Geology topics

Annemarie S. Baltay

Publications and source records attributed to Annemarie S. Baltay.

At least 37 records · Page 2Linked to original sources

USGS2024

GRAPES: Earthquake early warning by passing seismic vectors through the grapevine

Estimating an earthquake's magnitude and location may not be necessary to predict shaking in real time; instead, wavefield-based approaches predict shaking with few assumptions about the seismic source. Here, we introduce GRAph Prediction of Earthquake Shaking (GRAPES), a deep learning model trained to characterize and propagate earthquake shaking across a seismic network. We show that GRAPES’ internal activations, which we call “seismic vectors”, correspond to the arrival of distinct seismic phases. GRAPES builds upon recent deep learning models applied to earthquake early warning by allowing for continuous ground motion prediction with seismic networks of all sizes. While trained on earthquakes recorded in Japan, we show that GRAPES, without modification, outperforms the ShakeAlert earthquake early warning system on the 2019 M7.1 Ridgecrest, CA earthquake.

Shimane/HiroshimaPrefectures↗

USGS2024

Summary of the discussions during 2023 SSA topical meeting on “Future Directions for Physics-Based Ground Motion Modeling”

The Seismological Society of America (SSA) topical conference, Future Directions for Physics‐Based Ground Motion Modeling, was held in Vancouver, Canada, on 10–13 October 2023, co‐sponsored by the Seismological Society of Japan and co‐chaired by Annemarie Baltay of the U.S. Geological Survey and Hiroshi Kawase of Kyoto University. This meeting brought together many researchers and practitioners interested in modeling, observing, and utilizing ground‐motion models (GMMs). Scientists gathered to discuss complex kinematic and dynamic rupture simulation approaches, empirical representations of the earthquake source, site and path effects, physical modeling of the recording site, challenges for model extrapolation, and overall prediction accuracy and simulation validation. The four‐day meeting included many posters as well as oral presentations, with each session followed by lively discussion sections, upon which we report here.

Seismological Research Letters↗

USGS2024

Integration of rupture directivity models for the US National Seismic Hazard Model

Several rupture directivity models (DMs) have been developed in recent years to describe the near-source spatial variations in ground motion amplitudes related to propagation of rupture along the fault. We recently organized an effort towards incorporating these directivity effects into the USGS National Seismic Hazard Model (NSHM), by first evaluating the community's work and potential methods to implement directivity adjustments into probabilistic seismic hazard analysis (PSHA). Guided by this evaluation and comparison among the considered DMs, we selected an approach that can be readily implemented into the USGS hazard software, that provides an azimuthally varying adjustment to the median ground motion and its aleatory variability. This method allows assessment of the impact on hazard levels and provides a platform to test the DM amplification predictions using a generalized coordinate system, necessary for consistent calculation of source-to-site distance terms for complex ruptures. We give examples of the directivity-related impact on hazard, progressing from a simple, hypothetical rupture, to more complex fault systems, composed of multiple rupture segments and sources. The directivity adjustments were constrained to strike-slip faulting, where DMs have good agreement. We find that rupture directivity adjustments using a simple median and aleatory adjustment approach can impact hazard both from a site perspective and on a regional scale, increasing shaking off the end of the fault trace up to 30--40\% and potentially reducing it for sites along strike. Statewide hazard maps of California show that the change in shaking along major faults can be a factor to consider for assessing long-period (>ls) near-source effects within the USGS NSHM going forward, reaching up to 10--20\%. Finally, we suggest consideration of minimum parameter ranges and baseline requirements as future DMs are developed to minimize single approach adaptations, to enable more consistent application within both ground motion and hazard studies.

Earthquake Spectra↗

USGS2024

Incorporating intensity distance attenuation into PLUM ground-motion-based earthquake early warning in the United States: The APPLES configuration

We develop Attenuated ProPagation of Local Earthquake Shaking (APPLES), a new configuration for the United States West Coast version of the Propagation of Local Undamped Motion (PLUM) earthquake early warning (EEW) algorithm that incorporates attenuation into its ground-motion prediction procedures. Under APPLES, instead of using a fixed radius to forward-predict observed peak ground shaking to the area surrounding a seismic station, the forward-predicted intensity at a location depends on the distance from the station using an intensity prediction relationship. We conduct conceptual tests of maximum intensity distribution predictions in APPLES and PLUM using a catalog of ShakeMaps to confirm that the attenuation relationship in APPLES is appropriately modeling shaking distributions for West Coast earthquakes. Then, we run APPLES and PLUM in simulated real-time tests to determine warning time performance. Finally, we compare real-time alert behavior during the 2022 M 6.4 Ferndale, California, earthquake and other recent events. We find that APPLES presents two potential improvements to PLUM by reducing over-alerting during smaller magnitude earthquakes and by increasing warning times in some locations during larger earthquakes. APPLES can produce missed and late alerts in locations that experience shaking intensities close to the level used to issue alerts, so preferred alerting strategies with APPLES would use alert thresholds that are lower than the intensities targeted for EEW alerts. We find alerts using APPLES are also similar to those for the source-based approaches currently used in the ShakeAlert EEW system, which will make APPLES easier to integrate into the system.

California, Oregon, Washington↗

USGS2023

Rupture directivity of the 25 October 2022 Mw 5.1 Alum Rock earthquake

The 25 October 2022 M w 5.1 Alum Rock earthquake shows strong evidence for southeast rupture directivity along the central Calaveras fault (CCF), as indicated by observed ground motions and simulated kinematic ruptures. Peak ground accelerations (PGAs) and peak ground velocities (PGVs) are notably higher to the southeast, with an order of magnitude difference for stations at the same distance but different azimuths. In addition, PGAs are lower than that predicted by ground‐motion models by a factor of 3 on average in all the directions, indicating a low stress drop (∼1.57 MPa). Directivity function modeling and ground‐motion simulations both indicate rupture propagation to the southeast with rupture velocity between 2.3 and 2.5 km/s. We suggest that the southward rupture propagation and relatively low stress drop may be typical of M ∼5 earthquakes on this portion of the CCF.

Califronia↗

USGS2023

Applications of nonergodic site response models to ShakeAlert case studies in the Los Angeles area

In this study, we explore whether the Parker and Baltay (2022) site response models for the Los Angeles (LA) basin region can improve ground‐motion forecasts in the U.S. Geological Survey ShakeAlert earthquake early warning system (hereafter ShakeAlert). We implement the peak ground acceleration and peak ground velocity site response models of Parker and Baltay (2022) in ShakeAlert via the earthquake information to ground‐motion (hereafter eqinfo2GM) module, which predicts ground motions from the estimated earthquake parameters of magnitude, rupture length, and location. The nonergodic site response models for the greater LA area were developed using ground motions from 414 M 3–7.3 earthquakes in southern California. We test nonergodic ground‐motion forecasts for five earthquakes in the LA area: the 1994 M 6.7 Northridge earthquake, the 2008 M 5.4 Chino Hills earthquake, the 2019 M 7.1 Ridgecrest earthquake, the 2020 M 4.5 South El Monte earthquake, and a synthetic M 7.8 earthquake on the southern San Andreas fault from the ShakeOut scenario, which was the basis of a statewide emergency response exercise. From the test results, we find that with the nonergodic site response applied, ShakeAlert not only alerts larger areas but can also result in longer warning times in LA region. In addition, the modified Mercalli intensity (MMI) ground‐motion predictions generated by the ShakeAlert eqinfo2GM module are improved in accuracy when compared with the corresponding ShakeMap ground‐truth MMI when the nonergodic site response model is applied.

California↗

USGS2022

Estimates of k0 and effects on ground motions in the San Francisco Bay area

Ground‐motion studies are a key component of seismic hazard analyses and often rely on information of the source, path, and site. Extensive research has been done on each of these parameters; however, site‐specific studies are of particular interest to seismic hazard studies, especially in the field of earthquake engineering, as near‐site conditions can have a significant impact on the resulting ground motion at a site. There has been much focus on the constraint of site parameters and their application to seismic hazard studies, especially in the development of ground‐motion models (GMMs). Kappa is an observational parameter describing the high‐frequency attenuation of spectra, and its site contribution ( ⁠ κ 0 "> κ 0 ) has shown to be a good predictor of high‐frequency ground motions; however, measurements are often limited. In this study, we develop a κ 0 "> κ 0 dataset for the San Francisco Bay area (SFBA) by estimating κ 0 "> κ 0 for 228 stations, and we produce a continuous regional map of κ 0 "> κ 0 ⁠ . We find κ 0 "> κ 0 to range between 0.003 and 0.072 s, with larger values concentrating on the east, north, and south sides of the bay, and lower values concentrating on the west side. We also evaluate the robustness of κ 0 "> κ 0 as a site parameter and find it to correlate with peak ground acceleration. These estimates of κ 0 "> κ 0 can add predictive power to GMMs, thus increasing the accuracy of predicted ground motion and improving the robustness of ground‐motion studies in the SFBA.

California↗

USGS2022

Real-time earthquake detection and alerting behavior of PLUM ground-motion-based early warning in the United States

We examine the real‐time earthquake detection and alerting behavior of the Propagation of Local Undamped Motion (PLUM) earthquake early warning (EEW) algorithm and compare PLUM’s performance with the real‐time performance of the current source‐characterization‐based ShakeAlert System. In the United States (U.S.), PLUM uses a two‐station approach to detect earthquakes. Once a detection is confirmed, observed modified Mercalli intensity (MMI) distributions are forecast onto a regular grid, in which the preferred alert regions are grid cells with MMI 4.0+ forecasts. Although locations of dense station coverage allow PLUM to detect small ( M < 4.5) earthquakes typically not considered for EEW in the U.S., a PLUM detection on a small earthquake does not always generate an alert. This is because PLUM alerts are determined by current shaking distributions. If the MMI 4.0+ shaking subsides prior to detection confirmation by shaking at a second neighboring station, the prior MMI 4.0+ information will not be in the alert forecasts. Of the 432 M 3.0+ U.S. West Coast earthquakes in 2021, 33 produced ground motions large enough to be detected by PLUM. Twenty‐four generated MMI 4.0+ PLUM alerts, whereas ShakeAlert issued public EEW alerts for 13 of these earthquakes. We compare PLUM and ShakeAlert alert regions with ShakeMap and “Did You Feel It?” intensity distributions. Because PLUM alert regions surround stations observed to have strong ground motions (regardless of earthquake magnitude), PLUM alerts reliably include locations that experience significant shaking. This is not necessarily the case for ShakeAlert alert regions when there are large errors in magnitude or epicenter estimates. For two of the largest earthquakes in our real‐time dataset, the M 6.0 Antelope Valley and M 5.1 Petrolia earthquakes, the inclusion of PLUM would have improved real‐time ShakeAlert performance. Our results indicate that incorporation of PLUM into ShakeAlert will improve the robustness of the EEW system.

California, Oregon, Washington↗

USGS2022

What to expect when you are expecting earthquake early warning

We present a strategy for earthquake early warning (EEW) alerting that focuses on providing users with a target level of performance for their shaking level of interest (for example, ensuring that users receive warnings for at least 95 per cent of the occurrences of that shaking level). We explore the factors that can affect the accuracy of EEW shaking forecasts including site conditions (which can make shaking more or less varied), the share of earthquake hazard that comes from outlier events, and even the metric of shaking being forecast (with modified Mercalli intensity being the most uncertain quantity to forecast). To provide different locations with the same level of EEW performance, the alert issuance criteria will need to change based on these factors. Finally, we present a hierarchy of information approach with which to communicate EEW system design and performance to the public and other stakeholders, including how frequently alerts will be issued and how accurate those alerts will be.

Geophysical Journal International↗

USGS2022

How low should we alert? Quantifying intensity threshold alerting strategies for earthquake early warning in the United States

We use a suite of historical earthquakes to quantitatively determine earthquake early warning (EEW) alert threshold strategies for a range of shaking intensity targets for EEW in the U.S. West Coast. The current method for calculating alert regions for the ShakeAlert EEW System does not take into account variabilities and uncertainties in shaking distribution. As a result, if the modified Mercalli intensity (MMI) level used to determine the extent of the alert region (the alert threshold) is the same as the target intensity threshold, the alert region will be too small to include all locations that require alerts even if earthquake source parameters are estimated accurately. Missed alerts can be reduced by using a lower alert threshold than the target threshold. This expands the alert region, increasing the number of precautionary alerts issued to people who experience shaking below the target level. We determine alert thresholds that optimize this tradeoff between missed and precautionary alerts for target thresholds of MMI 4.0-6.0 using a ShakeMap catalog of 143 M 5.0-7.3 earthquakes as ground truth. We examine the quality of each alerting strategy relative to the target MMI, where we define alert quality metrics in terms of both the area and population alerted. Optimal alert thresholds maximize correct alerts while limiting most precautionary alerts to regions that are likely to still feel some shaking. We find these optimal alert thresholds also maximize warning times. This analysis presents a quantitative framework ShakeAlert can use to communicate alerting strategies and performance expectations to ShakeAlert users.

Earth's Future↗

USGS2022

Empirical map-based nonergodic models of site response in the greater Los Angeles area

We develop empirical estimates of site response at seismic stations in the Los Angeles area using recorded ground motions from 414 M 3–7.3 earthquakes in southern California. The data are from a combination of the Next Generation Attenuation‐West2 project, the 2019 Ridgecrest earthquakes, and about 10,000 newly processed records. We estimate site response using an iterative mixed‐effects residuals partitioning approach, accounting for azimuthal variations in anelastic attenuation and potential bias due to spatial clusters of colocated earthquakes. This process yields site response for peak ground acceleration, peak ground velocity, and pseudospectral acceleration relative to a 760 m/s shear‐wave velocity ( ⁠ V S "> V V S "> s ) reference condition. We employ regression kriging to generate a spatially continuous site response model, using the linear site and basin terms from Boore et al. (2014) as the background model, which depend on V S "> V V S "> s V S "> 30 ⁠ and depth to the 1 km/s V S "> V V S "> s isosurface. This is different from past approaches to nonergodic models, in which spatially varying coefficients are regressed. We validate the model using stations in the Community Seismic Network (CSN) that are in the middle of our model spatial domain but were not considered in model development, finding strong agreement between the interpolated model and CSN data for long periods. Our model could be implemented in regional seismic hazard analyses, which would lead to improvements especially at long return periods. Our site response model also has potential to improve both ground‐motion accuracy and warning times for the U.S. Geological Survey ShakeAlert earthquake early warning (EEW) system. For a point‐source EEW simulation of the 1994 M 6.7 Northridge earthquake, our model produces ground motions more consistent with the ground‐truth ShakeMap and would alert areas with high population density such as downtown Los Angeles at lower estimated magnitudes (i.e., sooner) than an ergodic model for a modified Mercalli intensity 4.5 alerting threshold.

California↗

USGS2022

Earthquake early warning for estimating floor shaking levels of tall buildings

This article investigates methods to improve earthquake early warning (EEW) predictions of shaking levels for residents of tall buildings. In the current U.S. Geological Survey ShakeAlert EEW system, regions far from an epicenter will not receive alerts due to low predicted ground‐shaking intensities. However, residents of tall buildings in those areas may still experience significant shaking due to the acceleration amplification caused by tall buildings’ dynamic behavior, as recently experienced by residents of the 52‐story building in downtown Los Angeles (DTLA) during the 2019 M 7.1 Ridgecrest earthquake. Using more than 400 recorded response data acquired from 77 instrumented buildings in California, here we compare the Federal Emergency Management Agency (FEMA) P‐58 and American Society of Civil Engineers (ASCE) 7‐16 simplified equations for peak floor acceleration (PFA), finding that the ASCE estimation is close to the median of data recorded in large and long‐distance events, whereas the current FEMA estimation is not suitable. In the second part of this article, four instrumented tall buildings in DTLA are extensively studied, and the performance of the simplified and response spectrum (RS) methods giving both an estimation of the free‐field horizontal peak ground acceleration (PGA) and pseudospectral acceleration is evaluated. The results show that the RS method is as accurate as the response history analysis as long as the ground‐motion RS is accurate, whereas the ASCE 7‐16 prediction is conservative. However, when ground‐motion RS or PGA is estimated for DTLA using a ground‐motion model (GMM), the performance of the RS method significantly degrades due to underestimation by the GMM at long periods. The results of this study imply that a nonergodic GMM, which may give more accurate prediction in Los Angeles, could improve the results for PFA when the building’s behavior is dominated by a few long‐period fundamental modes, as is the case for the 52‐story building in DTLA.

California↗

USGS2022

Alert optimization of the PLUM earthquake early warning algorithm for the western United States

We determine an optimal alerting configuration for the propagation of local undamped motion (PLUM) earthquake early warning (EEW) algorithm for use by the U.S. ShakeAlert system covering California, Oregon, and Washington. All EEW systems should balance the primary goal of providing timely alerts for impactful or potentially damaging shaking while limiting alerts for shaking that is too low to be of concern (precautionary alerts). The PLUM EEW algorithm forward predicts observed ground motions to nearby sites within a defined radius without accounting for attenuation, avoiding the earthquake source parameter estimation step of most EEW algorithms. PLUM was originally developed in Japan where the alert regions and ground motions for which alerts are issued differ from those implemented by ShakeAlert. We compare predicted ground motions from PLUM to ShakeMap‐reported ground motions for a set of 22 U.S. West Coast earthquakes of magnitude 4.4–7.2 and evaluate available warning times. We examine a range of prediction radii (20–100 km), thresholds used to issue an alert (alert threshold), and levels of impactful or potentially damaging shaking (target threshold). We find optimal performance when the alert threshold is close to the target threshold, although higher target ground motions benefit from somewhat lower alert thresholds to ensure timely alerts. We also find that performance, measured as the cost reduction that a user can achieve, depends on the user’s tolerance for precautionary alerts. Users with a low target threshold and high tolerance for precautionary alerts achieve optimal performance when larger prediction radii (60–100 km) are used. In contrast, users with high target thresholds and low tolerance for precautionary alerts achieve better performance for smaller prediction radii (30–60 km). Therefore, setting the PLUM prediction radius to 60 km balances the needs of many users and provides warning times of up to ∼20 s.

California, Oregon, Washington↗

USGS2022

Spatially continuous models of aleatory variability in seismic site response for southern California

We develop an empirical, spatially continuous model for the single-station within-event ( ϕ SS ) component of earthquake ground motion variability in the Los Angeles area. ϕ SS represents event-to-event variability in site response or remaining variability due to path effects not captured by ground motion models. Site-specific values of ϕ SS at permanent seismic network stations were estimated during our previous work [1]. Here we first fit a model to ϕ SS conditioned on the time-averaged shear wave velocity in the upper 30 m (V S30 ). We observe that stations on soft soil have larger average variability in site response than those on rock, especially at short periods. We use regression kriging to spatially interpolate ϕ SS to an even grid spacing, using the V S30 -scaling as the background model [2]. This improves on previous work that used ordinary kriging for interpolation [1]. We find that ϕ SS ranges from about 0.1 to 0.4 natural log units in our study area, representing variations in site response at single locations of a factor of 1.1 up to a factor of 1.5. There is both greater variability and more coherency in the variability for short-period site response than for long periods. We recommend using these ϕ SS models with the site response models of [1] for applications where quantification of variability is needed, such as probabilistic seismic hazard analyses.

California↗

USGS2022

Estimates of kappa in the San Francisco Bay area

Site characterization is a critical component of seismic hazards studies, especially in the development and use of ground motion models (GMMs). One such parameter, kappa ( Κ 0 ), represents local site attenuation and effectively describes regional variations in ground motion [1]. However, estimates of Κ 0 are limited. We estimate the site parameter Κ 0 for 296 broadband and accelerometer stations in the San Francisco Bay area using 225 local events with M>3.5. We first invert for kappa on the high-frequency slope of every record at a station, which represents full attenuation effects [2]. We then perform weighted regression of kappa versus distance to obtain expected kappa at the site ( Κ 0 ). We find that Κ 0 varies extensively in the San Francisco Bay area, with values ranging from 0.004-0.113 s, thus representing the heterogeneous geologic conditions in the region. Κ 0 estimates from this study will be useful to consider in future seismic hazard analyses and ground motion studies in the Bay Area.

California↗

USGS2021

The impact of 3D finite‐fault information on ground‐motion forecasting for earthquake early warning

We identify aspects of finite‐source parameterization that strongly affect the accuracy of estimated ground motion for earthquake early warning (EEW). EEW systems aim to alert users to impending shaking before it reaches them. The U.S. West Coast EEW system, ShakeAlert, currently uses two algorithms based on seismic data to characterize the earthquake’s location, magnitude, and origin time, treating it as a point or line source. From this information, ShakeAlert calculates shaking intensity and alerts locations where shaking estimates exceed a threshold. Several geodetic EEW algorithms under development would provide 3D finite‐fault information. We investigate conditions under which this information produces sufficiently better intensity estimates to potentially improve alerting. Using scenario crustal and subduction interface sources, we (1) identify the most influential source geometry parameters for an EEW algorithm’s shaking forecast, and (2) assess the intensity alert thresholds and magnitude ranges for which more detailed source characterization affects alert accuracy. We find that alert regions determined using 3D‐source representations of correct magnitude and faulting mechanism are generally more accurate than those obtained using line sources. If a line‐source representation is used and magnitude is calculated from the estimated length, then incorrect length estimates significantly degrade alert region accuracy. In detail, the value of 3D‐source characterization depends on the user’s chosen alert threshold, tectonic regime, and faulting style. For the suite of source models we tested, the error in shaking intensity introduced by incorrect geometry could reach levels comparable to the intrinsic uncertainty in ground‐motion calculations (e.g., 0.5–1.3 modified Mercalli intensity [MMI] units for MMI 4.5) but, especially for crustal sources, was often less. For subduction interface sources, 3D representations substantially improved alert area accuracy compared to line sources, and incorrect geometry parameters were more likely to cause error in calculated shaking intensity that exceeded uncertainties.

Bulletin of the Seismological Society of America↗

USGS2021

The PLUM earthquake early warning algorithm: A retrospective case study of West Coast, USA, data

The PLUM (Propagation of Local Undamped Motion) earthquake early warning (EEW) algorithm differs from typical source-based EEW algorithms as it predicts shaking directly from observed shaking without first deriving earthquake source information (e.g., magnitude and epicenter). Here, we determine optimal PLUM event detection thresholds for U.S. West Coast earthquakes using two data sets: 558 M3.5+ earthquakes (California, Oregon, Washington; 2012–2017) and the ShakeAlert test suite of historic and problematic signals (1999–2015). PLUM computes Modified Mercalli Intensity ( I MMI ) using velocity and acceleration data, leveraging co-located sensors to avoid problematic signals. An event detection is issued when the observed I MMI exceeds a given threshold(s). We find a two-station detection method using I MMI trigger thresholds of 4.0 and 3.0 for the first and second stations, respectively, is optimal for detecting M4.5+ earthquakes. PLUM detected 79 events in the 2012–2017 data set, reporting (not including telemetry or alert dissemination) detection times on par, and sometimes faster than current EEW methods (mean 8 s; median 6 s). As expected, detection times were slower for the older 1999–2015 earthquakes ( N = 21; mean 11 s; median 6 s) when station coverage was sparser. Of the 31 PLUM detected M5+ events (10 2012–2017; 21 1999–2015), theoretically 20 (∼65%) could provide timely warnings. PLUM issued no false detections and avoided issuing detections for all calibration/anomalous signals, regional and teleseismic events. We conclude PLUM can successfully identify I MMI 4+ shaking from local earthquakes and could complement and enhance EEW in the U.S.

British Columbia, California, Oregon, Washington↗

USGS2020

Shaking is almost always a surprise: The earthquakes that produce significant ground motion

Although small earthquakes are expected to produce weak shaking, ground motion is highly variable and there are outlier earthquakes that generate more shaking than expected—sometimes significantly more. We explore datasets of M 0.5–8.3 earthquakes to determine the relative impact of frequent, smaller-magnitude earthquakes that rarely produce strong ground motion, to rare, large earthquakes that always cause strong shaking. We find that the natural variability of ground motion, combined with the Gutenberg–Richter magnitude–frequency relationship, ensures that most occurrences of any ground motion come from earthquakes of smaller magnitude than expected, often > 2 magnitude units smaller. This holds even for very strong shaking ( > 20% g ), suggesting that M < 7 earthquakes could be a significant source of damage.

Seismological Research Letters↗