip along the Knoll Mountain fault began ca. 16 Ma and continued to at least 8 Ma, with coeval deposition of sediment of the Blanchard, Knoll, Cave, and Bloody Gulch members of the Humboldt Formation. After or ca. 8 Ma, but prior to ca. 5 Ma, the locus of faulting stepped basinward with the development of the Henry and Summer Camp intrabasin faults. This was accompanied by a shift of the basin depocenters to the hanging wall of these faults with coeval deposition of the Eagle Flat member of the Humboldt Formation. Slip along the intrabasin faults likely temporally overlapped with the inception of subsidence of the Snake River Plain to the north, although drainage in the Knoll basin was still to the south during this time. Following development of the intrabasin fault system between 8 and 5 Ma, but prior to ca. 3 Ma, several fundamental changes took place in the Knoll basin, beginning with the cessation of faulting followed by widespread erosion, deposition of the older alluvium, and ultimately reversal of drainage direction from south flowing to north flowing. The age of the older alluvium relative to the reversal of drainage direction is uncertain and may either predate, postdate, or be synchronous with the reversal.

The Knoll Mountain fault represents the northern end of a >190-km-long, west-dipping fault system, the Knoll-Ruby fault system. This fault system, and its hanging-wall basin, the Knoll-Ruby basin, was a prominent structure during initial collapse of the Nevadaplano in northeastern Nevada from ca. 16 to 8 Ma or later, and its central part facilitated partial exhumation of metamorphic rocks in the Ruby–East Humboldt–Wood Hills metamorphic core complex. Fault-dip angles along the length of the Knoll-Ruby fault system ranged from low to high angle, and on the basis of modern analogues, we infer that this fault system was probably segmented.

After 8 Ma, but prior to 5 Ma, during the waning stages of extension along the Knoll-Ruby fault system, a series of intrabasin faults developed, resulting in shifting of the depocenter of the Knoll-Ruby basin to the hanging walls of these faults. The development of the intrabasin faults preceded or overlapped with the integration of streams to form the easternmost reaches of the Humboldt River system, which thereafter appears to have caused widespread fluvial erosion.

Extinction of the intrabasin faults and the Knoll-Ruby fault system by ca. 3 Ma was followed by a southward shift in tectonism away from the subsiding Snake River Plain, and new faults were produced that cut and dismembered the central and southern parts of the Knoll-Ruby fault system and the Knoll-Ruby basin. Slip along these faults resulted in development of the modern Ruby Mountains, East Humboldt Range, southern Snake Mountains, and Pequop Mountains. Furthermore, footwall uplift associated with these faults (1) effectively ended any structural or hydrographic connection between the Knoll and Ruby basins, (2) facilitated migration of drainage divides, (3) locally produced antecedent stream segments of the Humboldt River across footwall uplifts, and (4) created a series of hydrographically closed basins east of the Ruby Mountains–East Humboldt Range.

Slip along faults in the Knoll Mountain region occurred before, during, and after passage of the hotspot at the longitude of Knoll Mountain. With the exception of significant faulting postdating passage of the hotspot, this faulting chronology is consistent in a general way with the space-time pattern of extension along the southern margin of the Snake River Plain east of Knoll Mountain. However, it is unknown if the rate of fault slip increased during passage of the hotspot.

ACKNOWLEDGMENTS

This paper is in honor of Arthur W. Snoke’s contributions to research on the diverse and complicated tectonic history of northeastern Nevada. Our research in northeastern Nevada has benefitted from discussions with Allen McGrew, Arthur Snoke, Karl Mueller, and Bill Bonnichsen. Preparation of tephra samples was done in a laboratory at the University of Utah (overseen by Frances H. Brown). Microprobe analyses were done at the University of Utah Electron Microprobe Laboratory under the direction of Barbara Nash. Austin Peay State University (APSU) research grants partially funded this research and Camilleri acknowledges a Spring 2015 APSU Faculty Development Assignment that facilitated the production of this paper and related research. We sincerely thank Associate Editor Graham Andrews, Luke Beranek, and an anonymous reviewer for insightful comments and suggestions that helped to improve the paper.

1Supplemental Item. Additional information about granitic clasts in the Humboldt Formation. Please visit http://doi.org/10.1130/GES01318.S1 or the full-text article on www.gsapubs.org to view the Supplemental Item.
16 figures; 3 tables; 1 supplemental file.

APPENDIX 1. DESCRIPTION OF THE EAGLE FLAT MEMBER OF THE HUMBOLDT FORMATION

Stratigraphic Position of the Eagle Flat Member and Estimates of its Thickness

The Eagle Flat member is restricted to the hanging wall of the Henry fault and the nature of its stratigraphic position and its depositional contact with the other members of the Humboldt Formation is not immediately apparent because no exposures of this contact were found. However, structural relationships and composition of the Eagle Flat member indicate that the Eagle Flat member overlies the Bloody Gulch member. The Eagle Flat member contains clasts of the welded units of the Cougar Point Tuff (Fig. 5), indicating that it is at least younger than the ca. 12.7–10.5 Ma Cougar Point tuffs (Perkins et al., 1995, 1998; Perkins and Nash, 2002; Bonnichsen et al., 2008). Furthermore, structural relationships across the Henry fault indicate that the Eagle Flat member should be stratigraphically above the Bloody Gulch member. Geometrically, a normal fault that cuts basin fill should place younger strata in the hanging wall against older strata in the footwall. Consequently, it is reasonable to assume that, because the Henry fault places the Eagle Flat member in the hanging wall against the Bloody Gulch member in the footwall (see cross section in Fig. 3B), the Eagle Flat member is younger than the Bloody Gulch member.

Sedimentary Characteristics of the Eagle Flat Member

The Eagle Flat member mostly consists of white to light tan, medium- to thick-bedded sandstone with minor amounts of conglomerate and siltstone (Fig. 5). Sandstone units are fine- to very coarse-grained and are commonly massive, with rare horizontal laminations and trough cross-stratification. Other sedimentary features include diagenetic silicification, horizontal zones of root casts, and large burrows to 8 cm in diameter and 1 m in length. Sandstone framework grains are angular to subangular, poorly sorted, and are dominantly composed of volcanic glass shards with minor amounts of quartz, feldspar, chert, biotite, and magnetite. Sandstone units contain minor amounts of subangular, granule- to cobble-sized clasts of chert, granite, siltstone, and welded Cougar Point Tuff. In addition, the Eagle Flat member contains vertebrate fossils that include mice, rabbits, camel, and rhinoceros (Macdonald, 1949; Fig. 11).

Conglomerate units occur as lenses and medium to thick beds, and dominantly have a matrix-supported texture. A few conglomerate units display clast-supported texture and crude horizontal stratification. Framework clasts are 2–20 cm in diameter, subangular, and are composed of chert, granite, siltstone, and welded Cougar Point Tuff (Fig. 5B). Matrix material within conglomerate units is typically angular to subangular, poorly sorted, and is mostly composed of volcanic glass shards with minor amounts of quartz, feldspar, chert, biotite, and magnetite. Siltstone units are thinly bedded and commonly display internal laminations. Minor soft-sediment deformation is present in some beds.

Depositional Environments of the Eagle Flat Member

Sedimentary structures observed in the Eagle Flat member indicate slow to fast, channelized to nonchannelized water flows with occasional debris flows. The immature nature of clasts suggests that they were transported a short distance from their source area. Horizontal zones of root casts and diagenetic silicification, along with the occurrence of fossils of large grazing animals, indicate the formation of soil and the presence of substantial vegetation. Overall, the Eagle Flat Member is interpreted to have been deposited in fluvial channels and flood plains adjacent to steep highlands.

Paleocurrents in the Eagle Flat Member

Sedimentary structures that permit assessment of paleocurrents are sparse in the Eagle Flat member. However, 11 paleocurrent readings were obtained from trough cross-stratified sets in sandstone beds, with sets ranging from 5 to 30 cm in height (see Fig. A1). The mean value of the 11 paleoflow directions is 164° ± 23°. The mean flow direction, along with the variations between the individual flow readings, indicates a general south to southeast flow of streams during the deposition of the Eagle Flat Member.

APPENDIX 2. DESCRIPTION OF THE OLDER ALLUVIUM

The older alluvium is composed predominantly of unconsolidated to poorly cemented, subangular to subrounded, poorly sorted gravel and sand. Gravel clasts include granite, chert, limestone, siltstone, and welded tuff. Sand grains are composed of quartz, feldspar, chert, biotite, and volcanic glass shards. Internally, the older alluvium is generally massive, and the top of the alluvium, in places, contains desert-pavement hardgrounds that form planar sloping surfaces that are dissected by modern streams. Some of the hardground surfaces are remnants of an older regional, nearly continuous surface that gently sloped from the margins of the mountain ranges down to the central area of the basin, and others appear to be younger strath terraces cut into this older surface by incision of the modern north-draining stream system in the Knoll basin.

Supplementary data