Sediment cores were collected from five locations in Jenny Lake in 2013, 2015, 2016, and 2021 using multiple types of piston coring systems deployed on cables from the frozen lake surface (Figure 2; Supplementary Table S1; for details of field and analytical methodology, see SI; Larsen et al., 2016; Larsen et al., 2019). The sediment sequence between all sites can be correlated using distinct changes in lithostratigraphy. This study focuses primarily on the Central Basin site (JEN16-1 and JEN21-1) (Figure 2). In 2016, one pair of offset and overlapped cores (JEN16-1A and JEN16-1B) were recovered from this location through repeated 2-m drives using a Uwitec coring system. Following collection, density and magnetic susceptibility were measured at continuous 5 mm intervals using Geotek core scanners at the National Lacustrine Core Facility (Laccore), University of Minnesota. In 2021, a third core (JEN21-1) was collected from this site using similar techniques to extend the composite record deeper into the sediment fill. The JEN16-1 and JEN21-1 cores were sub-sampled and analyzed for dry bulk density, total organic matter content (OM), and sand concentration. Total OM (weight % OM) was measured at continuous 2 cm intervals using loss on ignition (LOI) at 550°C for 4 h. To determine sand concentration (weight % sand grains >125 μm), 3 cm³ of wet sediment was extracted from the cores using a volumetric sampler at continuous 5 cm intervals and at visible disturbance horizons. The sub-samples were transferred into beakers and submerged in concentrated 30% hydrogen peroxide for 24–48 h to disaggregate the sediment and dissolve organic material before being washed and wet sieved through a 125 μm sieve and allowed to dry. The concentration of grain sizes >125 μm (fine sand) was determined by dividing the mass of the residual sediment by the initial sample volume.

Turbidite deposits in Jenny Lake sediments were identified and described by Larsen et al. (2019) using visual observations and changes in sediment physical character. Here, we perform additional analyses of the six largest turbidites (those greater than 5 cm thick) common to all Central Basin (JEN61-1 and JEN21-1) cores to evaluate fine-scale changes in grain size distributions, mineralogical composition, and sedimentology. Grain size distributions through each turbidite deposit were measured at 5 mm intervals using a Malvern Mastersizer laser diffraction particle size analyzer coupled with a wet dispersion unit at the California State University at Fullerton and at Occidental College. At each interval, approximately 1 cm³ of sediment was extracted and pretreated to remove organic material using oxidation treatment through repeated baths in 30% hydrogen peroxide solution heated to 150°C. Once organic material was removed, samples were rinsed with distilled water and wet-transferred into beakers prior to being analyzed. An internal standard was run after every 10 samples to verify instrument accuracy and repeatability (Fullerton standard: silica carbide powder (n = 96, median grain size = 13.1 μm, σ = 0.03 μm), Occidental standard: boron carbide powder (n = 54, median grain size = 12.2 μm, σ = 0.02 μm)). Statistical parameters of the grain size distributions, including skewness and geometric standard deviation (GSD; a measure of grain sorting) were calculated using the following equations: G S D = σ = ∑ X i d i −   d ¯ 2 ∑ X i S k e w = ∑ X i d i −   d ¯ 3 σ 3 ∑ X i where X i = the number of particles in the ith size class, d i = the median size of the ith size class, and d ¯ = the mean particle size.

Changes in elemental abundance through the turbidite deposits were determined at 2 mm resolution by scanning X-Ray Fluorescence (sXRF) using an Avaatech core scanner at the Scripps Institution of Oceanography, University of California at San Diego. XRF measurements were made at voltages of 10 kV using an exposure time of 10 s and current of 500 mA (Addison et al., 2013). We performed a Principal Component Analysis (PCA) that incorporated sediment grain size metrics (D50, sand %, clay %, silt %, geometric standard deviation (GSD), and skew), elemental abundance (K, Si, Al, Ti, Fe, Ca), density, and magnetic susceptibility data for each turbidite event using the PCA function within the R package FactoMineR with default settings (Lê et al., 2008; see SI text). To directly compare variations of the different data through each deposit, all properties were placed on a common depth scale of 2 mm-intervals by interpolating the MS and grain size data using the approx function in R. This function eliminates the possibility of introducing bias into the PCA by not including missing values.

The ages of individual turbidites at Jenny Lake were first determined by Larsen et al. (2019) by dating material deposited within 200 years of the top and/or bottom contact of each deposit. Here, we improve the age control at the JEN16-1 site by adding six additional radiocarbon ages and transfer the updated chronology to the new JEN21-1 core (for additional geochronological information and methods, see SI text, Supplementary Figure S5; Supplementary Tables S2, S3).

Sediment cores recovered from Jenny Lake range in length and approximate basal age (Supplementary Table S1). Cores collected from sites in the deep central basin (JEN16-1 and JEN21-1) and outer shelf (JEN15-4 and JEN13-2) penetrate down to strongly laminated sediments containing ice-rafted debris (IRD), indicating nearly complete recovery of the postglacial interval (Figure 3). These cores reveal a coherent two-component lithostratigraphic sequence of clay-rich, laminated or massive light gray to gray glacial-lacustrine sediments overlain by an upper unit of mostly homogenous or faintly laminated olive brown to brown non-glacial sediments (Figure 3; see Supplementary Figures S1–S4 for individual core images). A gradational contact separates these two sediment packages and reflects the decreasing influence of glacial erosion on lake sedimentation and shift toward non-glacial conditions at the end of the Pleistocene. The depth to this transition and other stratigraphic markers increases with water depth and proximity to the inflow delta, confirming that sediment accumulation rates in the central basin are greater than the surrounding shelf and that Cascade Canyon has been the dominant source of sediment to the lake.

Interstratified within the general lithostratigraphic sequence are two prominent rhyolitic air-fall tephra layers and a series of other event deposits (Figure 3). The tephra layers have been geochemically characterized and assigned to the Mazama (7.6 ka; Egan et al., 2015) and Glacier Peak (13.6 ka; Kuehn et al., 2009) ash beds (Larsen et al., 2016). These tephra layers aid in the stratigraphic correlation between cores taken throughout the lake. The other event deposits range in thickness, sedimentary texture, and frequency. At site JEN16-2, proximal to the inflow delta, the upper sediment sequence contains numerous sand layers ranging in thickness from 1 to 10 cm and composed predominantly of fine to coarse sand grains with abundant woody terrestrial plant material (Figure 3; Supplementary Figure S2). The layers do not present consistent sedimentary structure or grading but tend to have a moderately sharp basal contact of coarse sand and a more diffuse upper contact where organic material (e.g., twigs, conifer needles, cone fragments, etc.) contained in a sandy matrix grade into the overlying host sediments. Apart from these qualities, the organic-rich sand deposits do not show a strong stratigraphic pattern and are absent in cores from other sites, indicating these chaotic deposits are spatially restricted to the region near the inflow delta.

In addition to the sand layers at JEN16-2, all core sites contain well-defined, normally-graded turbidite deposits (Figure 3). These deposits range in thickness and are easily correlated between sites around the lake based on their characteristic stratigraphic expression and the individual chronologies respective to each core. The relative thickness of individual turbidite deposits is greatest in the central basin and at site JEN16-2, where they are up to 48-cm-thick (Figure 3). The cores taken from the distal eastern and southern regions of the outer shelf contain fewer turbidites and they are relatively thin compared to their correlative deposits in the central basin. One exception to this trend is the thick turbidite (39-cm-thick) present in the glacial unit at JEN15-4 (Figure 3).

Average sediment density values and OM concentrations at the Central Basin site are 1.6 ± 0.3 g/cm³ (range: 1.1–2.2 g/cm³; n = 2097) and 6.1% ± 3.8% (range: 0.5%–17.5%; n = 843), respectively (Figure 4). The concentration of siliciclastic sand (grains >125 µm) ranges from 0 to 1.5 g/cm³ (n = 304). A long-term pattern of decreasing density and increasing organic content toward the core top tracks changes in seismic stratigraphy and core lithostratigraphy (Figure 4). In general, the glacial unit is characterized by dense, minerogenic sediments deposited during the overall phase of deglaciation. The lithologic transition from glacially influenced sedimentation to non-glacial sedimentation that occurs near the start of the Holocene reflects catchment stabilization and reduced minerogenic sediment flux (Larsen et al., 2016).

The most notable features of the Jenny Lake sediment record are the rapid changes in density, sand content, and organic matter associated with the series of turbidites (Figure 4). These deposits are identified in core section by recognizable changes in sediment texture. In seismic imagery, the turbidites are expressed as a series of high-amplitude seismic reflectors that can be traced around the basin (Figure 4; Larsen et al., 2019). Abrupt increases in sand and density and a decrease in organic content occur at the base of each turbidite. For most of the events, sediment density increases from background values of approximately 1.4 g/cm³ to maximum values of approximately 2.0 g/cm³. Density values generally remain higher than background through the entirety of each deposit. Similarly, sediment OM concentration decreases from background values at the base of each turbidite, remains low through the deposits, and immediately returns to background values in overlying sediments (Figure 4). One exception to this trend in OM content is the thick event deposit at approximately 2.6 m depth, which contains a brief peak in OM that is associated with interstitial woody plant remains visible in core section (Figure 4). Outside of the defined spikes in sand and density associated with the turbidite deposits, sediments at this site in the central basin contain negligible sand content and background sedimentation during the non-glacial interval is dominated by deep-water hemipelagic silts (Figure 4).

The six most prominent turbidite deposits at the Central Basin site (JEN16-1 and JEN21-1) have thicknesses of approximately 34 cm, 8 cm, 12 cm, 12 cm, 24 cm, and 6 cm, and occurred at 13.9 ± 0.6 ka, 12.9 ± 0.1 ka, 10.3 ± 0.2 ka, 9.1 ± 0.1 ka, 8.4 ± 0.1 ka, 7.7 ± 0.1 ka, respectively (uncertainty reported as ± 2 SD) (Figure 5, Supplementary Tables S1, S3). We note that, despite the addition of six new radiocarbon ages, the calculated age and uncertainties for these deposits are statistically identical to those previously reported by Larsen et al. (2019), with only nominal differences. For clarity, we apply these turbidite ages when referring to the individual event deposits (Figure 5).

In all cases, the top and bottom contacts of the event deposits are sharp and clearly defined from background sediments by changes in textural properties and grain size distributions (Figure 6). Changes in the fractional abundance of grain sizes and median grain size (D₅₀) also define a stratigraphic pattern within each event deposit composed of 1) a thick coarse-grained basal unit of mostly sand, 2) a middle unit of mostly silt and very fine sand, and 3) a comparatively thin, fine-grained cap unit made of very fine to fine silt and clay (Figure 6). The boundaries between successive units are relatively sharp indicating strong segregation of particle sizes during discrete phases of deposition. The coarse basal sand unit is relatively well-sorted and contains D₅₀ values between approximately 85 μm and approximately 265 μm, depending on the deposit (Figure 6). In comparison, the D₅₀ in background lake sediments is generally between 15 and 25 μm (Figure 6). Above the sand unit, D₅₀ and grain sorting decrease step-wise into the middle silt unit and again into the upper clay unit, with the finest grain sizes (D₅₀ values of 2.5–15 μm) and poorest sorting comprising the clay-rich cap layer (Figure 6).

Principal Component Analysis (PCA) can be performed in stratigraphic assessments to characterize and statistically differentiate units (e.g., Praet et al., 2020). Here, the PCA results show broadly consistent associations between elemental abundance, grain size, and other sediment parameters across all events (Figure 7). The primary principal component (PC1) captures between 53.95% and 77.19% of multivariate variability and is predominantly controlled by sediment grain size with positive and negative PC1 values corresponding to fine and coarse grain sizes, respectively. The second principal component (PC2) captures between 10.98% and 26.95% of multivariate variability and appears to be influenced by secondary elemental relationships. A common feature of the PCA is the clustering of sand content, D₅₀, and density in negative PC1 space while clay content and silt content generally plot together near positive PC1 (Figure 7). The elemental data relationships are less consistent. Elemental abundance generally aligns with positive PC1, suggesting a greater abundance of these elements in clay- and silt-sized particles, which tend to cluster closest to positive PC1 and near magnetic susceptibility. Geometric standard deviation (GSD) and skewness plot consistently between positive PC1 and negative PC2 (Figure 7).

The PCA results also highlight a consistent relationship between stratigraphic depth within the event deposits and variations in the sedimentary parameters captured by PC1 and PC2. The basal (coarse-grained) and upper (fine-grained) portions of the events are consistently dominated by PC1 while mid-depths show variable associations with PC2 (Figure 7). This relationship indicates that sedimentary characteristics of the basal unit and top unit are best correlated with PC1, with a minimal but variable contribution of positive and negative PC2, while those at intermediate depths show decreasing association PC1 and an increasing association with PC2 (Figure 7).

The geomorphologic and tectonic setting of the range-front Teton lakes make them particularly well-suited to record seismic activity. Moreover, the lake records are continuous and can be established with high chronological resolution, providing an ideal complement to the nearby paleoseismic trench exposures. Steep subaqueous slopes and relatively high rates of sediment delivery to the lakes results in the replenishment of lacustrine and deltaic sediment supply available for successive earthquake-triggered failures and provides ample stratigraphic spacing of individual turbidite deposits in core section—thus aiding in their identification and ability to be accurately dated.

We suggest that the characterization of earthquake-generated turbidites at Jenny Lake provides a model for their formation in this setting and establishes a framework for identifying, characterizing, and correlating earthquake-generated deposits in nearby basins. Similar piedmont lakes (Phelps, Taggart, Bradley, and Leigh) span a broad section of the fault and together, present a rare opportunity to develop multiple cross-evaluated, temporally-long (14 ka), and detailed earthquake records (Figure 1). In the following discussion, we present the Jenny Lake sediment record in the context of improving the postglacial paleoseismic history of the Teton fault with an emphasis on coseismic sedimentary processes in and around the piedmont lakes.

The generalized stratigraphy of Jenny Lake turbidites is consistent and begins with a sharp basal contact with underlying sediments followed by a well-defined sedimentary sequence (Figure 6). Based on patterns in grain size distributions and in accordance with descriptions of coseismic turbidite stratigraphy observed in similar lake settings (e.g., Beck, 2009), we categorize the turbidites into three primary components—a basal coarse sand unit, a middle silt unit, and an upper clay-rich cap. These sub-divisions are clearly differentiated by changes in grain sizes, which suggest a step-wise decrease in the energy of the depositional environment during the overall phase of deposition (e.g., Bertrand et al., 2008) (Figure 6). Distinctive relationships between textural and geochemical properties of each component are highlighted for the 10.3 ka turbidite event (Figure 8). The sand unit (“Unit I”) is relatively thick, has a sharp erosive basal contact, maximum grain size values, and high density (Figure 8). The properties of this unit are indicative of deposition by a relatively high-velocity current. For most events, it is well-sorted and reveals either mostly massive grading or a slight fining-upward pattern. One exception to this trend is the sand unit in the 9.1 ka event, which contains a strong fining-upward pattern (Figure 6).

We suggest the coarse sediments associated with the sand unit are sourced primarily from the steep subaqueous slopes around the lake perimeter, and particularly from the inflow delta. Steep slopes are an important characteristic of Teton lakes with regard to sediment transport mechanisms. Slope failures at Jenny Lake are revealed by bathymetric imagery, which shows numerous sediment piles at the base of the moraine along the lake’s perimeter and landslide deposits (Figure 2; Larsen et al., 2019). On their western shores, the Teton fault runs along hillslopes at the mouths of the respective glacial valleys and directly across the lake inflow deltas (Figure 1). The Gilbert-style deltas are composed of relatively coarse sediments with a steeply-dipping foreslope that descends to the central deep basin of each lake. Seismic imagery and core stratigraphy at Jenny Lake confirm that sediment transport to the lakes occurs primarily through the inflow delta system. We suggest that high slope angles and the continuous supply of sediment sourced from the catchment render the submerged delta fronts unstable and vulnerable to failure, including in response to strong ground shaking during earthquakes. In such settings, delta slope failures that result in a mass flow can result in diagnostic turbidite deposits in the basin (Beck, 2009; Sabatier et al., 2022). Due to the hypsometry of the Teton lake basins and their position along the range front, delta failure and/or destabilization of other subaqueous slopes would generate a high-density flow and the immediate delivery and focusing of coarse sediments to the lake’s central basin.

The next stratigraphic component is the silt unit (“Unit II”) (Figure 8). It is relatively thick, composed of fine silts to very fine sands, and reveals either massive grading or a modest fining-upward pattern, depending on the deposit (Figure 6). The boundary between the underlying sand unit and this unit is relatively well-marked by changes in grain size and other parameters, including a slight decrease in grain sorting and density and increased K, Ti, and Fe content (Figure 8). The finer texture of this unit reflects a reduction in the energy of the depositional environment. In some settings, co-seismic turbidites contain a layer of reworked terrestrial organic material interstitial to the sand and/or silt units (e.g., Bertrand et al., 2008). At Jenny Lake, the only turbidite that contains appreciable organic material is the 8.4 ka event (Figure 6). As discussed below, this particular turbidite is associated with a contemporaneous lake-terminating landslide event, which we suggest may have contributed terrestrial plant material to the deposit.

The clay-rich cap is the top unit (“Unit III”) of the turbidite sequence and expressed as a relatively thin layer of clay to fine silts (Figure 6). The clay cap is characterized by relatively poor sorting and modestly increased density and Fe and K content (Figure 5). We suggest the notable decrease in grain sorting of this unit may reflect the contribution of coarser grains due to the resumption of background sedimentation. The very fine-grained sediment in this layer is likely sourced from previously deposited lake sediments, including the glacial lithostratigraphic unit. We suggest previously-deposited clays and fine silts are mobilized during strong ground shaking, resuspended in the water column, and redistributed throughout the lake due to turbidity currents and seiching, with the finest particles settling last (i.e., during the ice-covered season when the absence of wave energy enables the settling of fine grains). A seiche wave -triggered by coseismic subaqueous slope failures, subaqueous fault displacement, and/or landsliding into the lake-would act to remobilize additional clastic material from near-shore regions and/or trigger backslope failures of the impounding moraine. Seiching may have also contributed to the observed focusing of the deposit in the central basin and increased segregation between fine- and coarser-grained particles, which resulted in the distinct 3-component stratigraphy (e.g., Bertrand et al., 2008; Beck, 2009).

Given their potential to exhibit similar features, differentiating between flood and earthquake-generated turbidites remains an important objective of investigations into event deposits in lake sediments (e.g., Praet et al., 2020; Vandekerkhove et al., 2020; Sabatier et al., 2022). Some studies benefit from the identification and characterization of individual turbidites that were generated by documented historical events (St-Onge et al., 2004; Beck, 2009; Daxer et al., 2022; Fan et al., 2023). However, without an historical flood or earthquake event to use as a sedimentary fingerprint, differentiation between these two triggers is primarily based on recognition of distinctive sedimentological characteristics and basin-scale distribution patterns associated with inferred sedimentary processes (e.g., Beck, 2009; Praet et al., 2020; Vandekerkhove et al., 2020). For example, differences between the quasi-stable versus quasi-instantaneous flow and depositional style associated with flood and earthquake-derived deposits, respectively, can help in attributing a trigger mechanism. Strong ground shaking during an earthquake is expected to induce subaqueous slope failure and a short period of high hyperpycnal flow that gradually wanes (Mulder and Alexander, 2001). Previous studies demonstrate that shaking-induced delta collapse and/or failures of other subaqueous slopes result in a sharp basal contact and a relatively coarse sand unit (e.g., Bertrand et al., 2008). In contrast, the hyperpycnal flow induced by floods typically lasts longer and is somewhat steadier in flow velocity (Alexander and Mulder, 2002; Mulder et al., 2003). Additionally, flood water is more likely to “bypass” the coarse sediments of the delta front, which results in a comparatively fine-grained deposit that commonly lacks an erosive basal contact (Mulder and Alexander, 2001; Beck, 2009). Some studies of flood deposits in lake sediments observe a distinctive two-unit structure consisting of a silt and clay layer (Mulder and Chapron, 2011). While earthquake deposits are expected to show variable distribution and thickness within the basin (Vandekerkhove et al., 2020), the hyperpycnal flow during a flood event carries sediment across the basin, resulting in a more-even basin-wide spatial distribution and thickness from mixing and settling evenly throughout the lake (Praet et al., 2020; Vandekerkhove et al., 2020).

The stratigraphical and statistical (PCA) similarity of turbidite deposits at Jenny Lake suggests a common trigger mechanism (Figures 6, 7). We infer the trigger to be earthquake activity along the Teton Fault, which is evidenced by the distinct three-component stratigraphy and strong basal contact of the turbidites, their spatial distribution and variable thickness, and their temporal agreement with trench investigations. We would expect flood-generated deposits in Teton lakes to be comparatively fine-grained, lack an erosive basal contact or strong three-component stratigraphy, and display a more-uniform distribution throughout the basin. However, we acknowledge the degree of similarity between flood and earthquake-generated deposits are highly contingent on the local geomorphic system (e.g., Sabatier, 2022) and that a universal expression across all Teton Lakes cannot be assumed. We therefore suggest careful correlation and analysis of a multi-lake record may aid in improving our understanding of the nuanced relationship between basin morphology and the stratigraphic expression of co-seismic turbidites.

Hillslopes in and around the Teton Range are susceptible to landslides (e.g., Love et al., 1992; Crosby, 2019; Johnson et al., 2022). Here, we refer to landslides broadly as incorporating a range of mass movement events that describe flows, slides, and falls of earth materials. Evidence of past landslides can be observed on the steep canyon walls in the interior of the mountain range and along the eastern range front, where in places, they transect the Teton fault scarp and/or run out into lake basins (Crosby, 2019; Zellman et al., 2019). For example, the floor and walls of Cascade Canyon are blanketed by colluvium, talus, and alluvial fan deposits that have accumulated since deglaciation approximately 15 ka. Landslides occurring over this timescale have affected large portions of the Canyon and are considered to be important agents of clastic sediment production and transport (e.g., Tranel et al., 2015; Johnson et al., 2022).

Above the western shore of Jenny Lake, prominent landslide scars are visible in the hillslopes adjacent to the fault trace (Figure 2). One landslide south of Cascade Canyon originated on a steep bedrock slope in Late Archean crystalline rocks, and another originated in poorly consolidated moraine deposits and colluvium to the north (Love et al., 1992; Figure 2). The runout paths of both features can be mapped into the lake’s central basin where they terminate as block and fan deposits visible in bathymetric and seismic imagery. High-amplitude seismic reflectors extending laterally from these deposits indicate the two events were coincident with and/or contributed to widespread deposition of a layer of anomalously dense sediments around the lake.

Larsen et al. (2019) dated the two landslides above Jenny Lake using ¹⁰Be inventories of boulder and bedrock surfaces, which yielded mean ages of approximately 8.1 ka and approximately 14.0 ka for the two landslides, respectively (Figure 2). The landslides ages are coincident with two radiocarbon-dated turbidites in the Jenny Lake sediment sequence and preliminary ages of two turbidites identified in nearby lakes Bradley and Taggart (Blumm et al., 2019; Figure 1). Both of these turbidites are remarkably thick at the Central Basin core site (24-cm-thick and 34-cm-thick, respectively) but otherwise present a similar stratigraphic expression to other turbidites at Jenny Lake, suggesting a similar process of formation. The earlier event is also fully expressed in core JEN15-4 recovered from the distal eastern portion of the lake, on a shelf approximately 23 m above the central basin (Figures 2, 3). At this site, the coarse basal unit of the 13.9 ka turbidite contains rounded clasts of gneiss and quartzite cobbles, up to 8 cm in diameter, that were likely sourced from the backslope of the terminal moraine due to seiching (Larsen et al., 2019). There is no lithologic source of quartzite cobbles in Cascade Canyon but they are a common component of the terminal moraine complex, which was developed in part by the buildup of quartzite-rich outwash sourced from regions to the north of Jackson Hole during the Pinedale glaciation (Pierce and Good, 1992). The younger 8.4 ka turbidite deposit is contained in the Central Basin cores and at other core sites. In contrast to all other turbidites, this deposit is notable for an unusually high amount of interstitial woody plant remains that are clearly visible in core section and register as increased OM and decreased density values (Figures 3, 4). We suggest the woody plant material incorporated in this deposit was delivered to the lake directly by the landslide and by seiching, which could have remobilized OM from nearshore regions and focused it in the central basin. We further speculate that the lack of organic matter contained in the older 13.9 ka turbidite is a function of the age of this deposit, which occurred soon after ice retreat from the lake basin when the hillslopes around the lake, and particularly the freshly deposited moraines, may have supported only minimal woody vegetation.

The available evidence strongly suggests that the two lake-terminating landslides at Jenny Lake occurred coincident with the timing of earthquake events and contributed clastic material and organic remains to the resulting turbidite deposits. We note that past landslides throughout the Tetons were most likely driven by both seismic and aseismic factors (e.g., glacial over-steepening, extreme precipitation events, wildfires, etc.). However, we demonstrate that in some cases, deep-seated landslides in the Tetons may have been triggered by major earthquake events. The potential for strong earthquakes to trigger slope failures presents significant hazards to visitors of Grand Teton National Park and to local infrastructure, yet an incomplete record of both postglacial seismicity and landsliding impedes our ability to characterize the relationship and magnitude of these coupled hazards. We suggest future work combining paleoseismic histories with efforts to construct landslide inventories and chronologies could help elucidate patterns of past seismically-induced landslide activity in the Tetons.

The relationship between local seismic intensity and turbidite properties (thickness, grain size) has been explored in various lake settings around the world (e.g., Beattie and Dade, 1996; Van Daele et al., 2015; Daxer et al., 2022) and some studies have quantitatively estimated paleomagnitude using such relationships (e.g., Hibsch et al., 1997; Rodriguez-Pascua et al., 2003; Moernaut et al., 2014; Van Daele et al., 2015; Daxer et al., 2022). It is intuitive that stronger ground shaking can mobilize greater sediment volumes and result in thicker turbidites (e.g., Goldfinger et al., 2012; Moernaut et al., 2017), and that ruptures resulting in delta failure can yield especially coarse deposits in proximal basins (e.g., Beck, 2009). However, despite these first-order controls, the nuanced relationship between turbidite character and paleoearthquake magnitude are likely to be site-specific and influenced by numerous additional factors. For example, coseismic turbidite properties (thickness, grain size distributions, sharpness of basal contact) are largely controlled by the interaction of basin morphology with the many tectonic, climatic, and geomorphic processes that influence local sediment supply across a range of timescales (e.g., production, transport, and deposition of sediment in and around the lake) (e.g., Bouma, 2004; Howarth et al., 2012; Wilhelm et al., 2016; Praet et al., 2020). Increased sediment supply maximizes the amount of sediment available to be mobilized during slope failure and therefore maximizes potential turbidite thickness. Basin morphometry can influence turbidite thickness as the size, shape, and proximity to the sediment source influence the volume of sediment delivered to any particular location in the basin. Similarly, changes in climate can influence turbidite properties by governing the production, transport, and character of sediment through the catchment over a variety of timescales. Sediment supply is also influenced by other factors such as the time elapsed since the last earthquake event and/or the recurrence intervals of other disturbance events that control sediment flux (e.g., glaciations, floods, wildfires, landslides, etc.).

These factors discussed above, and many others highlight the importance of varying sediment supply when relating turbidite properties at a given core site to seismic intensity. As stated earlier, we suggest the Teton lake inflow deltas are a key source of sediment remobilized during earthquake events. We further suggest that delta failure and other slope failures are not constrained by sediment influx. This is demonstrated by the relatively high accumulation rates and potentially by the repeated sand units in the upper sequence of JEN16-2 (Figure 2). The chaotic sand units occur during a prolonged interval of inferred seismic quiescence and are only present proximal to the delta front (Figures 2, 3). We speculate that these chaotic deposits reflect episodes of minor delta failure at Jenny Lake during times when the inflow delta is not periodically reset by seismic shaking and thus susceptible to slope instabilities triggered by aseismic factors (e.g., sediment overload, storms, etc.). Therefore, assuming delta failure is not limited by sediment influx, we deduce a first-order relationship between turbidite properties and the degree of ground shaking. This potential relationship may be strengthened by comparison to other proximal lake basins. A similar approach was followed byMoernaut et al. (2014), who used the composite turbidite thickness across multiple lake basins to estimate the paleomagnitude of megathrust events in Chile. Additionally, a comparison of the thickness of synchronously deposited events across lake basins may aid in determining local seismic intensity, epicenter location and/or rupture length. Thus, while we acknowledge that turbidite thickness at any single location within a lake may be partly influenced by earthquake magnitude, we suggest that the stratigraphic expression and presence of synchronous turbidite deposits in multiple lake basins positioned along fault strike may be a more meaningful indicator. Future work that develops and correlates additional paleoseismic records across multiple range-front Teton lakes and trench datasets may provide a framework for understanding and interpreting the intricate relationship between turbidite properties and seismic intensity.