Earthquake Science Center Seminars

The discovery of unusual deformation phenomena like tectonic tremor, episodic tremor and slip, and slow earthquakes has renewed interest in friction and rheology of active faults. Indirect evidence of elevated pore pressure suggests that fluids likely play an important role. Many of the more unusual deformation modes appear to occur in the transition from brittle to ductile deformation and therefore can involve complex interactions of chemical, mechanical and poroelastic processes. We look at laboratory deformation tests of rate- and state-friction parameters, especially near critical loading stiffness. In this region, complex deformation patterns appear that suggest nonlinear fault properties. These include periodic stress oscillation (i.e., slow slip and creep episodes), period doubling, and spontaneous jumps between attractors in limit cycles. Many of these features can be understood in terms of rate- and state-friction, but some cannot.

We investigate spatial and temporal relations between an ongoing and prolific seismicity cluster in central Washington state, near Entiat, and the December 14, 1872 Entiat earthquake, the largest historic crustal earthquake in Washington east of the Cascades. A fault scarp produced by the 1872 earthquake lies within the Entiat cluster, and the locations and areas of both the cluster and the estimated 1872 rupture surface are comparable. Seismic intensities and the 1 to 2 m of coseismic displacement suggest a magnitude range between 6.5 and 7.0 for the 1872 earthquake. Aftershock forecast models for (1) the first several hours following the 1872 earthquake, (2) the largest felt earthquakes from 1900 to 1974, and (3) the seismicity within the Entiat cluster from 1976 through 2016 are also consistent with this magnitude range. Based on this aftershock modeling, most of the current seismicity in the Entiat cluster could represent aftershocks of the 1872 earthquake. Other earthquakes, especially those having long recurrence intervals, have long-lived aftershock sequences, including the Mw 7.5 1891 Nobi earthquake in Japan, with aftershocks continuing 100 years after the mainshock. Although we do not rule out ongoing tectonic deformation in this region, a long-lived aftershock sequence can account for these observations.

Dr. Astaneh after briefly summarizing the performance of the 1936 Bay Bridge during the 1989

Loma Prieta earthquake and discussing seismic performance of the 1936 West Suspension

Spans, will focus on the 2013 SAS Bay Bridge and share his findings on the causes and

consequences of more than 15 design, construction and inspection problems that have resulted in

making the seismic resistance of the new SAS Bay Bridge debatable, its duration of construction

unacceptably long, and its cost overruns astronomical. He will discuss the reasons why the SAS

Bay Bridge ended up being so problematic and the lessons learned from this project, with

emphasis on the questionable seismic performance of the main tower and the bridge as a whole

with more than 2,200 hydrogen-embrittled anchor rods and bolts used in the most seismically

critical connections of the new bridge. The talk will conclude with proposed repair and retrofit

plans that if implemented can mitigate the existing seismic hazards posed by the inevitable

fracture of the hydrogen-embrittled anchor rods and bolts and the local buckling of the tower

shafts, none being an acceptable seismic performance.

Correlations of ambient noise have been shown to be a useful tool for imaging Earth structure at a wide range of scales. Correlation functions are particularly useful on area not sampled by earthquake data. While noise based analyses were previously made with surface waves, body waves can also be extracted from correlation function even at teleseismic distances.

In this study, we analyzed the P-wave phases that travel under the North Atlantic Ocean between the US and Europe. We exploit the large set of data deployed in these regions, choosing year 2014 since the Transportable Array (USArray) was located in the Eastern part of the US. The final dataset contains roughly 900 stations distributed on both sides of the ocean.

The PcP phase reflected at the Core Mantle Boundary is particularly well reconstructed in the mid-period range (3-8 seconds) between distant arrays. We use a vespagram analysis on a set of sub-networks and compare the relative PcP to P phase arrivals with traveltime predictions to compute an image of the Core Mantle Boundary under the Northern Atlantic Ocean.

The earthquake activity in Oklahoma and Kansas that began in 2008 reflects the most widespread instance of induced seismicity observed to date. Given the continued high rates of seismicity in this area, including three M5+ earthquakes in 2016, it is imperative to develop methods that forecast future earthquake hazard accurately. In this work, we develop a method that combines a fluid pressurization model with rate-and-station friction theory to forecast rates of induced seismicity. Using injection well data from 893 saltwater disposal wells completed in the Arbuckle aquifer, we established a detailed description of the temporal and spatial variability in stressing conditions over the 21.5-year period from January 1995 through June 2017. Seismicity rate forecasts generated based on these stressing conditions replicate many salient features of seismicity in Oklahoma and southern Kansas including the onset of the sequence, the timing of the peak seismicity rate, and the reduction in seismicity following decreased disposal activity. Comparing our hydromechanical earthquake rate model to observational and empirical models that are commonly used to forecast seismicity rates, we find that our model outperforms them for 1-year forecast durations for the years 2008 through 2016.

A continually increasing number of high-quality digital strong-motion records from stations of the National Strong Motion Project (NSMP) of the U.S. Geological Survey (USGS), as well as data from regional seismic networks within the U.S., calls for automated processing of strong-motion records with human review limited to selected significant or flagged records. The NSMP has developed the Processing and Review Interface for Strong Motion data (PRISM) software to meet this need. PRISM automates the processing of strong-motion records by providing batch-processing capabilities. The PRISM software is platform-independent (coded in Java), open-source, and does not depend on any closed-source or proprietary software. The software consists of two major components: a record processing engine composed of modules for each processing step, and a graphical user interface (GUI) for manual review and processing. To facilitate the use by non-NSMP earthquake engineers and scientists, PRISM (both its processing engine and GUI components) is easy to install and run as a stand-alone system on common operating systems such as Linux, OS X and Windows. PRISM was designed to be flexible and extensible in order to accommodate implementation of new processing techniques. Input to PRISM currently is limited to data files in the Consortium of Organizations for Strong-Motion Observation Systems (COSMOS) V0 format, so that all retrieved acceleration time series need to be converted to this format. Output products include COSMOS V1, V2 and V3 files as: (i) raw acceleration time series in physical units with mean removed (V1), (ii) baseline-corrected and filtered acceleration, velocity, and displacement time series (V2), and (iii) response spectra, Fourier amplitude spectra and common earthquake-engineering intensity measures (V3). All computing features have been thoroughly tested.

The paleoseismic record from crustal faults in the Puget Lowland in western Washington State contains ~30 earthquakes that cut the late Pleistocene glacial surface. This rich dataset, produced by decades of scientific effort by the USGS and collaborators, has dramatic implications for seismic hazard and risk in this densely-populated and economically-vital area, and also provides insight into earthquake processes and fault interaction in crustal fault arrays. However, before the fruits of this labor can be snacked upon, more thorough characterization of these paleoearthquakes is necessary. To these ends, I present a three-pronged approach of ongoing research into this characterization:

1) Magnitude estimation: Paleoseismological techniques measure earthquake displacement at a study site, which is then used in scaling relationships to infer plausible magnitude ranges for each event. However, the fault geometry can add additional constraints to the magnitude as well. By using Bayesian methods that incorporate both rupture length and point displacement, we decrease both the magnitude and the uncertainty in the magnitude estimation for Puget Lowland earthquakes.

2) Earthquake recurrence and survival analysis: The expected time until the next earthquake may depend on the time since the last earthquake, but the form of this dependence is not always obvious. Using the tools of survival analysis (borrowed from biology and engineering), we can calculate the probability of earthquake occurrence in western Washington given the time since the last event based on the statistics of previous events, rather than by assuming a standard statistical model for earthquake recurrence. This empirical technique lets us directly incorporate phenomena such as earthquake triggering and clustering (which are clearly observed in the Puget Lowland) that standard recurrence models cannot.

3) Time-dependent seismic hazard and loss analysis: The paleoearthquake record from the Puget Lowland provides quite strong evidence for earthquake triggering and fault interaction to a degree that invalidates the assumptions of time- and space-independence used in classical PSHA. In particular, records of clustered M7+ events within a few years to decades imply that communities with greatly damaged infrastructure from one event may face successive episodes of strong ground shaking. I present very preliminary work describing how expected seismic hazard and loss within a 50-year window may be strongly affected by earthquake clustering, and plans for incorporating cumulative infrastructure damage and repair into PSHA.

The possibility that a large portion of the Southern San Andreas Fault (SSAF) could be activated in a M>7.5 earthquake remains one of the most important issues for the correct evaluation of the seismic hazard in Southern California. The commonly accepted southern termination of the SSAF is halfway along the eastern margin of the Salton Sea, near Bombay Beach. The area between the southern end of the SSAF and the northern end of the Imperial Fault (IF) is known as the Brawley Seismic Zone (BSZ). The BSZ is characterized by small- to medium-sized earthquakes, which often manifest as seismic swarms. Recent cases of such swarms in this area include the 2012 Brawley sequence (Hauksson et al., 2013), as well as a more recent case in the fall of 2016 (Hauksson et al., 2017) that occurred immediately adjacent to the southern terminus of the SSAF. In spite of this seismic activity, to date there is no known trace of the SSAF extending south of Bombay Beach, or a trace of the IF extending north to link the two large faults. For that reason, the SSAF and IF are often considered to have little to no seismic interaction. However, paleoseismological evidence of coincident slip at the Dogwood site on the IF (Rockwell et al., 2011) and Coachella Valley (Philibosian et al., 2011) on the SSAF suggest that a through-going rupture might be possible. A major event involving the activation of both SSAF and IF could reach a magnitude of up to 8. Such an event could disrupt Interstate 8, which has long been considered to be a "safe" path in the worst-case scenario of the much anticipated "big one" in southern California. The main topics of this talk are: 1) the possibility of joint rupture of the Southern San Andreas and Imperial Faults, including the role of intersecting Cross faults; 2) the possibility of a major SSAF event being triggered by a small swarm-type earthquake in the BSZ. We are targeting the above questions by performing spontaneous dynamic rupture models that investigate a large spectrum of geometric and stress configurations. Our preliminary results show that, under certain prestress conditions, a through-going rupture is possible via a joint SSAF and IF activation. The implementation of the cross faults in the model reveals a wide variety of interaction possibilities that might affect rupture propagation. Finally, we are exploring a rupture scenario for which a moderate event will initiate on a cross fault (e.g., as part of a local seismic swarm) and propagate on the adjacent main fault, and thus cascade into a large SSAF earthquake.

To better understand how earthquake-generated shaking mobilizes and redistributes submarine sediments I used seismic data from the 2011-2015 offshore Cascadia Initiative and permanent onshore seismic networks to measure three simple site-response measures (PGV, PGA and energy density), in low (.02-1 Hz) and high (1-10 Hz) frequency passbands. These metrics show similar patterns, and vary by factors of almost 10 to 60 for low and high frequencies, respectively, across all of Cascadia. Site-response variability is primarily longitudinal, with the caveat that few observations sample the continental shelf in southern Cascadia. Offshore site-response estimates have nearly opposite patterns at low and high frequencies, which broadly may be attributed to sediment resonance and attenuation. However, the patterns are not visually or statistically fully consistent with the mapped sediment distribution and only weakly consistent with peaks in horizontal to vertical spectral ratios often attributed to resonance. This undoubtedly partly reflects sediment thickness model inaccuracies, but the topography likely has a profound impact on the shaking, evident in an abrupt change in the longitudinal trend of site-response along most of the Cascadia margin. This change coincides with the base of the steepest slopes that border the shelf. Offshore site-response exceeds that offshore in both passbands, which should be accounted for when extrapolating from onshore ground motions to estimate those offshore. The patterns of site-response alone do not shed light on what frequency passband is most important for triggering slope failures and remobilizing sediments, but if low frequencies are the most effective triggers, suggest that where slopes are steepest site-response facilitates triggering. Finally, site-response patterns neither support nor permit rejection of the hypothesis that the southerly decrease in great earthquake recurrence intervals estimated by Goldfinger et al., [2012, 2013, 2016] and Priest et al. [2017] is attributable to inherently stronger shaking going southward, though significant north-south variations in site-response are not apparent.