Earthquake Science Center Seminars

In the past decade, Oklahoma has experienced an unprecedented 100-fold increase in seismicity rates, including 4 events with Mw >= 5. These earthquakes affect infrastructure that was designed with little to no consideration of seismic loads, leading to some damages and widespread unrest in the public. Our goal was to assess the time-dependent nature of the seismic hazard and risk in Oklahoma and identify how much of the seismicity rate increase is related to wastewater injection, a widely used process in the oil and gas industry.

To that end, we develop a semi-empirical model to simulate the observed seismicity given the injection time-history. The proposed earthquake recurrence model is a modified version of the Gutenberg-Richter relation and builds upon the Seismogenic Index model [Shapiro et al., 2010] and the stressing-rate dependency of the time lag between injection and changes in the seismicity rates. First, we employ this model to test the statistical significance of the prevailing hypothesis that the seismicity rate changes are driven by pore-pressure changes related to wastewater injection. The results show that 84% of the recent felt seismicity in Oklahoma can be associated with wastewater injection at a 95% confidence level.

We also forecast seismicity rates given future injection scenarios and analyze the observed magnitude-frequency distributions, arguing that the reported elevated values of the Gutenberg-Richter b-value are an artifact of the finiteness of the pore-pressure perturbation areas around the injection wells. In terms of forecasting power, the model is able to predict the evolution of the seismicity burst starting in 2014, both in terms of timing and magnitude, even when only using seismicity data through 2011 for the calibration. Under a conservative scenario of injection rates, the seismicity is expected to reach pre-2009 levels after 2025, while the probability of a potentially damaging event remains at substantial levels.

Finally, we perform a probabilistic assessment of the time-dependent seismic hazard in Oklahoma and incorporate these results into an integrated seismic risk model to assess the evolution of the statewide potential for earthquake-induced losses. The resulting seismic hazard maps reach at times levels found across the San Andreas fault, illustrating the incompatibility of the seismic provisions in the state with the observed seismicity in recent years. Our direct-loss estimates are in reasonable agreement with the paid insurance claims, but show significant sensitivity to the ground motion model selection. During the peak of seismicity in 2015, the seismic risk was 275 times higher than the background level, with the majority of the losses originating from damages to non-structural elements.

While substantial progress has been made in understanding how and where strain accumulates on faults, enabling the identification of fault segments that are more likely to host earthquakes, the mechanical behavior of some key features of the earthquake rupture remains unclear. For instance, difficulties arise when evaluating the precise effect of fault geometry during earthquake rupture. While physics-based modeling offers a first, elegant, explanation for the main features of surface data, residuals are often interpreted solely as the results of the system's own complexity. In this presentation, I will present examples, from fault roughness to thrust mechanics, of how complexity can also emerge from "simple" physics-based models to help us understand key features of earthquake physics.

Observations of surface deformation caused by earthquakes provide insights into earthquake mechanisms and rheological structure. So far, geodetic observations have been mostly limited to shallow earthquakes since the amplitude of surface deformation from deep events is considered relatively small. In this talk, I will show the detections of post-seismic signals from deep earthquakes, which open up new opportunities of understanding mantle viscosity and deep rupture processes. We examine the GPS data of one of the largest deep earthquakes ever recorded, 2018 Mw 8.2 Fiji earthquake which occurred at ~600-km depth. Taking advantage of data processing techniques such as independent component analyses, we detect a large scale post-seismic deformation that has been taking place for more than one and half years. I will discuss what we can learn about the viscosity structure and the existence of afterslip from the overall directionality and amplitude of the post-seismic deformation.

In addition to earthquakes, seismometers record ground motions generated by various environmental surficial processes, the ocean, the atmosphere, and anthropogenic activities, and the continuous seismic records are mainly these non-earthquake signals. However, these signals are complex and difficult to interpret, leaving their source mechanisms poorly understood. Therefore, these signals are often considered as noise and have received limited observational or theoretical attention. Here, using ten years of continuous records from more than 2000 seismic stations spanning the whole U.S. continent and a novel surface wave detection method, we identify and locate various unusual environmental seismic sources. In this seminar, I will show the discovery of abundant submarine landslides in the Gulf of Mexico. Specifically, we find 85 previously unknown submarine landslides in the Gulf of Mexico. Ten of the 85 landslides occurred spontaneously without preceding earthquakes, while the remaining 75 occurred nearly instantaneously after the passage of surface waves from distant earthquakes, and hence were likely dynamically triggered by the earthquakes. I will also discuss a newly discovered geophysical phenomenon - stormquakes, involving the coupling of the atmosphere-ocean and solid Earth. We discovered that large storms such as hurricanes and Nor'easters can excite coherent transcontinental seismic surface waves as large as those excited by M3.5 earthquakes. These stormquakes are fundamentally different from previously reported atmosphere-ocean-solid Earth couplings that produce incoherent seismic noise. These discoveries suggest the richness of the continuous seismic wavefield and hint future research directions.

We investigate how the choice of alerting threshold for a given shaking intensity target affects the quality of ShakeAlert earthquake early warning (EEW) alerts using a suite of earthquakes in the west coast of the United States. The ShakeAlert system takes a source-based approach to EEW, where the alert region for a specified modified Mercalli intensity (MMI) is computed from ground-motion models using magnitude and location estimates as the main inputs. Currently, median-expected ground motions are used in the alert region calculation, which is unable to reliably alert all of the regions that experience shaking at the MMI level used to generate the alert. To ameliorate this, EEW systems can expand the size of the alerting region for a target MMI level (alert target) by choosing a lower MMI for the alert region computation. However, this comes with the tradeoff of increasing the number of inaccurate alerts issued to people who experience shaking below the alert target. We aim to find optimal MMI thresholds for alert targets ranging between MMI 4.0-6.0 using a ShakeMap catalog of 140+ M5.0-7.3 earthquakes in California, Oregon, and Washington. We compute the quality of each alerting strategy relative to the alert target in terms of the area and population alerted. Additionally, we evaluate the alert quality of the current ShakeAlert alerting thresholds for this range of MMI alert targets. We further discuss how this MMI threshold alerting strategy is equivalent to incorporating ground-motion-model uncertainty into the alert region computation, and we compare the optimal MMI thresholds found by this analysis with the combined standard deviation of the ground-motion models used by ShakeAlert.

This seminar by Dr. Morton Ender is part of the ESC Equity, Inclusion, and Diversity Working Groups efforts as we seek to improve life for underrepresented groups in the Earth Sciences and our Center and understand the impacts of earthquakes on different parts of our society. We hope you will join us for his seminar and discussion to help us build our communal understanding of these issues in the U.S. The abstract is below.

Seven major social characteristics are consistently contested in American society: race/ethnicity; sex/gender; sexual orientation; religious affiliation; age; socio-economic status; and (dis)ability. I will briefly discuss these seven areas in American society and in the U.S. military. Additionally, I will touch upon intersectionality, rugged individualism, and the ongoing citizenship revolution in the U.S. I will leave ample time for Q&A and discussion. This presentation originates from two publications: Morten G. Ender and Betsy Lucal. "Social Inequality and the United States Army: The (Un)Lucky Seven," in R.E. Tully and B. Chilton (eds.) Equality - More or Less (Hamilton Books, 2020) See: https://rowman.com/isbn/9780761871170 and http://www.bard.edu/news/releases/pr/fstory.php?id=2886 and David E. Rohall, Morten G. Ender, Michael D. Matthews (eds.). (2017). Inclusion in the U.S Military: A Force for Diversity. (Lexington Books) See: https://rowman.com/ISBN/9781498528603/Inclusion-in-the-American-Military-A-Force-for-Diversity.

I will report on the tectonic framework, seismicity, and monitoring efforts related to the March 31, 2020 Mw 6.5 Stanley, Idaho earthquake. The earthquake sequence has produced both strike-slip and dip-slip motion, and defines the northern limits of the Sawtooth normal fault. This fault separates the Centennial Tectonic Belt, a Basin and Range subprovince, from the Idaho batholith to the west and Challis volcanic belt to the north and east. I will show evidence for a kinematic link between the east-dipping Sawtooth fault and the west-dipping Lost River fault. These opposing faults have recorded four of the five M>=6 Idaho earthquakes from the past 100 years, including 1983 Mw6.9 Borah Peak and the 1944 M6.1 and 1945 M6.0 Seafoam earthquakes. Geological and geophysical data are consistent with fault boundaries, where segmentation may be influenced by preexisting geologic structures. I will show that the limits of both the Sawtooth and Lost River faults extend north beyond their mapped extent, are influenced by the relic Trans-Challis fault system, and that seismicity within this region will likely continue for the coming years. Ongoing seismic monitoring efforts will lead to an improved understanding of ground shaking potential and active fault characteristics.

The forearc mantle wedge plays a critical role in the geodynamics of subduction zones. From five perspectives, I will highlight its thermal, petrologic, and mechanical states and how it affects megathrust slip behaviour. (1) Heat flow data and seismic imaging indicate that the forearc mantle wedge is cold, referred to as the "cold nose", in sharp contrast with the hot arc-backarc region. With fluids supplied by the dehydrating slab, hydrous minerals form in the ultramafic cold nose. (2) Maintaining the cold and stable thermo-petrologic state requires the forearc wedge to be fully decoupled from the subducting slab and does not participate in mantle wedge flow. This is supported by the lack of seismic anisotropy in the cold nose as inferred from local-earthquake shear wave splitting analysis in the Japan Trench subduction zone. (3) The cold state gives rise to high stiffness, affecting postseismic deformation following megathrust earthquakes. The mechanical contrast of the cold nose with the rest of the mantle wedge is clearly reflected in geodetically observed postseismic uplift between the cold nose and the volcanic arc. (4) In warm-slab subduction zones, a very high degree of serpentinization of the tip area of the mantle wedge is expected to diminish permeability. The hydrological consequence is inferred to foster a geological condition for episodic tremor and slip (ETS) downdip of, but separated from, the megathrust seismognic zone. (5) In colder-slab subduction zones, serpentinite derived from the base of the forearc mantle wedge affects the mechanics of the megathrust fault zone, resulting in complex seismogenic behaviour downdip of the Moho-megathrust intersection. I will use the 2010 M=8.8 Maule, Chile, earthquake as an example to show how this process may affect coseismic slip, stress drop, and aftershock distribution under the specific P/T condition in this area.

Geotechnical engineers have noted that in many places there is significant variability in the near surface over much shorter spatial scales than what is measured by traditional techniques. Over the past decade new seismic sensing technologies such as distributed acoustic sensing have enabled continuous, high-density seismic acquisition over long distances. The ability to install cables quickly or to plug into existing telecommunications infrastructure have enabled engineering geophysics around infrastructure and in urban areas. Further lowering costs, these low-labor acquisitions have been analyzed with passive seismic methods for near-surface imaging and earthquake engineering. While data are now incredibly easy to acquire, processing these data has been a challenge both due to large data volumes (multi-terabyte per day) and due to challenging noise environments in urban areas. New algorithms including machine learning are increasingly required to analyze these data. These new techniques will be discussed in the context of several recent experiments with applications in earthquake engineering and near-surface geohazards.

Distributed Acoustic Sensing (DAS) is an emerging photonic method that transforms standard telecommunication optical fibers into meter-scale, watershed-aperture geophysical arrays. Such an idea has numerous possibilities in geoscience and civil engineering, for example in urban seismology, marine geophysics, and the cryosphere where observation is limited by access. The fomenting potential of DAS is belied by the method's sparse calibration to-date. Developing an understanding of DAS instrument response is necessary for seismic moment tensor inversion, full-wavefield inversion, attenuation studies, and geodetic measurements of absolute strain. In this talk, evidence from recent laboratory and field experiments will be used to characterize DAS in terms of its frequency sensitivity, amplitude response, phase response, self-noise, and dynamic range. Behavior in the laboratory does not necessarily apply in the field due to fiber coupling and cable construction. To address this, two in situ field techniques will be proposed that use available seismological signals for calibration. Results will be shown from three field campaigns: FORGE geothermal microseismicity, regional and teleseismic earthquake observations in rural California, and microseisms recorded on the seafloor of Monterey Bay. I will conclude by summarizing the status of industry calibration standards, and discuss how greater awareness of DAS instrument response including inter-array comparisons could benefit the community.