The location and geometry of the boundary between the Sierra Nevada microplate and the transtensional Walker Lane belt of the Basin and Range Province in the Lake Tahoe area have been debated. Two options are that the active structural boundary is (1) a few km west of Lake Tahoe, along the northwest-trending Tahoe-Sierra frontal fault zone (TSFFZ) or (2) within Lake Tahoe, along the largely submerged, north-trending West Tahoe–Dollar Point fault zone (WTDPFZ).

Emerald Bay, a famous scenic locality at the southwest end of Lake Tahoe, is at the juncture between the TSFFZ and the WTDPFZ. There, utilizing high-resolution, multibeam-echosounder maps and derived bathymetric profiles, detailed field studies on land are integrated with bathymetric data and remotely operated vehicle observations to clarify the existence and activity of faults and sedimentology of the bay. Results include the most detailed structural maps of glacial moraines and the bottom of Lake Tahoe ever produced. Glacial moraines on both sides of Emerald Bay clearly have been deformed by normal displacements on faults within the TSFFZ and the WTDPFZ. Tectonic geomorphic features include scarps along moraine crests, locally back-tilted crests, and tectonic reversal of moraine crests, where older, higher moraines locally lie at lower elevations than younger, lower moraines. The alignment of crests of lateral moraines shows that dextral slip has not occurred during or since late Pleistocene glaciations.

On the floor of Emerald Bay, submerged youthful faults that correspond to onshore faults that displace glacial moraines have numerous distinct, well-preserved, postglacial fault scarps, for which the vertical component of slip (vertical separation) is estimated.

This study clearly demonstrates that the TSFFZ is the active structural boundary of the Sierra Nevada microplate and that the TSFFZ has a higher rate of slip than the WTDPFZ. It also provides evidence for complex range-front evolution, with both zones of normal faults active concurrently at various times.


New studies at Emerald Bay, a world-famous, scenic locale in the Lake Tahoe basin, California-Nevada, clarify the relationship between two important fault zones along the eastern edge of the Sierra Nevada microplate. The Lake Tahoe basin (Figs. 1A and 1B), a complex half-graben, is part of the Walker Lane belt, possibly an incipient plate boundary, and a region of dextral transtensional deformation between the internally unfaulted Sierra Nevada and the extensional Basin and Range Province to the east. Numerous studies in the Walker Lane belt have established the following (see, for example, Bormann et al., 2016; Busby, 2013, 2016; Carlson et al., 2013; Dixon et al., 2000; Faulds and Henry, 2008; Hammond et al., 2011; Hammond and Thatcher, 2007; Lifton et al., 2013; Rood et al., 2011a; Schweickert et al., 2004; Surpless et al., 2002; Taylor and Dewey, 2009; Unruh et al., 2003; Wesnousky et al., 2012): The Walker Lane currently takes up about one fifth of the dextral displacement between the Pacific and North American plates. Kinematic complexity characterizes the Walker Lane, with various parts dominated by normal faults, dextral or sinistral strike-slip faults, and/or vertical-axis rotations of crustal blocks. Additionally, in some parts of the Walker Lane, discrepancies have been noted between long-term, geologically determined slip rates and those calculated from instantaneous GPS geodetic studies.

The Lake Tahoe basin, a normal-fault domain, is bounded to the north and south by domains that have conjugate dextral and sinistral faults and earthquake focal mechanisms (Fig. 1B; Schweickert et al., 2004). This study focuses primarily on the Tahoe-Sierra frontal fault zone (TSFFZ), a complex, northwest-trending zone of faults one to five km west of Lake Tahoe (Fig. 1B). Observations are also provided on the southern part of the north-trending West Tahoe–Dollar Point fault zone (WTDPFZ), a largely submerged and relatively simple zone of normal faults within and adjacent to Lake Tahoe.

Statement of the Problem

Normal faults along the western edge of the Lake Tahoe basin provide insights into the evolution of range-front faults along an incipient, transtensional plate boundary. The nature and continuity of faults in the area within and south of Emerald Bay, however, have been debated for more than a decade (Schweickert et al., 2000a, 2000b, 2004; Kent et al., 2005; Schweickert and Lahren, 2006; Dingler et al., 2009; Brothers et al., 2009; Howle et al., 2012; Maloney et al., 2013; Kent et al., 2016; Pierce et al., 2017). Due in part to this ongoing debate, along with the structural importance of the faults and public interest in the Emerald Bay area, the locale deserves a full, accurate characterization.

In the Lake Tahoe basin, which fault zone—the northwest-trending TSFFZ or the north-trending WTDPFZ—represents the active structural boundary of the Sierra Nevada microplate (Fig. 1A). The location and orientation of the active structural boundary are important because kinematic models for transtension (Dewey, 2002; Dewey et al., 1998; Taylor and Dewey, 2009) indicate that they set a kinematic boundary condition (e.g., they determine the direction of maximum instantaneous extension, Xi) for transtensional deformation in the Walker Lane belt to the east. The active boundary may also pose a significant seismic hazard. Criteria used here to identify the active structural boundary are that it should be marked by faults that have large displacements and are currently active.

Additionally, what is the kinematic history of these normal fault systems along the eastern edge of the microplate? Have they experienced exclusively dip-slip displacement, or have they experienced components of dextral slip? Has activity along these faults migrated progressively from the range front out into the basin, as observed along many range fronts in the Basin and Range Province (e.g., Koehler and Wesnousky, 2011; McCalpin, 2009; Personius et al., 2017; Wesnousky et al., 2005), or has fault activity alternated back and forth between the range front and the basin (e.g., Wallace, 1987)? When both the TSFFZ and the WTDPFZ are considered, are long-term, geological slip rates for the Lake Tahoe basin consistent with instantaneous slip (or strain) rates, or does a discrepancy exist?

Regional Setting

The Tahoe-Sierra frontal fault zone (TSFFZ), whose trend varies from N26° to 45°W (334–315° AZ), extends over 100 km from near Echo Lakes to the Mohawk Valley fault (Figs. 1B and 2; Howle et al., 2012; Schweickert et al., 2000a, 2000b, 2004; Schweickert, 2009; Sylvester et al., 2012). The footwall of the fault zone consists of granitic rocks of the largely unfaulted Sierra Nevada microplate (Howle et al., 2012; Saucedo, 2005; Schweickert et al., 2000b, 2004; Schweickert, 2009; Surpless et al., 2002; Unruh et al., 2004). In places, the granitic rocks are nonconformably overlain by cover sequences of Paleogene and Neogene volcanic rocks, Pleistocene glacial deposits, and, locally, lacustrine deposits (e.g., Saucedo, 2005; Sylvester et al., 2012).

Between ca. 3.5 and 2 Ma, east-side-down displacement on numerous subparallel branch faults of the TSFFZ beheaded various west-flowing Sierran drainages leaving wind gaps along the present crest (Fig. 2B; Schweickert, 2009). Normal displacement along the TSFFZ established the steep, east-facing Sierran escarpment that controlled the development of younger, east-flowing drainages and Pleistocene glacial valleys (Fig. 2A).

The TSFFZ includes several overlapping, northwest-trending, en echelon segments (Howle et al., 2012). From south to north, these include Twin Peaks, Ellis Peak, Rubicon Peak, Mount Tallac, and Echo Peak segments (Figs. 2A and 2B). Each segment, through long-term normal displacements, has developed high-standing bedrock facets (Fig. 2A) that are underlain mainly by granitic bedrock. Two million-year-old and older lacustrine sediments, which were deposited in a lake that predated modern Lake Tahoe are found in the hanging wall of the TSFFZ over much of its length (Fig. 2B; Kortemeier et al., 2018; Lopez et al., 2004; Moore et al., 2006; Schweickert et al., 2005).

The north-trending West Tahoe–Dollar Point fault zone (WTDPFZ), as used in this report, is submerged along a 25-km-long stretch northward from near Emerald Bay to Dollar Point (Fig. 2A), north of which it continues for about another 10 km on land (Howle et al., 2012; Schweickert et al., 2004; Schweickert et al., 2000a, 2000b, 2004; Sylvester et al., 2012).

The combined 1380 m height of footwall facets on both fault zones two to five kilometers north of Emerald Bay gives that area the greatest structural relief of any area within the Lake Tahoe basin.

Geology of the Emerald Bay Area

Emerald Bay, a submerged glacial valley, is well known for its glacial geology and scenery. During parts of its history, it has been a moraine-dammed lake, although at the current level of Lake Tahoe, the two lakes have merged. Prominent lateral moraine complexes that trend N35–45°E (035–045° AZ) form the steep slopes along the sides of Emerald Bay and Cascade Lake (Figs. 35) and are ∼2.2 km in length. Glacial till within the moraines consists principally of boulders of granodiorite with a sandy matrix. Scenic, State Route 89 (CA 89) traverses high-standing glacial moraines that enclose the bay.

Emerald Bay and Cascade Lake, together with their glacial moraines, span most of the faults of the TSFFZ (Figs. 4 and 5). Emerald Bay lies along a prominent right step in the TSFFZ, where the steep range front steps 1.8 km eastward from Eagle Falls toward Emerald Point (Howle et al., 2012; Figs. 25).

The Mount Tallac segment (as used here) of the TSFFZ, which includes at least five subparallel branches or splays within granitic bedrock, extends ∼20 km N30°W (330° AZ) along the prominent, 1-km-high range front from near the south end of Fallen Leaf Lake northwestward past the southwest end of Emerald Bay (Howle et al., 2012; Figs. 2, 4, and 5; Fig. S1 in the Supplemental Material1). Eagle Falls and Cascade Falls (Figs. 3 and 4) both lie along this prominent escarpment. It is noteworthy that both deep, U-shaped, glacial valleys of Eagle and Cascade creeks upstream from the falls do not align with the downstream parts of the valleys. The upstream valleys (Figs. 3 and S1 [text footnote 1]), which have been carved into granitic bedrock, appear to have been displaced in a dextral sense relative to the valleys below the falls. This geomorphic relationship suggests that faults within the Mount Tallac segment experienced some dextral displacement during Pleistocene time. Work in progress suggests that a similar relationship exists along many of the glacial valleys north of Emerald Bay (Fig. 2B).

The Stony Ridge fault (SRF) lies between the Mount Tallac fault and the Rubicon Peak fault (Howle et al., 2012; Figs. 4 and 5). Northwest of Emerald Bay, the SRF lies entirely within granitic bedrock, although it displaces glacial moraines near small cirques high on the east flank of Stony Ridge (Figs. 2A and 4). Howle et al. (2012) reported that branches of the fault southeast of Stony Ridge displace lateral moraines on the north side of Emerald Bay and then enter Emerald Bay. Beneath the bay and described in detail here are what appear to be fault scarps that cut through and around Fannette Island and that also form a steep, submerged bedrock escarpment on the northeast side of the island. About 200 m southeast of Fannette Island, these faults are buried beneath modern sediment and apparently form a broad flexure. Southeastward from there, several scarps representing branches of the SRF cross and displace the medial moraine complex south of Emerald Bay.

In the Rubicon Peak segment of the TSFFZ, Howle et al. (2012) reported that two main branches of the Rubicon Peak fault (RPF) cross and displace moraines on the north side of Emerald Bay, and they interpreted a large landslide on the bottom of the bay to have been cut by youthful scarps (Figs. 4 and 5). Where the RPF continues southeastward across the medial moraine complex on the southeast side of Emerald Bay, moraine crests define a monoclinal flexure (Fig. 4). About 200 m southeast from the Emerald Bay moraine complex, the probable RPF passes beneath postglacial fluvial deposits northeast of Cascade Lake. One kilometer farther southeast, prominent scarps along the RPF cut moraines on the south side of Cascade Lake (Figs. 4 and S3 [footnote 1]).

Landslides are present along much of the range front defined by the TSFFZ, particularly along the Mount Tallac and Rubicon Peak segments (Howle et al., 2012; their figures DRF2, DRF5, and DRF16).


The following sections present new maps, profiles, and descriptions of: (1) glacial moraines, (2) modern sedimentology, and (3) submerged scarps on the floor of Emerald Bay. Significantly, the moraine crests at Emerald Bay provide plentiful evidence for dip-slip normal displacements, while the submerged scarps allow vertical separation and extension rates to be estimated for many of the normal faults.

Reliable age control on glacial moraines in the region that encompasses Emerald Bay includes results of Howle et al. (2012) at Meeks Creek, 8 km northwest, and Pierce et al. (2017) on right-lateral moraines at Cascade Lake, 2 km south (Fig. 2A). Additionally, Rood et al. (2011b) determined ages on postglacial deposits on the eastern flank of the Carson Range, 25 km east of Emerald Bay. As discussed in the document in the Supplemental Material [footnote 1], we adopt ages of 23.5 ± 3 ka to 20.5 ± 0.6 ka for Tioga-age moraines (Howle et al., 2012).

Structure of the Glacial Moraines

Lateral moraines are referred to as left and right lateral, respectively, when viewed downstream or down canyon (Howle et al., 2012). The left-lateral moraines of Emerald Bay (north side) are draped upon granitic bedrock, whereas the right-lateral moraines (south side) are free standing and separate Emerald Bay and Cascade Lake. Moraines of different ages have been differentiated during field mapping using criteria described by McCaughey (2003) and Howle et al. (2012) (see document in Supplemental Material [footnote 1]). In this study, Tahoe-age moraines (Qta) and two sets of Tioga-age (Qti) moraine crests have been distinguished, an older Qti-1 crest and a younger, inset Qti-2 moraine crest. The Tioga-age crests have similar weathering characteristics and subequal heights (Howle et al., 2012; see document in Supplemental Material [footnote 1]).

Structural maps and topographic profiles of the moraines (Figs. 69) have been prepared using established criteria for tectonic geomorphology of faulted moraines (Howle et al., 2012; Schweickert et al., 2004; see document in Supplemental Material [footnote 1]). Careful field examination has allowed recognition of very small scarplets (1–2 m in height) that are not evident in the light detection and ranging (LiDAR) imagery. The terminology used for scarps, scarp height (SH), and vertical separation (VS) is explained in Figure S4 (footnote 1).

Scarp heights for subaerial scarps cutting moraines have been estimated in the field using a tape and compass and using topographic profiles, as discussed in later sections. All estimates are reported to the nearest half meter and to the nearest foot.

Faults and Scarps Cutting the Left-Lateral Moraines (Figs. 6 and 7)

East-facing fault scarps and backtilted moraine crests are prominent where the left-lateral moraines have been cut and displaced by the Mount Tallac, Stony Ridge, and Rubicon Peak faults (Howle et al., 2012; Fig. 6A).

Near the southwest end of the Qta moraine, large gullies along the eastern branch of the Mount Tallac fault have removed the crest; the remnant of the crest east of the large gullies is ∼27 m (89 ft) lower than the part southeast of the gullies, reflecting post-Tahoe normal displacement along an eastern branch of the Mount Tallac fault.

The Stony Ridge fault (SRF) has two distinct branches that have produced scarps of similar height in the Qti-1 moraine (Figs. 6A and 6B). A few tens of meters northeast of the Stony Ridge fault, the Qta moraine has been buried beneath the Qti-1 moraine (this is called tectonic reversal, which indicates significant fault displacement occurred between the deposition of Qta and Qti-1 moraines; see document in Supplemental Material [footnote 1]).

Along both branches of the Rubicon Peak fault (RPF), prominent, northeast-side-down scarps are within the Qta and Qti-1 moraines (the Qti-2 moraine crest where crossed by the RPF has been removed by landsliding [Figs. 6A and 6C]). Field estimates for heights of two scarps in the Qti-1 moraine are 8 and 6 m (∼26 and 20 ft), respectively (the topographic profile in Fig. 6C gives estimates of 9 and 6 m [30 and 20 ft] for heights of the two scarps), in general agreement with the Howle et al. (2012) estimate of VS for the scarps cutting Qti-1.

The submerged West Tahoe fault (WTF; the southern part of the WTDPFZ) continues southward onshore near the northeast end of the left-lateral moraine complex (Fig. 7). There, a bedrock escarpment ∼90 m (297 ft) high separates Qta, Qti-1, and Qti-2 crests to the southwest from much lower Qti-1 and Qti-2 moraine crests near Emerald Point; Qta is not exposed in the hanging wall east of the escarpment, due to tectonic reversal. In contrast, the Qti-1 moraine crest projects across this escarpment with little apparent displacement. Two possible explanations are: (1) that at least 90 m of post-Qta and pre-Qti-1 normal displacement occurred along the fault at the base of the escarpment, or (2) the Qta glacier flowed down a preexisting bedrock escarpment, and further fault displacement occurred prior to deposition of the Qti-1 moraine. We tentatively favor the second hypothesis. A few tens of meters northeast of the bedrock escarpment (Fig. 7), an eastern branch of the WTF has noteworthy, fresh scarps, as mapped by Howle et al. (2012).

Faults and Scarps Cutting the Right-Lateral Moraines (Figs. 8 and 9)

The high-standing, 1-km-long, remnant of the Qta moraine near the range front has a strongly backtilted segment in the hanging wall of the eastern branch of the Mount Tallac fault (Fig. 8A). The Qta remnant has several small, 2–3-m-high scarps along its crest, and at its northeast end, the moraine terminates in a 15-m-(50-ft)-high triangular facet along the eastern branch of the Stony Ridge fault (SRF).

Near CA 89 and Inspiration Point (Figs. 8A and 8B), a 10-m-(33-ft)-high scarp along an unnamed fault separates granodiorite capped by Qti-2 in the footwall from backtilted Qti-2 moraine in the hanging wall. This scarp may mark a separate branch of the MTF.

Along the western branch of the SRF, the Qti-2 crest has been removed where a deep gully cuts into the moraine (Figs. 8A and 8B). The Qti-2 crests on opposite sides of the gully have a difference in elevation of ∼10.5 m (35 ft).

The Qti-2 crest has not been displaced or disrupted adjacent to the large triangular facet at the end of the Qta moraine (Figs. 8A and 8B). This observation strongly suggests that major displacement on the eastern branch of the SRF, which resulted in tectonic reversal of Qta, was post-Qta and pre-Qti-2. The Qti-2 moraine is the only moraine exposed for a distance of ∼240 m northeast of the SRF, where it is surmounted by a very narrow stretch of CA 89.

Near the switchbacks along CA 89 (Figs. 8A and 8C), the slope of the Qti-2 crest steepens markedly down canyon. This steeper section of the moraine crest was interpreted by Howle et al. (2012) as a fault-propagation fold, formed during the post-Qti-2 interval, along concealed (or blind) branches of the RPF. A few tens of meters northeast of the fold, the Qti-1 and Qta crests are at a slightly lower elevation than the Qti-2 crest, and then both Qti-1 and Qta moraines rise gradually northeast to elevations greater than that of Qti-2. The latter observation indicates growth of the fold (or fault displacement) occurred during the interval between deposition of the Qti-1 and Qti-2 moraines, as well as in post-Qti-2 times.

A fault or faults cut through the Qta moraine 0.6 km (0.4 mi) northeast of the switchbacks on CA 89 (Figs. 8A and 8C). More than one interpretation of the faults is possible. One option is that a branch of the West Tahoe fault (WTF) may continue southeastward from Emerald Point across the bay (where a submerged scarp is present; see below) toward the large Qta moraine remnant northeast of CA 89 (Howle et al., 2012; Figs. 8 and 9). Alternatively, the fault cutting the Qta moraine may be a continuation of a submerged fault marked by the southeast ridge within Emerald Bay (discussed later, Figs. 8 and 9).

Faults near Eagle Point (Fig. 9), which lie in the hanging wall of the WTF, are not discussed further; more work is necessary in that area (see document in Supplemental Material [footnote 1]).

Each fault in the glacial moraines shows evidence of significant post-Tahoe displacement and postglacial displacement (Table 1). As discussed later, faults and scarps cutting the moraines align closely with scarps on the floor of the bay.

Bathymetry and Sedimentology of Emerald Bay


High-resolution bathymetry of Emerald Bay has been mapped in two campaigns, the first in 2009, using techniques described by Howle et al. (2012), and the second, in 2011, 2012, and 2013, during which bathymetric mapping used an autonomous vessel, “SWATH,” built by C. Kitts and students, Santa Clara University, Mechanical Engineering Group (see document in Supplemental Material for details [footnote 1]).

Continuous video images of the bottom of Emerald Bay were scanned along ∼34 traverses across scarps along the bottom of Emerald Bay over 11 days in May and September 2011, 2012, 2013, and 2014, utilizing the remotely operated vehicle (ROV) Triton (Mechanical Engineering Department, University of Santa Clara) (Table 2, Fig. 10).


Multibeam-echosounder bathymetry for Emerald Bay (Figs. 10 and S5A and S5B [footnote 1]) reveals a shallow bottom near the mouth of the bay; the bottom slopes gently (slightly over 3°) southwest toward the deep (>65 m [>200 ft]), relatively flat, central part of the bay. The southwestern end of Emerald Bay between the SRF and the mouth of Eagle Creek is a shallow shelf (∼3–30 m [10–100 ft] deep) upon which a shallow, sandy delta is developing nearshore, and muddy sediment is accumulating south of Fannette Island. Nearly all Eagle Creek sediment that bypasses the delta is routed through the channel south of Fannette Island, because the channel to the north is partially blocked by bedrock promontories and by submerged recessional moraines (Fig. 11; see below). Both glacial till and talus exist in places, especially near steep escarpments.

In the central basin, near the presumed trace of the RPF (Figs. 10 and S5 [footnote 1]), measured water depths range from ∼60–65 m (198–215 ft). The basin floor in that area is flat and featureless and is blanketed by water-rich, muddy sediment, derived from currents passing south of Fannette Island. In the northeastern part of the basin, sand derived from the subaerially exposed lateral moraines blankets the bottom.

The deep, central basin is interrupted along its northern margin by a large, post-Qti-2 landslide (Figs. 4 and 8; Howle et al., 2012). This landslide and several other smaller landslides in Emerald Bay developed along mapped faults and resulted from failure of steep sidewalls of lateral moraines.

Published seismic-reflection profiles (Dingler et al., 2009; Maloney et al., 2013) depict depocenters in places including subbasins east of the SRF and subbasins east of the RPF. In profile section, each depocenter is wedge shaped, with strata thickening to the southwest. These depocenters are interpreted here to be fault-bounded half grabens in the hanging walls of major normal faults. In the subbasin between the SRF and RPF, Maloney et al. (2013; their figure 10; shown schematically in Fig. 13) reported that the sediments may reach thicknesses greater than 25 m (82 ft) (see document in Supplemental Material for discussion). East of the RPF, one subbasin may have a sediment thickness greater than ∼10 m (33 ft), and a more easterly subbasin along the WTF may have a thickness greater than ∼5 m (17 ft) (see also Dingler et al., 2009). New interpretations of relations among these depocenters and mapped faults are discussed below.

Maps and Bathymetric Profiles on Submerged Scarps within Emerald Bay

Submerged scarps within Emerald Bay were first reported by Howle et al. (2012). Because several authors cited earlier have maintained that submerged faults are not evident within Emerald Bay, however, and to verify the existence and nature of scarps and active faults as reported by Howle et al. (2012), all of these features have been examined and mapped using ROV dives, and bathymetric profiles have been constructed from the multibeam-echosounder data using QT mapper. Profiles were constructed perpendicular to scarps at their highest points.

ROV observations reveal that the submerged scarps are generally better preserved than subaerial scarps, probably due to a lack of slope wash and mass wasting of scarps in the lacustrine environment. Debris slopes, colluvial wedges, and wash slopes (Fig. S4B [footnote 1]), which are common to subaerial scarps (McCalpin and Nishenko, 1996; McCalpin, 2009; Wallace, 1977), are not developed on most of the submerged scarps and very poorly developed on others. Talus has accumulated locally near the bases of some high bedrock scarps (Fig. 11), probably a result of freeze-thaw frost wedging along episodically exposed upper parts of the scarps.

ROV Dives and Scarp Profiles along a Submerged Branch of the Mount Tallac Fault (MTF)

Two ROV dives and a bathymetric profile revealed a probable scarp at the eastern edge of the Eagle Creek delta (Figs. 8, 11, and 12A). Along dive 2011-H, the flat, sandy bottom of the shelf, at 22–25 m (73–82 ft) depth, rises gradually westward to 15 m (50 ft) depth. Between 15–10 m (50–33 ft) depth, a possible scarp is in loose sand. Profile 1 (Figs. 8, 11, and 12A) was constructed across the submerged scarp ∼75 m (248 ft) south of dive 2011-H, where the scarp height reaches a maximum. The scarp face has a maximum slope angle of 27°, a remarkably steep face in loose sand, much steeper than typical delta fronts (which rarely exceed 5°–6°; Patruno et al., 2015), suggesting the scarp is a youthful feature. A rough estimate of maximum scarp height from the profile, which includes the (unknown) height of the delta front is ∼17 m (∼56 ft; discussed later). This youthful, submerged scarp at the edge of the Eagle Creek delta aligns with the eastern branch of the MTF to north and south (Figs. 6 and 8). Dive observations and profile 1 suggest that significant postglacial, even Holocene, normal displacement has occurred on the eastern branch of the MTF.

ROV Dives and Scarp Profiles along Submerged Parts of the Stony Ridge Fault (SRF)

The granodiorite bedrock high marked by Fannette Island resembles a classic roche moutonnée (Easterbrook, 1999), with a relatively gentle up-canyon slope and a very steep, down-canyon slope. Howle et al. (2012) indicated that the steep slope facing down canyon also has a tectonic origin, however, because they interpreted two prominent sets of scarps around Fannette Island as fault scarps along the SRF. Maloney et al. (2013; like Howle et al., 2012) depicted a short fault segment along the east side of Fannette Island on their maps but referred to it as part of the “West Tahoe–Dollar Point fault” (a designation with which we disagree, as discussed below). The scarps were examined in this study to determine whether they are related to faults.

Eastern (main) scarps. The eastern scarps, the more prominent of the scarps, were imaged in four ROV dives (Fig. 11), including, from north to south, 2011 dive E, 2012 dive 2A, 2013 dive 5, and 2011 dive A (Fig. 11; Table 2). Bathymetric profiles 4–6 were constructed across this scarp (Figs. 12D–12F).

The eastern scarp (Fig. 11) is ∼200 m (660 ft) in length and extends well beyond the limits of granodiorite bedrock; it dies out ∼100 m (330 ft) north and 50 m (165 ft) southeast of the island. The central part of the scarp is dominated by a steep, east-facing, wall of granodiorite (see photos in Figs. S6A and S6B [footnote 1]) that extends from ∼64 m (210 ft) depth up to lake level and continues nearly 30 m (66 ft) above lake level. Glacial polish on the granodiorite surface was observed at 24 m (80 ft) depth in one dive, indicating that Tioga glaciers covered parts of the scarp.

The nearly horizontal surface in the hanging wall of the eastern scarp is underlain by mud at depths of ∼53–64 m (175–213 ft; Fig. 11). This deposit extends directly to the base of the nearly vertical wall of granodiorite, with no evidence of a colluvial wedge; however, in some places, loose granitic boulders dislodged from the cliff above rest upon mud.

Near the eastern tip of Fannette Island (Fig. 11), the main scarp splits into two, with a high-standing bedrock scarp striking S40°W (220° AZ; past points 2011-A-8 and A-10) and a less prominent bedrock scarp continuing due south; the south-trending scarp face consists of granodiorite and, then, a few tens of meters farther south, of bouldery till.

In some places, lower parts of the main bedrock scarp are mantled with postglacial, bouldery talus (Fig. S6C [footnote 1])—some with angular boulders up to 1.5 m (5 ft) in diameter. Over-steepened slopes have developed in the talus at the base of the scarp, suggesting that the most recent fault displacement postdates accumulation of the talus.

At sites 2011 A-5, A-8, and A-10 at the bases of both scarps, slabs of dark, reddish, iron-oxide–cemented breccia lie within mud (Fig. S6D [footnote 1]). These rocks lack foliation and are interpreted to be spalled-off slabs of crush breccia formed by frictional slip along the fault (see Sibson, 1977). In dive 2012-2A along the main escarpment (Fig. 11), a similar breccia was observed as a steeply dipping veneer on the steep granodioritic face at a depth of ∼45 m (∼150 ft).

At its northern tip (site 2011-E-3), the main scarp is developed entirely in talus shed from the north edge of the island, some with angular boulders up to 1.8 m (6 ft) across. Again, an over-steepened slope in talus indicates that the talus has been displaced by the most recent rupture along the fault. In profile 4 (Fig. 12D) a few meters south of this site, the scarp face in talus has a slope angle near 60°, and the scarp has a vertical separation of ∼5.5 m (18 ft). One hundred meters (330 ft) south of profile 4, the same scarp has vertical separations of 4.5–5 m (15–16 ft) (profiles 5 and 6; Figs. 12E and 12F).

About 100 m (330 ft) southeast of Fannette Island, beneath the south channel, a probable monoclinal (fault-propagation) fold developed in unconsolidated sediments along the SRF (Howle et al., 2012). Water depths range from ∼70 m (231 ft) in the hanging wall to ∼30 m (100 ft) in the footwall, with a scarp-like slope ∼30–40 m (100–132 ft) in height. Several ROV dives traversed this slope in mid-channel southeast of Fannette Island, where the hanging wall at ∼61–64 m (200–210 ft) depth consists of flat-lying muddy sediment. The entire slope (profile 7, near the southern edge of Fig. 11; Figs. 12G and S5 [footnote 1]) is smooth and draped with muddy sediment indistinguishable from that of the hanging wall. The mud is covered with algal masses, is water-rich, lacks strength, and is easily disturbed by propellers on the ROV. Abrupt steps or slope breaks were not observed in the slope. Either the muddy deposits have been draped across a preexisting fault, or the muds have been warped during displacement, as proposed by Howle et al. (2012).

The eastern scarps are indeed fault scarps developed along the eastern branch of the SRF. This is borne out by the facts that the scarps extend north and south well beyond the limits of the bedrock, align with mapped branches of the SRF in the moraines, and sediment in the hanging wall adjacent to the scarp consists of muddy sediment rather than glacial till. In a few places, postglacial, bouldery talus from Fannette Island stands in sharp relief against muddy deposits of the hanging wall (Fig. S6C [footnote 1]), arguing for youthful (postglacial, e.g., post–14 ka) displacement along the scarps. The escarpment formed largely by displacement along the SRF, however, as indicated by the presence of crush breccia along its lower parts. Higher parts of the escarpment comprise a sheer wall of granodiorite, parts of which are glacially polished. The presence of youthful scarps both north and south of the gently sloping escarpment in the mid-channel (Figs. 8A and 11) supports the conclusion that the slope is indeed an expression of a fault-propagation fold.

A published seismic-reflection profile (Maloney et al., 2013; Fig. 13), in addition to depicting a depocenter in the hanging wall of the SRF, also appears to show that the inferred Tsoyowata ash horizon has been displaced. If so, post–ca. 7.7 ka (and pre–5.0 ka) displacement has occurred on this branch of the SRF.

Western scarps projecting through Fannette Island. Prominent scarps both north and south of Fannette Island (Fig. 11), ∼120 m (400 ft), west of the main scarps, are described briefly below.

South side of island. A submerged granodioritic promontory projects southeastward from the western part of Fannette Island (Fig. 11). Its eastern margin is a prominent fault-related scarp >11.5 m (>39 ft) in height (profile 3, Fig. 12C). The hanging wall of this scarp is a gently sloping, nearly flat surface at 30 m (100 ft) depth, underlain by sand. The lower part of the scarp face exposes bouldery talus resting in places upon till. Locally, the talus forms an over-steepened slope and appears to have been displaced during a youthful slip event.

North side of island. A steep, east-facing escarpment (Fig. 11) lies beneath the north channel on the north side of Fannette Island. This scarp, which is underlain by granodiorite, has a height of 15.5 m (51 ft) and a VS of 11 m (36 ft) (profile 2, Fig. 12B). The lower part of the scarp face is a steep, bouldery talus slope (with a slope angle up to 35°). As on the scarp to the south, the steep, sloping surface of the talus in places suggests that the deposit has been cut or deformed by displacement on the fault.

On Fannette Island, exposed granodiorite is intensely fractured but lacks any obvious topographic expression of a west-dipping normal fault. This observation suggests that any pre-glacial bedrock scarp on the upstream side of the roche moutonnée surface may have been eroded smooth by Tahoe and Tioga glaciers, and that post-Tioga displacements are not detectable in the fractured granodiorite.

As with the eastern (main) set of scarps, the submerged scarps projecting through the western part of Fannette Island clearly represent fault scarps both because they cut talus deposits and because they align with prominent scarps developed in moraines along the western branch of the SRF (see Figs. 6A, 8A, and 11).

ROV Dives and Profiles on Submerged Scarps along the Rubicon Peak Fault (RPF)

The large, post-Qti-2 landslide near the north shore of Emerald Bay (mentioned earlier; Fig. 14) has two sets of submerged, north-trending, east-facing, scarp-like features developed in glacial till that was displaced from the left-lateral Qti-2 moraine. These scarps have been examined with the ROV to determine whether they are fault-related topographic features (Howle et al., 2012) or are related to landslide lobes.

The western set of east-facing scarps (Fig. 14) begins ∼20 m (66 ft) southeast of the pier at the boaters’ campground continues, southward within the central part of the landslide, and includes three southwest-trending splays near its southern end. About 125 m (412 ft) east of the western scarps, the eastern scarps, which are near the eastern exposed limits of the landslide, include three discontinuous, arcuate, east-facing segments. Both sets of scarps continue to the southern exposed limit of the landslide but cannot be traced into deeper water, suggesting that the scarps predate the late Holocene muds of the central basin.

ROV dives and profiles on the western set of scarps. The scarps are abrupt, east-facing walls of bouldery till, typically 3 m (10 ft) or more in height (see photos in Figs. S7A and S7B [footnote 1]). The hanging wall east of the scarps is a smooth, flat, sandy surface, with a gentle slope to the east and south, developed upon the landslide. Westward, beyond the top of the scarp, the bottom (footwall), which is the surface of the landslide, is again level, smooth, and mantled by sand.

Bathymetric profiles indicate that all of the scarps have typical fault geometry, with flat, smooth surfaces on the hanging walls and footwalls and abrupt, steep scarp faces. Scarp profiles 8 and 9 (Figs. 14, 15A, and 15B) reveal scarp faces with maximum slopes of 18° to 39° and vertical separations (VS) of 3.5 m (11 ft) and 2.5 m (8 ft). The scarps near the southern edge of the exposed landslide are less prominent (Figs. 15C and 15D).

ROV dives and profiles on the eastern set of scarps. The scarp faces are held up by abrupt, steep walls of granodioritic boulders varying from ∼0.5–2.5 m (1.6–8 ft) in diameter (Figs. S7C and S7D [footnote 1]) in glacial till displaced onto the floor of the bay by the large landslide in Figure 14. In places, loose boulders have tumbled down from the scarp face and rest upon sand at the base of the scarp. A possible bevel at the top of one scarp (Fig. S7D) suggests that more than a single slip event may have produced the scarp. A flat, smooth, sandy bottom east of the arcuate scarps (e.g., in the hanging wall) slopes gradually to the east and south. The footwall west of the scarps is a flat, smooth surface underlain by sand. In bathymetric profiles (Figs. 15E–15G), these scarps, like the western scarps, all have geometry typical of fault scarps, with flat surfaces on hanging walls and footwalls and with abrupt, steep scarp faces (maximum slope angles of scarp faces range from ∼21° to 32°). The arcuate scarp segments have vertical separations of 2 m (7 ft), 4 m (13 ft), and 6.5 m (21 ft), increasing in magnitude from north to south.

Interpretation of the scarps in the large landslide. The ROV data and bathymetric profiles support the conclusion that these submerged scarps are fault scarps. The surface of the landslide, which is mantled by sand, is relatively flat and smooth, with no evidence for lobes within the landslide. The submerged scarps also correlate favorably with mapped scarps and faults along the RPF within moraines to the north, and to a feature in a profile of Maloney et al. (2013; Fig. 13) interpreted here as a buried branch of the RPF. Because the scarps developed in the large landslide derived from the left-lateral Qti-2 moraine (Howle et al., 2012), the scarps postdate at least part of the deglaciation (e.g., they are post–21 ka and possibly post–14 ka). Furthermore, the facts that scarp faces are steep, no colluvium mantles their lower parts, and postglacial sand mantles both hanging walls and footwalls but not the scarp faces themselves also attest to youthful, postglacial displacement.

ROV Dives and Profiles across the Southeast Ridge

A bathymetric feature in the east-central part of Emerald Bay and here referred to as the “southeast ridge” (see Figs. 6A, 8A, and 10) was interpreted as a north-facing fault scarp by Howle et al. (2012). The ridge is linear, ∼0.2 km (660 ft) in length, and trends west-northwest, perpendicular to Qti recessional moraines along the right-lateral moraine complex (Fig. 8A). ROV observations support the interpretation of Howle et al. (2012), because bouldery Qti till is exposed along the scarp face. Several unusual features of this scarp include its short, exposed length, its linearity, and the fact that it strikes at a high angle to the RPF. It also shows a gently south-tilted footwall slope. LiDAR images suggest that a continuation of this fault to the southeast may disrupt and offset crests of Qti-2 recessional moraines (Fig. 8A; discussed earlier); conceivably, the southeast ridge could be a link between the RPF and the southern part of the WTF.

Profiles and ROV Dives along Scarp Possibly Related to the West Tahoe Fault (WTF)

The bedrock escarpment along the West Tahoe fault (WTF) near Emerald Point lacks bathymetric expression within Emerald Bay. However, Howle et al. (2012) mapped a short, ∼140-m-(460-ft)-long, submerged, northwest-striking scarp (shown in Figs. 8A and 9) on the flat bottom of the bay ∼0.8 km (0.5 mi) northeast of the submerged landslide discussed above, as part of the WTF. Our observations support that conclusion. In plan view, the scarp is slightly arcuate, with at least two subtle steps. ROV dives (2014 dives 1 and 2) revealed that this scarp has a steep northeast flank (scarp face) exposing fresh Qti till, with prominent granodiorite boulders, and a gently sloping southwestern flank. Smooth, postglacial sand mantles both the footwall and hanging wall. Profile 17 (Figs. 8A and 15J) shows a vertical separation of only ∼1 m (3 ft) and a scarp-face slope angle of 20°. A few tens of meters north of the profile, the scarp height is 2 m (7 ft). Development of this scarp postdated Qti-2 (post–21 ka), but it is unknown if it postdated complete deglaciation at 14 ka. Along strike ∼200 m (660 ft) southeast of the scarp, a kink along the crest of an onshore Qti recessional moraine (see Fig. 9) may represent deformation related to displacement on the WTF.

Summary of ROV Dives and Detailed Maps

ROV dives, together with echosounder bathymetry, confirm the fault mapping and conclusions of Howle et al. (2012) and in particular the continuation of active branches of the MTF, SRF, RPF, and WTF beneath the floor of Emerald Bay (Fig. 16). Youthful scarps exist along all the faults, and crush breccia is preserved along the SRF.

Comparison of Results with Seismic-Reflection Profiles in Emerald Bay

As noted previously, some authors have stated that active faults do not exist within Emerald Bay (Dingler et al., 2009; Kent et al., 2006), based upon unfaulted, horizontal layering in seismic-reflection profiles. Although precise locations and scales for the profiles were not provided, enlargements of the seismic profiles have been compared with new bathymetric images and the locations of scarps described above.

The seismic-reflection profile sketched in Figure 13 (from Maloney et al., 2013), as noted earlier, appears to image the eastern branch of the SRF and possibly contains evidence for post-Tsoyowata displacement. This profile also appears to intersect a branch of the RPF and depicts a wedge-shaped deposit in the hanging wall of the fault. Other seismic profiles of Dingler et al. (2009) and Kent et al. (2006) image the large postglacial landslide cut by the RPF discussed above (Fig. 14) as an amorphous mass with no internal reflections. The lack of a well-layered succession in the landslide makes recognition of faults problematic, because in seismic-reflection profiles, the existence of steeply dipping faults is usually inferred from offsets of planar layering (Alcalde et al., 2017; McCalpin, 2009; Yeats et al., 1996). Subtle bathymetric steps in the top surface of the landslide are evident in the seismic profiles, however, and these most likely are the scarps shown in Figure 14. Prominent, unfaulted, horizontal layering in other parts of the seismic profiles consists dominantly of sediment younger than ca. 8 ka (Maloney et al., 2013). From this comparison, the seismic-reflection profiles appear to be compatible with the existence of the scarps and faults described by Howle et al. (2012) and this study.

Estimates of Slip Rates

Heights of scarps and vertical separations have been estimated in many places in this study. Most scarps in the glacial moraines have been excluded from VS analysis, however, because, in most cases, the hanging wall has been backtilted toward the fault scarp. In such cases, VS cannot be determined precisely (Howle et al., 2012; McCalpin, 2009).

There are many uncertainties in determinations of VS, including correct placement and dip of faults (see discussions in Rood et al., 2011a; Wesnousky et al., 2005; also see Fig. S4 [footnote 1]). In this study, normal faults are assumed to dip an average of ∼60°, following many other published works (e.g., Gilbert, 1890, 1928; Howle et al., 2012; Personius et al., 2017; Rood et al., 2011a; Wesnousky et al., 2005); this value is consistent with bedrock scarp faces in Emerald Bay that dip 50°–75°. Howle et al. (2012) used 3D modeling and three-point solutions to obtain a mean of 62 ± 12° for seven measurements of fault dip along the RPF.

The exact point of intersection of a fault surface with a scarp also may affect the VS estimate, especially in cases where hanging-wall and footwall surfaces are not parallel. Many other studies in unconsolidated deposits have assumed that fault surfaces project to midpoints or steepest parts of scarps (Howle et al., 2012; Koehler and Wesnousky, 2011; Personius et al., 2017; Wesnousky et al., 2005). In this study, most submerged scarp faces expose bedrock, talus, or glacial till, and colluvial wedges do not cover the lower parts of scarps. Therefore, fault surfaces are projected in most cases to intersect the profiles near the bases or toes of the scarps (Figs. S4C and S4D [footnote 1]).

Below and in Table 3, for estimation of vertical separation rates (VSR), an age of 21 ka is assumed for Qti-2 moraines, and 14 ka is assumed for the age of final deglaciation.

Approximate vertical separation rates are provided below, using estimated vertical separations and broad age estimates, together with previously published rates. The estimates from this study, which are conservative minimum values, are approximate, owing to uncertainties in age constraints and factors mentioned above. Rates therefore are reported to one significant figure only (e.g., 0.36 mm/yr is reported as 0.4 mm/yr, etc.). Nevertheless, these data provide a basis for comparison with other published data for slip rates within the Lake Tahoe basin and the Walker Lane belt.

Mount Tallac Fault (MTF)

Several strands of the MTF lie within bedrock west of Emerald Bay, but youthful markers have not been identified along these branches, and therefore it is unknown if youthful displacements have occurred. However, latest Quaternary to Holocene displacement likely occurred along the eastern branch of the MTF where it bounds the Holocene delta of Eagle Creek (Figs. 4, 8A, and 11). There, the prominent scarp (Fig. 12A, profile 1) is assumed to be post–14 ka (e.g., postglacial retreat).

Because the height or relief of the unfaulted or unmodified delta and the foreset slope are unknown, VS on this scarp is tentatively approximated as follows (see profile 1, Fig. 12A). The slope break is reconstructed by extending the footwall surface and the scarp face until they intersect. Two 6° sloping lines (defining an upper limit to delta foreset slopes [Patruno et al., 2015]) are constructed from the slope break and base of the scarp. An approximate VS of ∼13 m (44 ft) is obtained as the vertical height between the 6° sloping lines. A rough VSR estimate of ∼0.9 mm/yr is then obtained. This is three times the VSR of 0.3 ± 0.1 mm/yr (post–23.5 ka) at a youthful scarp along the MTF, ∼5 km southeast of Emerald Bay, between Cascade Lake and Fallen Leaf Lake (Fig. 18; Howle et al., 2012). At Emerald Bay, owing to the uncertainties in estimating VS, half of the estimated VSR, 0.45 mm/yr (rounded to 0.5 mm/yr), is taken as a conservative estimate of minimum VSR for the eastern branch of the MTF.

A VSR estimate of 1.4 ± 0.7 mm/yr was reported for the MTF where it forms a very high scarp on the right-lateral Tioga moraine of Cascade Lake (Pierce et al., 2017; Fig. S3 [footnote 1]; those authors considered this scarp to be the “West Tahoe fault”). At that site, we mapped three scarps ranging from 1–7 m (3–23 ft) high in the Qti-2 moraine (Fig. S3; Howle et al., 2012). As discussed in the document in the Supplemental Material, it seems likely that the very high bedrock scarp discussed by Pierce et al. (2017) represents both pre-and post-Tioga displacement.

Eight km (4.8 mi) southeast of Emerald Bay, Howle et al. (2012) concluded that the submerged faults at the head of Fallen Leaf Lake are splays of the MTF. On one of these faults, an estimated “vertical deformation” rate of 0.4–0.7 mm/yr was reported by Brothers et al. (2009; Fig. 18).

Stony Ridge Fault (SRF)

No previous estimate of VSR has been made for the SRF. Some scarps along the submerged eastern branch of the SRF are younger than the 14 ka deglaciation. The vertical separation of 5.5 m (18 ft) at the north end (Figs. 11 and 12D, profile 4) and a limiting age of 14 ka produce a minimum VSR of 0.4 mm/yr along the fault (Table 3). Actual rates may be somewhat higher, because Holocene displacement may have occurred (see discussion above). The morphological similarity of scarps along the submerged western branch of the SRF to those on the eastern branch suggests that both branches have similar VSRs. If so, then the combined, minimum, VSRs from both branches would be ∼0.8 mm/yr.

Rubicon Peak Fault (RPF)

For a conservative estimate of VSRs, the submerged scarps are assumed to be younger than 21 ka, the age for Qti-2. On the western branch of the RPF (Figs. 14 and 15A–15C), scarps have VS up to ∼3.5 m (11 ft), yielding a 0.16 (0.2) mm/yr (post–21 ka) VSR. On the eastern branch (see Figs. 14 and 15E–15G), submerged scarps have VS of up to 6.5 m (21 ft), yielding a minimum VSR of 0.3 mm/yr (post–21 ka). The estimated VSRs on the submerged scarps combine to give an aggregate minimum VSR of ∼0.5 mm/yr (post–21 ka). This result is consistent with the estimate of Howle et al. (2012) where the RPF displaced the left-lateral Qti-1 moraine of Emerald Bay.

Southeast Ridge

Scarp heights along the southeast ridge, a scarp cutting post–Qti-2 till, range from 3.5–5.5 m (11–18 ft; see Figs. 15H and 15I). VS cannot be estimated directly, however, because the footwall has been tilted to the south. A very rough estimate of VS of ∼3 m (10 ft) and a maximum limiting age of 21 ka for Qti-2, yield a minimum VSR of 0.1 mm/yr (with large uncertainties).

West Tahoe Fault (WTF)

Estimates of VSR have not previously been reported for the southern part of the WTF. The Qti-2 moraine northwest of Emerald Point (Fig. 7) has at least three scarps with approximate heights ranging from ∼2–3.5 m (7–11 ft). The hanging wall is backtilted, however, and no estimate of VSR is made there. The youngest Qti-2 recessional moraine (Fig. 7) has been cut by two scarps, ∼2 m and 1.5 m (7 ft and 5 ft) in height (because the crests are approximately horizontal, SH there approximates VS). Using a limiting age of 21 ka and a combined VS of 3.5 m, a minimum VSR of 0.2 mm/yr is obtained for the WTF at Emerald Point. The post–21 ka submerged scarp along the WTF, with a VS of ∼1 m (3 ft) from profile 17 (Fig. 15J), yields an approximate minimum VSR of <0.1 mm/yr.

For the WTF, ∼12 km (7 mi) north of Emerald Bay near Sugar Pine Point, a “vertical deformation rate” of 0.43–0.81 mm/yr has been reported (location shown in Fig. 18; Brothers et al., 2009; Dingler et al., 2009). If the fault surface is projected toward the midpoint of the scarp (Fig. 17), an improved estimate of VS is ∼8 m (26 ft). This fan surface, its associated canyon, and incised channels formed by backflow during and after a major tsunami (Moore et al., 2014), which gives limiting ages of 21–12 ka. These data (∼8 m VS in 21 ka) yield an improved estimate of minimum VSR of ∼0.38 (rounded to 0.4) mm/yr for the WTF at this site.

Rates for TSFFZ versus WTF and GPS Geodesy

At Emerald Bay, each major fault in the TSFFZ has a VSR greater than the WTF. Taken together, the various segments of the TSFFZ have an estimated combined vertical separation rate up to three times greater than that for the WTF. Available data also suggest that VSRs for both the RPF and the WTF decrease southward; this is consistent with the fact that these two fault zones die out as mappable scarps a few kilometers southeast of Emerald Bay.

Minimum extension rates along two transects of the TSFFZ, one near Meeks Bay and the other at Emerald Bay, calculated from VSR data (Table 3) are comparable, ∼1.1–1.2 mm/yr, in a direction N50–55°E (050–055° AZ; Fig. 18, inset). How do these estimates based upon geologic data compare to modern GPS results?

Modern strain rates across the Lake Tahoe basin based on GPS geodesy are ∼0.8–1.1 mm/yr in a direction S45°E (145° AZ) (Hammond et al., 2011; Wesnousky et al., 2012). Notably, the extension directions for the two geological transects are at right angles to strain directions calculated from geodesy. Wesnousky et al. (2012) noted that the GPS data provide evidence for ongoing dextral displacement between the Sierra Nevada microplate and ranges to the east of the Lake Tahoe basin, despite the fact that there is little direct evidence for strike-slip displacements in the region. They also observed that directions and magnitudes of fault slip from geologic data do not agree with GPS results, a conclusion supported by results here.


New observations demonstrate that the TSFFZ continues through the Emerald Bay area, is active, and owing to its large displacement, the fault zone represents the active structural boundary of the Sierra Nevada microplate in the region around the Lake Tahoe basin. These new results contradict statements and conclusions of several previous studies (e.g., Brothers et al., 2009; Dingler et al., 2009; Kent et al., 2006, 2016; Maloney et al., 2013; Schmauder, 2013; and Smith et al., 2013) that discounted the existence or importance of the TSFFZ.

As noted by many previous studies, the TSFFZ includes geologic features typical of major range-bounding normal fault systems throughout the Basin and Range Province. Such features include large topographic relief along the eastern front of the Sierra Nevada, high-standing bedrock facets, topographic control of east-flowing glacial valleys, and glacial moraines at the mouths of bedrock glacial valleys. The authors of the papers cited above were unaware of or disregarded the work of Howle et al. (2012), who made a compelling case for activity along the TSFFZ.

The TSFFZ, which has set the kinematic boundary conditions for transtension in the Lake Tahoe region, has had a long and complex history, as described briefly in an introductory section. Kinematic changes may have occurred at various times along this fault zone and may have led to some of the structural complexity. As interpreted above, during the Pleistocene, after large normal displacements had formed a high, northeast-facing escarpment, minor dextral displacement appears to have occurred along some of the faults. Normal slip then continued during and after the Tahoe glaciation and continues to the present.

The WTDPFZ is clearly younger than the TSFFZ, because 2–2.3 Ma basaltic volcanic rocks and lacustrine sediments are in both hanging-wall and footwall positions on the WTDPFZ (Kortemeier et al., 2018; Schweickert, 2009; Schweickert et al., 2004). Yet both fault systems appear to have been active during late Pleistocene and Holocene times (Howle et al., 2012; this study).

The evolution of these two normal fault systems has important implications for the migration of activity and evolution of range-front fault systems in the Basin and Range Province, supporting the conclusions of Wallace (1984, 1987). Although it is commonly assumed that range-front normal fault systems have a clear progression of activity from the range front toward the adjacent basin along with deactivation of the range-front system, Wallace’s (1984, 1987) work and this study reveal that some range fronts have a much more complex pattern of evolution, with activity both migrating from range front into the basin but also with concurrent activity occurring within both systems.

Within the Basin and Range Province and, in particular, the Walker Lane belt, discrepancies commonly exist between geodetically determined strain rates and slip rates determined from geologic measurements along component range-bounding faults (e.g., Bormann et al., 2016; Gold et al., 2014; Herbert et al., 2013; Lifton et al., 2013; Personius et al., 2017, Wesnousky et al., 2005, 2012). This study, like many previous reports, emphasizes the fact that geodetic strain rates do not necessarily reflect long-term slip rates on discrete faults.

Taylor and Dewey (2009) noted that, with the N35–45°W (320° AZ)–trending TSFFZ setting the kinematic boundary condition and a Sierra Nevada transport direction of ∼N50–60°W (305° AZ; Fig. 1B), the maximum instantaneous stretching direction (Xi) should be oriented roughly east-west. As noted by Taylor and Dewey (2009), the N-S–trending WTDPFZ is a normal fault zone formed perpendicular to Xi.

The reconciliation of long-term, northeast-southwest extension along the TSFFZ with northwest-southeast dextral transtension is still problematical. For a region ∼100 km in width involving seven N-S–trending Walker Lane basins, including the Lake Tahoe basin, Wesnousky et al. (2012) invoked a Ramsay-style brittle-ductile, dextral shear zone model (Ramsay, 1967; Ramsay and Huber, 1983). Such a shear zone model involves plane-strain, simple shear. Some modification of that model may be required because northwest-directed transtension involves constrictional strain rather than plane strain (Dewey, 2002; Taylor and Dewey, 2009) and requires a N-S component of shortening. Additionally, in a shear-zone model, all N-S–striking normal faults in the region, which may reflect the east-west Xi as in the Lake Tahoe basin (Taylor and Dewey, 2009), should have rotated progressively clockwise.

Alternatively, transtension in the region may be viewed as heterogeneous, with domains of extension alternating with zones of strike slip. Historic earthquakes in the Lake Tahoe region suggest that transtensional deformation is currently partitioned into domains of normal and conjugate strike-slip faults (Fig. 2B; Schweickert et al., 2004). The normal fault domains (such as the Lake Tahoe basin) accommodate east-west or east-northeast extension, and adjoining domains of conjugate strike-slip faults may accommodate the northerly component of displacement.


Extensive field mapping, analysis of LiDAR data and multibeam-echosounder bathymetric data, and numerous ROV dives in Emerald Bay have been combined to determine the character, relative age, and offsets along scarps of the three faults (and their branches) of the Tahoe-Sierra frontal fault zone (TSFFZ) and the WTF (southern continuation of the West Tahoe–Dollar Point fault zone [WTDPFZ]). These are the most detailed structural maps to date of glacial moraines and parts of the lake bottom in the entire Lake Tahoe region.

The new data confirm previous reports (e.g., Howle et al., 2012) that several important, northwest-striking, normal faults (Mount Tallac [MTF], Stony Ridge [SRF], and Rubicon Peak [RPF] faults, etc.) do pass beneath Emerald Bay and displace both lake-bottom sediments and glacial moraines. Additionally, the SRF is confirmed as an active fault lying in the footwall of the Rubicon Peak segment of the TSFFZ. The TSFFZ now has the most dense array of VSR data of any part of the microplate boundary. The continuity of the faults mapped by Howle et al. (2012) precludes any direct connections between the WTF and the Mount Tallac fault (MTF). Significantly, both glacial moraines and bottom sediments of Emerald Bay provide excellent records of late Quaternary to Holocene fault displacements. Submerged scarps, in particular, are exceptionally well preserved. Faults of the TSFFZ have a combined VSR of about 2.1 mm/yr.

The TSFFZ, a complex zone with several en echelon segments, each with numerous subparallel branches, forms the eastern edge of the rigid Sierra Nevada microplate. Geologic relations and patterns of range-front normal faulting indicate that the TSFFZ is the older, more long-lived normal fault system. The WTDPFZ is a relatively simple zone with a single main trace.

The fact that both fault zones have been active during late Quaternary to Holocene times has important implications for the migration of activity of range-front fault systems, emphasizing the importance of conclusions of Wallace (1987), who described examples from the Basin and Range Province (BRP) both where fault activity migrated basinward from range fronts and where fault activity has jumped back and forth from range front to the basin and back through time.

Howle et al. (2012) noted that the TSFFZ is important both for youthful normal fault displacements and for potential seismogenic earthquake ruptures in the Lake Tahoe basin. The activity of faults within the TSFFZ is important for another reason. During normal fault earthquakes, ground motion in hanging walls may be significantly greater than in footwalls (e.g., Anderson et al., 2000; Brune and Anooshehpoor, 1999). Because the head scar of the 10–12.5 km3 McKinney Bay (mega)landslide (Fig. 18) lies within the hanging wall of the TSFFZ (and is west of the WTDPFZ), one or more large slip events along the TSFFZ may have triggered the large-scale collapse event(s) and resultant megatsunami (Moore et al., 2014).


Many people contributed to this project. In particular, we gratefully acknowledge James Howle for his assistance in providing detailed LiDAR and sonar images and in preparing numerous scarp profiles for us. We also are grateful to Brant Allen, Tahoe Research Group; Jamie McCaughey, former M.S. student at University of Nevada, Reno (UNR); undergraduate and graduate mechanical engineering students at Santa Clara University; undergraduate students and graduate teaching assistants in UNR geology field camps in 2003, 2004, 2007, 2008, and 2009; and Brent von Twistern for acquisition of the detailed echosounder map of the floor of Emerald Bay. Financial support for ROV work from Santa Clara University is also gratefully acknowledged. Several anonymous reviewers and Geosphere editors provided comments that materially improved this report.

1Supplemental Material. Notes on geologic setting, mapping criteria for glacial moraines, tectonic geomorphology, new mapping of moraines of Emerald Bay and Cascade Lake, high-resolution bathymetry images, ROV observations. Figures show topography and bathymetry, comparisons of submerged scarps with subaerial scarps, and photos of submerged scarps. Please visit https://doi.org/10.1130/GES02022.S1 or access the full-text article on www.gsapubs.org to view the Supplemental Material.
Science Editor: Raymond M. Russo
Associate Editor: Jose M. Hurtado
Gold Open Access: This paper is published under the terms of the CC-BY-NC license.