Geology topics

William J. Stephenson

Publications and source records attributed to William J. Stephenson.

At least 37 records · Page 2Linked to original sources

USGS2021

Shear-wave velocity site characterization in Oklahoma from joint inversion of multi-method surface seismic measurements: Implications for central U.S. Ground Motion Prediction

We analyze multimethod shear (SH)‐wave velocity ( ⁠ V S "> V S ) site characterization data acquired at three permanent and 25 temporary seismograph stations in Oklahoma that recorded M 4+ earthquakes within a 50 km hypocentral distance of at least one of the 2016 M 5.1 Fairview, M 5.8 Pawnee, or M 5.0 Cushing earthquakes to better constrain earthquake ground‐motion modeling in the region. We acquired active‐source seismic data for time‐averaged V S "> V S to 30 m depth ( V S "> V S V S 30 "> 30 ) at 28 seismograph stations near the Fairview, Pawnee, and Cushing epicentral areas. The SH‐wave refraction travel times coupled with Rayleigh‐ and Love‐wave phase velocity dispersion were extracted and modeled in a nonlinear least‐squares (L2) joint inversion to obtain a best‐fit 1D V S "> V S versus depth profile for each site. At a subset of sites where the preferred L2 inverse model did not optimally fit each of the Love, Rayleigh, and SH travel‐time datasets, we explore application of simulated annealing in a joint inversion to find a more global solution. V S "> V S V S 30 "> 30 values range from 262 to 807    m / s "> 807 m / s for the preferred measured (in situ) V S "> V S profiles, or National Earthquake Hazards Reduction Program (NEHRP) site class D to B, and are broadly comparable with estimates from previous data reports in the region. Site amplification estimates were calculated next from 1D SH transfer functions of the preferred V S "> V S VS profiles and then compared against observed horizontal‐to‐vertical spectral ratios (HVSRs) from nearby seismograph stations. We generally see good agreement between the predicted in situ model and the observed HVSR resonant frequencies, with nominal amplifications between 2 and 10 within the 2–15 Hz frequency band. Next, using 40 known in situ V S "> V S V S 30 "> 30 measurements in the region, we demonstrate that the in situ V S "> V S V S 30 "> 30 values improve the fit for selected suites of ground‐motion models (GMMs) for M 4+ earthquakes within a 50 km hypocentral distance when compared with proxy methods, arguing for future development of GMMs implementing in situ V S "> V S profiles.

Oklahoma↗

USGS2020

Seismic reflection imaging of the low-angle Panamint normal fault system, eastern California

Shallowly dipping (<30°) low‐angle normal faults (LANFs) have been documented globally; however, examples of active LANFs in continental settings are limited. The western margin of the Panamint Range in eastern California is defined by a LANF that dips west beneath Panamint Valley and has evidence of Quaternary motion. In addition, high‐angle dextral‐oblique normal faults displace middle to late Quaternary alluvial fans near the range front. To image shallow (<1 km depth), crosscutting relationships between the low‐ and high‐angle faults along the range front, we acquired two high‐resolution P wave seismic reflection profiles. The northern, 4.6‐km‐long profile crosses the 2‐km‐wide Wildrose graben and the southern, 0.8‐km‐long profile extends onto the Panamint Valley playa, ~7.5 km S of Ballarat, CA. The profile across the Wildrose graben reveals a robust, low‐angle reflector interpreted to represent the LANF separating Plio‐Pleistocene alluvial fanglomerate and Proterozoic metasedimentary deposits. High‐angle faults interpreted in the seismic profile correspond to fault scarps on Quaternary alluvial fan surfaces. Interpretation of the reflection data suggests that the high‐angle faults vertically displace the LANF up to 80 m within the Wildrose graben. Similarly, the profile south of Ballarat reveals a low‐angle reflector, which appears both rotated and displaced up to 260 m by high‐angle faults. These results suggest that near the Panamint range front, the high‐angle faults are the dominant active structures. We conclude that at least at shallow (<1 km) depths, the LANF we imaged is not active today.

California↗

USGS2020

Graphical Dispersion Plot Editor (DPE) for seismic-site characterization by using multiple surface-wave methods

Introduction To understand the behavior of potentially damaging ground motions during earthquakes, seismic-site effects are routinely characterized by using the dispersion of surface waves. Many methods exist to measure dispersion; these methods have various advantages and disadvantages, but they all yield dispersion data that must be inverted for shear-wave velocity. This report presents a graphical tool for efficiently removing spurious data as well as combining data from multiple methods prior to inversion. The Dispersion Plot Editor (DPE) program presented here (version 1.5) is coded in Python 3, which is open source and platform independent. DPE accepts input dispersion data as one or more delimited text files. The program plots the data in useful forms, including both scattered points and an interpolated heat map. The user selects points to delete by drawing arbitrary shapes with the mouse cursor. After the spurious data are removed, the user may represent the acceptable data with a dispersion curve. The acceptable data and the representative dispersion curve are output as separate comma-delimited text files. Images of the plotted data and the representative dispersion curve may also be saved.

Open-File Report↗

USGS2020

Evidence for late Quaternary deformation along Crowley's Ridge, New Madrid seismic zone

The New Madrid seismic zone has been the source of multiple major (M ~7.0–7.5) earthquakes in the past 2 ka, yet the surface expression of recent deformation remains ambiguous. Crowleys Ridge, a linear ridge trending north‐south for 300+ km through the Mississippi Embayment, has been interpreted as either a fault‐bounded uplift or a nontectonic erosional remnant. New and previously published seismic reflection and shallow resistivity data show discontinuities at the ridge margins in Plio‐Pleistocene strata, yet the timing of most recent faulting and the lateral extent of these faults remain unknown. To assess Pleistocene‐to‐recent tectonic activity of Crowleys Ridge, we perform landscape‐scale geomorphic analyses, such as relief, slope, hypsometry, and drainage basin shape, on a 10‐m digital elevation model (DEM). North‐to‐south variations in geomorphic indices indicate Pleistocene‐to‐recent tectonic uplift of the southern ridge. Moreover, mapping on a <1‐m lidar‐derived DEM reveals scarps on late Pleistocene geomorphic surfaces. The scarps are primarily located along the southern ridge, trend parallel to the ridge margin discontinuously for 0.1–1 km, and vertically offset <56 ka surfaces 0.4 m with up to 6 m of tilting. These landscape‐scale patterns and scarps, integrated with discontinuities in the seismic reflection and resistivity data, provide evidence of low‐rate (<0.2 mm/year) late Quaternary tectonic activity along the southern segment of Crowleys Ridge. The interpretations agree with recent tectonic models suggesting southern Crowleys Ridge is a compressional step over in a right‐lateral fault system within the Reelfoot Rift.

Arkansas, Illinois, Kentucky, Missouri, Tennessee↗

USGS2018

Erratum to Broadband synthetic seismograms for magnitude 9 earthquakes on the Cascadia megathrust based on 3D simulations and stochastic synthetics, Part 1: methodology and overall results

No abstract available.

Bulletin of the Seismological Society of America↗

USGS2018

Paleoseismic results from the Alpine site, Wasatch fault zone: Timing and displacement data for six holocene earthquakes at the Salt Lake City–Provo segment boundary

To improve the characterization of Holocene earthquakes on the Wasatch fault zone (WFZ), we conducted light detection and ranging (lidar)‐based neotectonic mapping and excavated a paleoseismic trench across an 8‐m‐high fault scarp near Alpine, Utah, located < 1    km "> < 1 km <1 km south of the boundary between the Salt Lake City and Provo segments (SLCS and PS). We document evidence for six paleoearthquakes (AL6–AL1) from scarp‐derived colluvial wedges and crosscutting relations. A ground‐penetrating radar survey across the scarp resolved fault‐zone width, but not paleoearthquake stratigraphy. Bayesian (OxCal) modeling of 13 radiocarbon and 13 optically stimulated luminescence ages indicates that six earthquakes occurred ∼ 6.2 – 0.4    ka "> ∼ 6.2 – 0.4 ka ∼6.2–0.4 ka . Interseismic recurrence ranges from 0.2 to 1.8 ky (mean 1.2 ky). We estimate 6.5 ± 0.7    m "> 6.5 ± 0.7 m 6.5±0.7 m of cumulative vertical tectonic displacement across the ≥ 14 ‐ m ‐ wide "> ≥ 14 ‐ m ‐ wide ≥14‐m‐wide fault zone used near‐field observations of scarp‐derived colluvial‐wedge thicknesses, antithetic faulting, and graben backtilting. This is similar to our independent estimate of 6.5 ± 0.5    m "> 6.5 ± 0.5 m 6.5±0.5 m using far‐field observations of the offset ground surface and correlation of alluvial‐fan stratigraphy across the WFZ. These results suggest that colluvial‐wedge thickness at the Alpine site approximates one‐half the original west‐facing fault scarp height. Per‐event vertical displacements range from 0.8 to 1.2 m (mean 1.1 m), which we use to estimate surface rupture lengths that may exceed 50 km from earthquakes as large as moment magnitude ( M w "> M w Mw ) ∼ 7.0 "> ∼ 7.0 ∼7.0 . The late Holocene average vertical slip rate is 0.9    mm / yr "> 0.9 mm / yr 0.9 mm/yr ( 0.7 – 1.2    mm / yr "> 0.7 – 1.2 mm / yr 0.7–1.2 mm/yr range). Earthquake frequency has increased in the past ∼ 1    ky "> ∼ 1 ky ∼1 ky , whereas displacement per event has been similar for the past ∼ 6    ky "> ∼ 6 ky ∼6 ky , suggesting that strain accumulation is not the sole factor that controls the frequency and size of earthquakes at the SLCS–PS segment boundary. These findings can be used for a more nuanced characterization of earthquakes at the SLCS–PS boundary and improve earthquake hazard assessments along the Wasatch Front.

Utah↗

USGS2018

Holocene fault reactivation in the eastern Cascades, Washington

Significant uncertainty remains concerning how and where crustal shortening occurs throughout the eastern Cascade Range in Washington State. Using light detection and ranging (lidar) imagery, we identified an ∼ 5 ‐ km ‐ long "> ∼5‐km‐long lineament in Swakane canyon near Wenatchee, roughly coincident with a strand of the Entiat fault. Topographic profiles across the lineament reveal a southwest‐side‐up break in slope, with an average of 2–3 m of vertical separation of the hillslope surface. We consider a range of possible origins for this feature, including differential erosion across a fault‐line scarp, slope failure (sackung or landslide), and surface deformation across an active fault strand. Based on trenching, radiocarbon and luminescence dating, and ground‐penetrating radar (GPR) across the lineament, we conclude that warped saprolite observed in the shallow subsurface is most consistent with southwest‐side‐up folding caused by blind reverse faulting at depth. Following this reasoning, dating of overlying colluvial deposits suggests that at least one Holocene earthquake occurred on this strand of the southern Entiat fault, with an approximate vertical separation of ≥ 1    m "> m ≥1 m . GPR reveals up to 4 m of cumulative vertical separation of the saprolite, suggesting a history of multiple earthquakes on the structure. Taken in context with other potential fault‐related lineaments along the Entiat fault, our interpretation of Holocene earthquakes in Swakane canyon could suggest reactivation of longer sections of the Entiat fault, as well as of other bedrock faults in the eastern Cascades. Although active erosion and slow strain rates lead to a subdued geomorphic expression of recent deformation, we conclude that the reactivated Entiat fault represents a seismogenic structure that should be considered in regional seismic hazard analyses. The difficulty of recognizing low‐slip‐rate structures in forested and mountainous terrain underscores the importance of additional lidar surveys and geological and geophysical studies for fully understanding seismic hazard in regions with infrequent but potentially large earthquakes.

Washington↗

USGS2018

Broadband synthetic seismograms for magnitude 9 earthquakes on the Cascadia Megathrust based on 3D simulations and stochastic synthetics, Part 2: Rupture parameters and variability

We used a combination of 3D finite‐difference simulations ( < 1    Hz "> < 1 Hz ) and 1D stochastic synthetics ( > 1    Hz "> > 1 Hz ) to generate broadband (0–10 Hz) synthetic seismograms for numerous M w "> Mw 9 earthquake rupture scenarios on the Cascadia megathrust. Slip consists of multiple high‐stress‐drop subevents ( M w "> Mw 8) with short rise times on the deeper portions of the fault, superimposed on a background slip distribution with longer rise times. We identify key rupture parameters that control the intensity of ground shaking and resulting seismic hazard; these include the hypocenter location, the down‐dip limit of slip, the average rupture velocity, and the character (i.e., location, magnitude, and stress drop) of subevents. Extending the down‐dip limit of rupture to the top of the nonvolcanic tremor zone results in localized regions with a factor of 5–10 increase in spectral acceleration (SA) for periods < 5    s "> <5 s , compared to a rupture that is completely offshore. This is primarily due to the closer proximity of high‐stress‐drop subevents to inland locations when the rupture is allowed to extend deeper. Similarly, we find that the hypocenter location can result in a variation in the intensity of ground motions of a factor of > 10 "> >10 , due to the effects of rupture directivity (i.e., SA at periods > 1    s "> >1 s ). We also observe a coupling between rupture directivity and basin amplification. The intensity of ground motions is also strongly affected by the magnitude, stress drop, and location of high‐stress‐drop subevents, which are poorly constrained. Overall, our results quantify the effect of kinematic rupture parameters on ground motions for an M w "> Mw 9 earthquake in Cascadia and emphasize the need for further constraints on these parameters to improve seismic hazard estimates in the Pacific Northwest.

Bulletin of the Seismological Society of America↗

USGS2018

Broadband synthetic seismograms for magnitude 9 earthquakes on the Cascadia megathrust based on 3D simulations and stochastic synthetics, part 1: Methodology and overall results

We have produced a large set of broadband (0–10 Hz) synthetic seismograms for Mw 9.0 earthquakes on the Cascadia megathrust by combining synthetic seismograms derived from 3D finite‐difference simulations ( ⁠≤1 Hz⁠ ) with finite‐source, stochastic synthetics ( ⁠≥1 Hz⁠ ). We used a compound rupture model consisting of high stress drop M w 8 subevents superimposed on large, shallower slip with long‐slip duration, informed by observations of the M w 9.0 Tohoku, Japan, and M w 8.8 Maule, Chile, earthquakes. Thirty 3D simulations were run, considering a variety of rupture parameters, to determine the range of expected ground motions. For sites not in sedimentary basins, the spectral accelerations of the synthetics are similar to the BC Hydro ground‐motion prediction equations (GMPEs) for periods of 0.1–6 s, but exceed them at periods greater than 6 s. Response spectra from the synthetics at sites in the Seattle and Tacoma sedimentary basins show large amplifications of factors of 2–5 at periods of 1–10 s. This basin amplification is substantially larger than that found for crustal earthquakes in the Next Generation Attenuation‐West2 (NGA‐West2) GMPEs. Basin amplification is caused by basin‐edge generated surface waves and by amplification and focusing of S waves and surface waves by the 3D basin structure. The synthetic seismograms show effective average durations of strong motions of about 70 s for coastal sites, increasing to about 120 s at 200 km distance. We find that the interevent and intraevent standard deviations of the spectral amplitudes of the synthetics are larger for sites closer to the rupture, because they are more sensitive to the location of subevents and rupture directivity.

California, Oregon, Washington↗

USGS2018

Integrate urban‐scale seismic hazard analyses with the U.S. National Seismic Hazard Model

For more than 20 yrs, damage patterns and instrumental recordings have highlighted the influence of the local 3D geologic structure on earthquake ground motions (e.g., M "> M M 6.7 Northridge, California, Gao et al. , 1996 ; M "> M M 6.9 Kobe, Japan, Kawase, 1996 ; M "> M M 6.8 Nisqually, Washington, Frankel, Carver, and Williams, 2002 ). Although this and other local‐scale features are critical to improving seismic hazard forecasts, historically they have not been explicitly incorporated into the U.S. National Seismic Hazard Model (NSHM, national model and maps), primarily because the necessary basin maps and methodologies were not available at the national scale. Instead, the U.S. Geological Survey (USGS), its partners, and external groups developed urban seismic hazard maps (urban models and maps) that consider detailed site effects in local areas (e.g., Wong et al. , 2002 ; Cramer et al. , 2006 ; Frankel et al. , 2007 ; Graves et al. , 2011 ). The disconnect between the urban and national hazard models, however, means that the national models, which underlie U.S. building codes and other applications, do not make use of all of the scientific results informing earthquake ground‐shaking hazards. We recommend that future U.S. national seismic hazard assessment make use of all available regional information, including that in urban models. In this column, we describe the roles of and differences between the urban and national models, and discuss the obstacles to and benefits of integrating the urban models with the national model. Future progress on this issue will require further coordination and implementation efforts between the USGS and external partners.

Seismological Research Letters↗

USGS2018

3-D Simulations of M9 earthquakes on the Cascadia Megathrust: Methodology and results

No abstract available.

Conference Paper↗

USGS2017

P- and S-wave velocity models incorporating the Cascadia subduction zone for 3D earthquake ground motion simulations, Version 1.6—Update for Open-File Report 2007–1348

In support of earthquake hazard studies and ground motion simulations in the Pacific Northwest, three-dimensional P- and S-wave velocity ( V P and V S , respectively) models incorporating the Cascadia subduction zone were previously developed for the region encompassed from about 40.2°N. to 50°N. latitude, and from about 122°W. to 129°W. longitude. This report describes updates to the Cascadia velocity property volumes of model version 1.3 (V1.3), herein called model version 1.6 (V1.6). As in model V1.3, the updated V1.6 model volume includes depths from 0 kilometers (mean sea level) to 60 kilometers, and it is intended to be a reference for researchers who have used, or are planning to use, this model in their Earth science investigations. To this end, it is intended that the V P and V S property volumes of model V1.6 will be considered a template for a community velocity model of the Cascadia region as additional results become available. With the recent and ongoing development of the National Crustal Model, we envision any future versions of this model will be directly integrated with that effort.

Oregon, Vancouver, Washington↗

USGS2017

The story of a Yakima fold and how it informs Late Neogene and Quaternary backarc deformation in the Cascadia subduction zone, Manastash anticline, Washington, USA

The Yakima folds of central Washington, USA, are prominent anticlines that are the primary tectonic features of the backarc of the northern Cascadia subduction zone. What accounts for their topographic expression and how much strain do they accommodate and over what time period? We investigate Manastash anticline, a north vergent fault propagation fold typical of structures in the fold province. From retrodeformation of line- and area-balanced cross sections, the crust has horizontally shortened by 11% (0.8–0.9 km). The fold, and by inference all other folds in the fold province, formed no earlier than 15.6 Ma as they developed on a landscape that was reset to negligible relief following voluminous outpouring of Grande Ronde Basalt. Deformation is accommodated on two fault sets including west-northwest striking frontal thrust faults and shorter north to northeast striking faults. The frontal thrust fault system is active with late Quaternary scarps at the base of the range front. The fault-cored Manastash anticline terminates to the east at the Naneum anticline and fault; activity on the north trending Naneum structures predates emplacement of the Grande Ronde Basalt. The west trending Yakima folds and west striking thrust faults, the shorter north to northeast striking faults, and the Naneum fault together constitute the tectonic structures that accommodate deformation in the low strain rate environment in the backarc of the Cascadia Subduction Zone.

Washington↗

USGS2017

3D ground‐motion simulations of Mw 7 earthquakes on the Salt Lake City segment of the Wasatch fault zone: Variability of long‐period (T≥1 s) ground motions and sensitivity to kinematic rupture parameters

We examine the variability of long‐period ( T ≥1 s) earthquake ground motions from 3D simulations of M w 7 earthquakes on the Salt Lake City segment of the Wasatch fault zone, Utah, from a set of 96 rupture models with varying slip distributions, rupture speeds, slip velocities, and hypocenter locations. Earthquake ruptures were prescribed on a 3D fault representation that satisfies geologic constraints and maintained distinct strands for the Warm Springs and for the East Bench and Cottonwood faults. Response spectral accelerations (SA; 1.5–10 s; 5% damping) were measured, and average distance scaling was well fit by a simple functional form that depends on the near‐source intensity level SA 0 ( T ) and a corner distance R c :SA( R , T )=SA 0 ( T )(1+( R / R c )) −1 . Period‐dependent hanging‐wall effects manifested and increased the ground motions by factors of about 2–3, though the effects appeared partially attributable to differences in shallow site response for sites on the hanging wall and footwall of the fault. Comparisons with modern ground‐motion prediction equations (GMPEs) found that the simulated ground motions were generally consistent, except within deep sedimentary basins, where simulated ground motions were greatly underpredicted. Ground‐motion variability exhibited strong lateral variations and, at some sites, exceeded the ground‐motion variability indicated by GMPEs. The effects on the ground motions of changing the values of the five kinematic rupture parameters can largely be explained by three predominant factors: distance to high‐slip subevents, dynamic stress drop, and changes in the contributions from directivity. These results emphasize the need for further characterization of the underlying distributions and covariances of the kinematic rupture parameters used in 3D ground‐motion simulations employed in probabilistic seismic‐hazard analyses.

Utah↗

USGS2017

The Evergreen basin and the role of the Silver Creek fault in the San Andreas fault system, San Francisco Bay region, California

The Evergreen basin is a 40-km-long, 8-km-wide Cenozoic sedimentary basin that lies mostly concealed beneath the northeastern margin of the Santa Clara Valley near the south end of San Francisco Bay (California, USA). The basin is bounded on the northeast by the strike-slip Hayward fault and an approximately parallel subsurface fault that is structurally overlain by a set of west-verging reverse-oblique faults which form the present-day southeastward extension of the Hayward fault. It is bounded on the southwest by the Silver Creek fault, a largely dormant or abandoned fault that splays from the active southern Calaveras fault. We propose that the Evergreen basin formed as a strike-slip pull-apart basin in the right step from the Silver Creek fault to the Hayward fault during a time when the Silver Creek fault served as a segment of the main route by which slip was transferred from the central California San Andreas fault to the Hayward and other East Bay faults. The dimensions and shape of the Evergreen basin, together with palinspastic reconstructions of geologic and geophysical features surrounding it, suggest that during its lifetime, the Silver Creek fault transferred a significant portion of the ∼100 km of total offset accommodated by the Hayward fault, and of the 175 km of total San Andreas system offset thought to have been accommodated by the entire East Bay fault system. As shown previously, at ca. 1.5–2.5 Ma the Hayward-Calaveras connection changed from a right-step, releasing regime to a left-step, restraining regime, with the consequent effective abandonment of the Silver Creek fault. This reorganization was, perhaps, preceded by development of the previously proposed basin-bisecting Mount Misery fault, a fault that directly linked the southern end of the Hayward fault with the southern Calaveras fault during extinction of pull-apart activity. Historic seismicity indicates that slip below a depth of 5 km is mostly transferred from the Calaveras fault to the Hayward fault across the Mission seismic trend northeast of the Evergreen basin, whereas slip above a depth of 5 km is transferred through a complex zone of oblique-reverse faults along and over the northeast basin margin. However, a prominent groundwater flow barrier and related land-subsidence discontinuity coincident with the concealed Silver Creek fault, a discontinuity in the pattern of seismicity on the Calaveras fault at the Silver Creek fault intersection, and a structural sag indicative of a negative flower structure in Quaternary sediments along the southwest basin margin indicate that the Silver Creek fault has had minor ongoing slip over the past few hundred thousand years. Two earthquakes with ∼M6 occurred in A.D. 1903 in the vicinity of the Silver Creek fault, but the available information is not sufficient to reliably identify them as Silver Creek fault events.

California↗

USGS2017

3-D simulations of M9 earthquakes on the Cascadia Megathrust: Key parameters and uncertainty

Geologic and historical records indicate that the Cascadia subduction zone is capable of generating large, megathrust earthquakes up to magnitude 9. The last great Cascadia earthquake occurred in 1700, and thus there is no direct measure on the intensity of ground shaking or specific rupture parameters from seismic recordings. We use 3-D numerical simulations to generate broadband (0-10 Hz) synthetic seismograms for 50 M9 rupture scenarios on the Cascadia megathrust. Slip consists of multiple high-stress drop subevents (~M8) with short rise times on the deeper portion of the fault, superimposed on a background slip distribution with longer rise times. We find a >4x variation in the intensity of ground shaking depending upon several key parameters, including the down-dip limit of rupture, the slip distribution and location of strong-motion-generating subevents, and the hypocenter location. We find that extending the down-dip limit of rupture to the top of the non-volcanic tremor zone results in a ~2-3x increase in peak ground acceleration for the inland city of Seattle, Washington, compared to a completely offshore rupture. However, our simulations show that allowing the rupture to extend to the up-dip limit of tremor (i.e., the deepest rupture extent in the National Seismic Hazard Maps), even when tapering the slip to zero at the down-dip edge, results in multiple areas of coseismic coastal uplift. This is inconsistent with coastal geologic evidence (e.g., buried soils, submerged forests), which suggests predominantly coastal subsidence for the 1700 earthquake and previous events. Defining the down-dip limit of rupture as the 1 cm/yr locking contour (i.e., mostly offshore) results in primarily coseismic subsidence at coastal sites. We also find that the presence of deep subevents can produce along-strike variations in subsidence and ground shaking along the coast. Our results demonstrate the wide range of possible ground motions from an M9 megathrust earthquake in Cascadia, and the potential to further constrain key rupture parameters using geologic and geophysical observations, ultimately improving our estimation of seismic hazard associated with the Cascadia megathrust.

Conference Paper↗

USGS2017

Geophysical characterization of seismic station sites in the United States – The importance of a flexible, multi-method approach

Noninvasive geophysical site characterization methods were used in two recent projects to obtain shear-wave velocity (VS) profiles to a minimum depth of 30 m and the time-averaged VS of the upper 30 meters (VS30) at seismic station sites. These projects include the 2009 American Recovery and Reinvestment Act (ARRA) funded U.S. Geological Survey site characterization project for 191 sites in California and the Central-eastern United States (CEUS), and the 2012 Electric Power Research Institute (EPRI) funded project for 33 additional CEUS sites. These sites are located in rural to urban settings with topographic conditions ranging from relatively flat sedimentary basins to mountaintop ridges. About 60 percent of the ARRA sites and 80 percent of the EPRI sites are located on rock or have thin sediment cover over rock, including Quaternary volcanic rock, Tertiary sediments and sedimentary rock, and Mesozoic (or older) crystalline or sedimentary rock. The remaining sites consist of thick sequences of Quaternary sediments overlying older sediments and rock. ARRA sites were characterized using non-invasive active and passive surface-wave methods, including the horizontal-tovertical spectral ratio (HVSR) method and one or more of the following: spectral analysis of surface waves (SASW), multichannel analysis of surface waves (MASW; Rayleigh and Love waves) and, occasionally, array microtremor (linear and 2-D arrays) methods. P-wave seismic refraction data were also acquired at rock and shallow-rock sites. S-wave seismic refraction and/or Love-wave MASW methods were applied at sites where characterization proved difficult with Rayleighwave methods. Based on our experience from the ARRA project, we acquired Rayleigh- and Love-wave based MASW and P- and S-wave refraction data for the EPRI project at CEUS sites. The HVSR method was found to be useful for identifying shallow-rock sites and for evaluating the relative variability of the depth-to-rock interface beneath the seismic station and the testing array(s). The fundamental mode modeling assumption was generally valid at most of these sites; nevertheless, multi-mode or effective-mode modeling routines were occasionally required, particularly in the case of shallow high-velocity layers. Deep sediment sites were characterized using active and, when appropriate, passive surface-wave based methods. Rock and shallow sediment sites were generally more challenging to characterize than deep sediment sites. About 10 percent of rock sites could not be characterized using surface wave methods, thus these sites were characterized using body-wave refraction methods. Love wave methods were found to be more effective than Rayleigh wave methods at some rock and shallow-rock sites (e.g., sites with shallow rock and sites with a thin low-velocity, highly attenuating surface layer). Lateral velocity variability was found to be very common at rock and shallow-rock sites, often causing significant scatter in the surface-wave dispersion data. Seismic refraction models have demonstrated that it may not be unusual for VS30 to vary by 20 percent, or more, over small distances (several tens of meters) at such sites. Based on these experiences, it is important to consider the application of combinations of methods when using noninvasive geophysical approaches to characterize seismic site conditions.

Conference Paper↗

USGS2016

Continuity of the Reelfoot fault across the Cottonwood Grove and Ridgely faults of the New Madrid Seismic Zone

Previous investigators have argued that the northwest-striking Reelfoot fault of northwest Tennessee and southeastern Missouri is segmented. One segment boundary is at the intersection of the northeast-striking Cottonwood Grove and Ridgely strike-slip faults with the Reelfoot fault. We use seismic reflection and geologic mapping to locate and determine the history of the Reelfoot South fault across this boundary zone. One reflection profile revealed a southwest-dipping (81°) Reelfoot South reverse fault that displaces the top of the Paleozoic 65 m, Cretaceous 40 m, Paleocene 31 m, Eocene Wilcox Group 20 m, and Eocene Memphis Sand 16 m. A second reflection profile reveals a north-dipping (84°) reverse fault 4.3 km south of the Reelfoot South fault, which defines the southwest margin of the Tiptonville dome. A geologic profile of the base of the ∼3:1 Ma Upland complex (Mississippi River terrace alluvium) within theMississippi River bluffs reveals ∼6 m of displacement across the Reelfoot South fault. Similarly, Quaternary stream terrace distribution suggests ∼6 m of Reelfoot South hanging-wall (Tiptonville dome) uplift that is probably Holocene. Fault strike trends show the Reelfoot fault and its hanging-wall Tiptonville dome are not laterally offset across the Cottonwood Grove and Ridgely faults. The Reelfoot South fault northwest and southeast of the Cottonwood Grove and Ridgely faults has very similar vertical displacement on common stratigraphic marker horizons in the upper 900 m. These data indicate the Reelfoot fault/Tiptonville dome has acted as one continuous fault zone across the Cottonwood Grove and Ridgely faults since Late Cretaceous.

Missouri, Tennessee↗