Stations In Southern California Research Articles

Regional/Single Station Zenith Tropospheric Delay Combination Prediction Model Based on Radial Basis Function Neural Network and Improved Long Short-Term Memory

Atmospheric water vapor is an essential source of information that predicts global climate change, rainfall, and disaster-natured weather. It is also a vital source of error for Earth observation systems, such as the global navigation satellite system (GNSS). The Zenith Tropospheric Delay (ZTD) plays a crucial role in applications, such as atmospheric water vapor inversion and GNSS precision positioning. ZTD has specific temporal and spatial variation characteristics. Real-time ZTD modeling is widely used in modern society. The conventional back propagation (BP) neural network model has issues, such as local, optimal, and long short-term memory (LSTM) model needs, which help by relying on long historical data. A regional/single station ZTD combination prediction model with high precision, efficiency, and suitability for online modeling was proposed. The model, called K-RBF, is based on the machine learning algorithms of radial basis function (RBF) neural network, assisted by the K-means cluster algorithm (K-RBF) and LSTM of real-time parameter updating (R-LSTM). An online updating mechanism is adopted to improve the modeling efficiency of the traditional LSTM. Taking the ZTD data (5 min sampling interval) of 13 international GNSS service stations in southern California in the United States for 90 consecutive days, K-RBF, R-LSTM, and K-RBF were used for regions, single stations, and a combination of ZTD prediction models regarding research, respectively. Real-time/near real-time prediction results show that the root-mean-square error (RMSE), mean absolute error (MAE), coefficient of determination (R2), and training time consumption (TTC) of the K-RBF model with 13 station data are 8.35 mm, 6.89 mm, 0.61, and 4.78 s, respectively. The accuracy and efficiency of the K-RBF model are improved compared with those of the conventional BP model. The RMSE, MAE, R2, and TTC of the R-LSTM model with WHC1 station data are 6.74 mm, 5.92 mm, 0.98, and 0.18 s, which improved by 67.43%, 66.42%, 63.33%, and 97.70% compared with those of the LSTM model. The comparison experiments of different historical observation data in 24 groups show that the real-time update model has strong applicability and accuracy for the time prediction of small sample data. The RMSE and MAE of K-RBF with 13 station data are 4.37 mm and 3.64 mm, which improved by 47.70% and 47.20% compared to K-RBF and by 28.48% and 31.29% compared to R-LSTM, respectively. The changes in the temporospatial features of ZTD are considered, as well, in the combination model.

Dynamic modeling and real-time management of a system of EV fast-charging stations

Demand for electric vehicles (EVs), and thus EV charging, has steadily increased over the last decade. Studies suggest that fast-charging facilities are crucial for the EV market. However, there is limited fast-charging infrastructure in most parts of the world to support EV travel, especially long-distance trips. The goal of this study is to develop a stochastic dynamic simulation modeling framework of a regional system of EV fast-charging stations for real-time management and strategic planning (i.e., capacity allocation) purposes. To model EV user behavior, specifically fast-charging station choices, the framework incorporates a multinomial logit station choice model that considers station charging prices, expected wait times, and detour distances. To capture the dynamics of supply and demand at each EV fast-charging station, the framework incorporates a multi-server queueing model in the simulation. The study assumes that multiple fast-charging stations are managed by a single entity (public or private) and that the demand for these stations are interrelated (through the station choice model). To manage the system of stations, this study proposes and tests dynamic demand-responsive price adjustment (DDRPA) schemes based on station queue lengths. The study applies the modeling framework to a system of EV fast-charging stations in Southern California. The computational results indicate that DDRPA strategies are an effective mechanism to balance charging demand across fast-charging stations. Specifically, compared to the no DDRPA scheme case, the quadratic DDRPA scheme reduces average wait time by 26%, increases charging station revenue (and user costs) by 5.8%, while, most importantly, increasing social welfare by 2.7% in the base scenario. Moreover, the study also illustrates that the modeling framework can evaluate the allocation of EV fast-charging station capacity, to identify stations that require additional chargers and areas that would benefit from additional fast-charging stations.

Monitoring Human Activity at a Very Local Scale with Ground-Motion Records: The Early Stage of COVID-19 Pandemic in California, U.S.A., New York City, U.S.A., and Mexicali, Mexico

Abstract In this article, we analyze the change in anthropogenic seismic noise level within a frequency range of 4–14 Hz, through a survey of seismic stations in California, United States, New York City, United States, and Mexicali, Baja California, Mexico from early December 2019 to late April 2020. Our analysis shows that some stations recorded a drop in anthropogenic seismic noise during the COVID-19 pandemic, and the timing of the anthropogenic noise decrease typically correlates with the timing of a strict curtailment of personal and economic activity issued by the local government. In other locations, the drop in the anthropogenic seismic noise appears not to follow the lockdown timing perfectly. During our analysis, we observed that many stations did not record a drop during the early stage of COVID-19 pandemic. Of the 19 stations of the Southern California Seismic Network that were surveyed, we found that only five show a similar extent of drop in anthropogenic seismic noise comparable to the Christmas holiday break in 2019. This suggests that the human activity that caused seismic noise did not significantly reduce during the COVID-19 pandemic near most surveyed stations in southern California. A further analysis implies that the primary seismic noise source in southern California might be traffic, and the continuation of industrial traffic, such as cargo transportation, during the COVID-19 pandemic may be the reason why many stations did not record a noise drop. Our results show that the anthropogenic seismic noise recorded by seismic stations is capable of indicating human activity, and that this metric is, particularly, powerful in measuring how localized communities initially responded to the COVID-19 pandemic.

Assessing the performance of GPS tomography at retrieving water vapour fields during landfalling atmospheric rivers over southern California

AbstractAccurate monitoring of water vapour evolution is essential to gain a better understanding of the atmospheric river movement, as well as predicting the severity and life cycle of heavy precipitation. The use of hundreds of ground Global Positioning System (GPS) stations in southern California provided a good opportunity to monitor the water vapour changes associated with landfalling atmospheric rivers. Based on a tomographic technique, the 15 min‐resolution water vapour density fields of southern California were reconstructed for February 2019. The assessments by radiosonde and the European Centre for Medium‐Range Weather Forecasts' (ECMWF) Re‐Analysis Interim (ERA‐Interim) database reanalysis show that tomography can retrieve the density profiles with an overall accuracy of about 1.1 g·m–3. In addition, both assessments indicate that the tomographic solutions of altitudes > 6 km are not reliable as relative root mean square error can reach up to 500% in the uppermost layer. Nevertheless, tomographic integrated water vapour is consistent with those derived from radiosonde and ERA‐Interim because their discrepancies are basically < 3 mm. The tomographic density fields were then applied to monitor water vapour evolution associated with two atmospheric river landfalls that occurred on February 2 and February 13, 2019. Results show that both atmospheric rivers made landfall from the northwest and passed over southern California in a southeasterly direction. The series of density profiles reveals that the high‐elevation mountains can block the atmospheric river movement, resulting in the concentration of water vapour at the foot of the mountains.

Automated Estimation and Tools to Extract Positions, Velocities, Breaks, and Seasonal Terms From Daily GNSS Measurements: Illuminating Nonlinear Salton Trough Deformation

AbstractThis paper describes the methods used to estimate positions, velocities, breaks, and seasonal terms from daily Global Navigation Satellite System (GNSS) measurements. Break detection and outlier removal have been automated so that decades of daily measurements from thousands of stations can be processed in a few hours. New measurements are added, and parameters are updated every week. Model parameters allow separation of interseismic, annual, coseismic, and postseismic signals. Tools available through GeoGateway (http://geo-gateway.org) allow rapid visualization and analysis of these terms for results that can be subsetted in time or space. Results show highly variable and nonlinear motion for GPS stations in southern California. The variable motion is related to seasonal motions, distributed tectonic motion, earthquakes, and postseismic motions that can continue for years. In some areas results suggest that additional processes are responsible for the observed motions. In general, following earthquakes, stations return to their long‐term motions after 2–3 years, though some exceptions occur. The use of the tools shows nonlinear motion in the Salton Trough of southern California related to the 2010 M7.2 El Mayor‐Cucapah earthquake, 2012 Brawley earthquake swarm, and a creep event on the Superstition Hills fault in 2017.

Isolating and Suppressing the Spurious Non‐Diffuse Contributions to Ambient Seismic Field Correlations

AbstractCross correlation of fully diffuse wavefields averaged over time should converge to the Green's function; however, the ambient seismic field in the real Earth is not fully diffuse, which interferes with that convergence. We apply blind signal separation to reduce the effect of spurious non‐diffuse components on the cross‐correlation tensor of the ambient seismic field. We describe the diffuse component as having uncorrelated neighboring frequencies and equal intensity at all azimuths and an independent (i.e., statistically uncorrelated) non‐diffuse component arising from a spatially isolated point source for which neighboring frequencies are correlated. Under the assumption of linear independence of the spurious non‐diffuse wave outside the stationary phase zone and the constructive interference of noise waves within that zone, we can suppress the spurious non‐diffuse component from the noise interferometry. Our numerical simulations show good separation of one spurious non‐diffuse noise source component for either non‐diffuse Rayleigh or Love waves. We apply this separation to the Rayleigh wave component of the Green's function for 136 cross‐correlation pairs from 17 stations in Southern California. We perform beamforming over different frequency bands for the cross correlations before and after the separation and find that the reconstructed Rayleigh waves are more coherent. We also estimate the bias in Rayleigh wave phase velocity for each receiver pair due to the spurious non‐diffuse contribution.

Comprehensive Analysis of the Effects of Common Mode Error on the Position Time Series of a Regional GPS Network

This study uses the daily horizontal and vertical Global Positioning System (GPS) position time series of 41 regionally distributed GPS stations in Southern California covering the period of January 2011 to January 2017 to extract the common mode error (CME) in east-north-up (ENU) components with the principal component analysis (PCA) method. We assume the first component with common spatial responses is CME. Firstly, we focus on the effects of CME on the root mean square (RMS) of the GPS time series. The results indicate that after being filtered, CMEs can reduce the average RMS values by 42.9, 39.1 and 28.8% for ENU components, respectively. Secondly, we investigate the changes in noise type before and after spatio-temporal filtration. The outcome reveals that noise model of the GPS position time series can be mainly described by power law noise (PL) plus white noise (WH). After CME correction, 32, 12 and 13% of the stations’ noise types are altered for ENU components, respectively. Thirdly, we confirm that stations whose noise properties can be best described by PL + WH, PL + WH and FL + WH or RW (Random walk noise) + FL + WH show no obvious differences in velocity and velocity uncertainty. Stations whose noise properties are described by RW + FL + WH, PL + WH and RW + FL + WH show significant differences in velocity and velocity uncertainty. Therefore, for all GPS stations, excluding the ones whose noise properties are described by the RW + FL + WH model, we estimate the noise parameters and velocity before and after spatio-temporal filtration by fitting a PL + WH model. According to our findings, the PCA method does not appear to be efficient for horizontal components. The CME is superimposed on the stochastic process realization for the vertical components. The impact of CME on vertical velocity estimation cannot be ignored, and eliminating CME enhances the credibility of the determined velocity field. Finally, we discuss the correlation between CME and mass loadings. The results of the Pearson correlation analysis and wavelet coherence analysis indicate that CME cannot be interpreted by mass loadings in a regional GPS network and that it may be related to GPS technique errors.

Effect of removing the common mode errors on linear regression analysis of noise amplitudes in position time series of a regional GPS network & a case study of GPS stations in Southern California

The analysis of the correlations between the noise in different components of GPS stations has positive significance to those trying to obtain more accurate uncertainty of velocity with respect to station motion. Previous research into noise in GPS position time series focused mainly on single component evaluation, which affects the acquisition of precise station positions, the velocity field, and its uncertainty. In this study, before and after removing the common-mode error (CME), we performed one-dimensional linear regression analysis of the noise amplitude vectors in different components of 126 GPS stations with a combination of white noise, flicker noise, and random walking noise in Southern California. The results show that, on the one hand, there are above-moderate degrees of correlation between the white noise amplitude vectors in all components of the stations before and after removal of the CME, while the correlations between flicker noise amplitude vectors in horizontal and vertical components are enhanced from un-correlated to moderately correlated by removing the CME. On the other hand, the significance tests show that, all of the obtained linear regression equations, which represent a unique function of the noise amplitude in any two components, are of practical value after removing the CME. According to the noise amplitude estimates in two components and the linear regression equations, more accurate noise amplitudes can be acquired in the two components.

Seismogeodesy Using GPS and Low‐Cost MEMS Accelerometers: Perspectives for Earthquake Early Warning and Rapid Response

Abstract The seismogeodetic method computes accurate displacement and velocity waveforms by optimally extracting high‐frequency information from strong‐motion accelerometers and low‐frequency information from collocated Global Positioning System (GPS) instruments. These broadband observations retain the permanent (static) displacement, are immune to clipping and magnitude saturation for large earthquakes, and are sensitive enough to record P ‐wave arrivals. These characteristics make seismogeodesy suitable for real‐time applications such as earthquake early warning. The Scripps Institution of Oceanography (SIO) has developed an inexpensive microelectromechanical systems (MEMS) accelerometer package to upgrade established GPS stations. We compare the performance of our MEMS accelerometer with an observatory‐grade accelerometer using an experiment at the University of California San Diego Large High‐Performance Outdoor Shake Table. We show that the two types of accelerometers agree in frequency ranges of seismological and engineering interest and produce equivalent seismogeodetic estimates of displacement and velocity. To date, 27 SIO MEMS packages have been installed at GPS monitoring stations in southern California and the San Francisco Bay area and have recorded four earthquakes ( M 4.2, M 4.1, and two of M 4.0). The P ‐wave arrivals are distinguishable in the seismogeodetic observations at distances of up to ∼25 km away but not in the GPS‐only displacements. There is no significant permanent deformation for these small events. This study demonstrates the lower limit of detectability and that seismogeodetic waveforms can also be a reliable early confirmation that an event is not large or hazardous. It also raises the possibility of rapid magnitude estimation through scaling relationships.

Anomalously low amplitude of S waves produced by the 3D structures in the lower mantle

Direct S and Sdiff phases with anomalously low amplitudes are recorded for the earthquakes in Papua New Guinea by seismographs in northern America. According to the prediction by a standard 1D model, the amplitudes are the lowest at stations in southern California, at a distance and azimuth of around 95° and 55°, respectively, from the earthquake. The amplitude anomaly is more prominent at frequencies higher than 0.03Hz. We checked and ruled out the possibility of the anomalies appearing because of the errors in the focal mechanism used in the reference synthetic waveform calculations. The observed anomaly distribution changes drastically with a relatively small shift in the location of the earthquake. The observations indicate that the amplitude reduction is likely due to the 3D shear velocity (Vs) structure, which deflects the wave energy away from the original ray paths. Moreover, some previous studies suggested that some of the S and Sdiff phases in our dataset are followed by a prominent postcursor and show a large travel time delay, which was explained by placing a large ultra-low velocity zone (ULVZ) located on the core-mantle boundary southwest of Hawaii. In this study, we evaluated the extent of amplitude anomalies that can be explained by the lower mantle structures in the existing models, including the previously proposed ULVZ. In addition, we modified and tested some models and searched for the possible causes of low amplitudes. Full 3D synthetic waveforms were calculated and compared with the observations. Our results show that while the existing models explain the trends of the observed amplitude anomalies, the size of such anomalies remain under-predicted especially at large distances. Adding a low velocity zone, which is spatially larger and has less Vs reduction than ULVZ, on the southwest side of ULVZ, contributes to explain the low amplitudes observed at distances larger than 100° from the earthquake. The newly proposed low velocity zone mostly overlaps with the northern part of the Pacific large low shear velocity province (LLSVP) revealed in tomographic models. Although the very low amplitudes observed at a distance of about 95° remain unexplained, our results indicate that the boundary of the Pacific LLSVP is sharp, and the amplitude of S waves at these large distances is lowered by strong vertical and/or lateral deflection at the boundary toward the interior of the low velocity province.

Multimark: an R package for analysis of capture-recapture data consisting of multiple "noninvasive" marks.

I describe an open-source R package, multimark, for estimation of survival and abundance from capture-mark-recapture data consisting of multiple "noninvasive" marks. Noninvasive marks include natural pelt or skin patterns, scars, and genetic markers that enable individual identification in lieu of physical capture. multimark provides a means for combining and jointly analyzing encounter histories from multiple noninvasive sources that otherwise cannot be reliably matched (e.g., left- and right-sided photographs of bilaterally asymmetrical individuals). The package is currently capable of fitting open population Cormack-Jolly-Seber (CJS) and closed population abundance models with up to two mark types using Bayesian Markov chain Monte Carlo (MCMC) methods. multimark can also be used for Bayesian analyses of conventional capture-recapture data consisting of a single-mark type. Some package features include (1) general model specification using formulas already familiar to most R users, (2) ability to include temporal, behavioral, age, cohort, and individual heterogeneity effects in detection and survival probabilities, (3) improved MCMC algorithm that is computationally faster and more efficient than previously proposed methods, (4) Bayesian multimodel inference using reversible jump MCMC, and (5) data simulation capabilities for power analyses and assessing model performance. I demonstrate use of multimark using left- and right-sided encounter histories for bobcats (Lynx rufus) collected from remote single-camera stations in southern California. In this example, there is evidence of a behavioral effect (i.e., trap "happy" response) that is otherwise indiscernible using conventional single-sided analyses. The package will be most useful to ecologists seeking stronger inferences by combining different sources of mark-recapture data that are difficult (or impossible) to reliably reconcile, particularly with the sparse datasets typical of rare or elusive species for which noninvasive sampling techniques are most commonly employed. Addressing deficiencies in currently available software, multimark also provides a user-friendly interface for performing Bayesian multimodel inference using capture-recapture data consisting of a single conventional mark or multiple noninvasive marks.

Accuracy enhancement of GPS time series using principal component analysis and block spatial filtering

This paper focuses on performance analysis and accuracy enhancement of long-term position time series of a regional network of GPS stations with two near sub-blocks, one block of 8 stations in Cascadia region and another block of 14 stations in Southern California. We have analyzed the seasonal variations of the 22 IGS site positions between 2004 and 2011. The Green’s function is used to calculate the station-site displacements induced by the environmental loading due to atmospheric pressure, soil moisture, snow depth and nontidal ocean. The analysis has revealed that these loading factors can result in position shift of centimeter level, the displacement time series exhibit a periodic pattern, which can explain about 12.70–21.78% of the seasonal amplitude on vertical GPS time series, and the loading effect is significantly different among the two nearby geographical regions. After the loading effect is corrected, the principal component analysis (PCA)-based block spatial filtering is proposed to filter out the remaining common mode error (CME) of the GPS time series. The results show that the PCA-based block spatial filtering can extract the CME more accurately and effectively than the conventional overall filtering method, reducing more of the uncertainty. With the loading correction and block spatial filtering, about 68.34–73.20% of the vertical GPS seasonal power can be separated and removed, improving the reliability of the GPS time series and hence enabling better deformation analysis and higher precision geodetic applications.

Ground motion prediction of realistic earthquake sources using the ambient seismic field

Predicting accurate ground motion is critical for earthquake hazard analysis, particularly in situations where sedimentary basins trap and amplify seismic waves. We exploit the information carried by the ambient seismic field to extract surface‐wave Green's functions between seismic stations and to predict long‐period ground motion from earthquakes. To do so, we modify the surface impulse response to correct for the source depth and for the double‐couple focal mechanism. These corrections are derived under the assumption that material properties in the immediate vicinity of the source depend only on depth. Using this local 1‐D assumption, we solve the surface‐wave eigenproblem and compute the fundamental‐mode displacement eigenfunctions to express the surface‐wave excitation at the source. We validate this technique, which we call the virtual earthquake approach, by comparing computed seismograms with earthquake waveforms from four moderate earthquakes that occur near broadband stations in southern California. The depth and mechanism corrections show clear improvements of the predicted ground motion relative to the surface impulse response.

Seismic velocity structures in the southern California plate-boundary environment from double-difference tomography

SUMMARY We present tomographic images of crustal structures in the southern California plate-boundary area, with a focus on the San Jacinto fault zone (SJFZ), based on double-difference inversions of earthquake arrival times. Absolute arrival times of 247 472 P and 105 448 S wave phase picks for 5493 earthquakes recorded at 139 stations in southern California are used. Starting with a layered 1-D model, and continuing in later iterations with various updated initial models, we invert the data for Vp and Vs in a 270 km long, 105 km wide and 35 km deep volume around the SJFZ using a spatially variable grid with higher density around the SJFZ. The examined volume stretches from Cajon Pass to the northernmost Imperial Fault Zone and includes portions of the southern San Andreas Fault (SAF), the Elsinore Fault and the Brawley Seismic Zone in the Salton Trough. Because differential traveltimes used in the double-difference inversions are most sensitive to near-source structures, we obtain high resolution around the earthquake sources. After 30 iterations we improve the average traveltime misfit by a factor of 16. Though ray coverage is limited at shallow depths, we obtain detailed images of seismic velocities from 3 to 20 km throughout much of the study area. Our final velocity results show zones of low-velocity and anomalous Vp/Vs ratios associated with various fault strands and sedimentary basins, along with clear velocity contrasts across the SJFZ and the southern SAF. The velocity reductions in fault zone regions are generally highest in geometrically complex areas (up to 30–50 per cent in the top few kilometres), are higher for Vs than for Vp, and follow a flower-type pattern with depth. In the central section of the SJFZ, from the San Jacinto valley to the trifurcation area, the northeast side of the fault has generally higher seismic velocities than the southwest block. The obtained contrasts of Vp are more persistent and higher (up to 20 per cent) than the contrasts of Vs (up to 15 per cent), although the differences may stem (at least partially) from the higher resolution of Vp images. In the SJFZ sections to the northwest and southeast, there are patches with reversed velocities contrasts especially in the shallow crust near the San Jacinto Valley and other basins. Along the Banning fault there is no clear velocity contrast. For the southern SAF, the northeast side has generally lower seismic velocities in the seismogenic zone with patches of contrast reversals in the shallow crust. In the Brawley Seismic Zone, the northeast side has somewhat lower velocities in the ∼20 km section near the southern SAF and higher velocities farther to the southwest. The imaged features have important implications for various aspects of earthquake and crustal dynamics in the region.

Real‐time inversion of GPS data for finite fault modeling and rapid hazard assessment

Responses to recent great earthquakes and ensuing tsunamis in Sumatra, Chile, and Japan, with the resulting loss of life and damage to infrastructure demonstrate that our ability to ascertain the full extent of slip of catastrophic earthquakes and their tsunamigenic potential in the first minutes after the initiation of rupture is problematic. Regional GPS networks such as those in western North America and Japan are complementary to seismic networks by being able to directly measure displacements close to the source during large earthquakes in real time. We report on rapid modeling of two large earthquakes, the 2003 Mw 8.3 Tokachi‐oki earthquake 100 km offshore Hokkaido Island using 356 GEONET stations and the 2010 Mw 7.2 El Mayor‐Cucapah earthquake in northern Baja California using 95 CRTN stations in southern California about 75 km northwest of the epicenter. Working in a simulated real‐time mode, we invert for finite fault slip in a homogeneous elastic half‐space using Green's functions obtained from Okada's formulation. We compare two approaches: the first starts with a catalog of pre‐defined faults, while the second uses a rapid centroid moment tensor solution to provide an initial estimate of the ruptured fault plane. In either case, we are able to characterize both earthquakes in less than two minutes, reducing the time necessary to obtain finite fault slip and moment magnitude for medium and greater earthquakes compared to traditional methods by an order of magnitude.

High-resolution imaging of the deep anisotropic structure of the San Andreas Fault system beneath southern California

SUMMARY Geological and geodetic records show that motion is distributed over a broad and complex system of faults in southern California, with many large earthquakes occurring off the main San Andreas fault. This unusually complex plate boundary geometry is imposed by the big bend of the San Andreas Fault and the dynamics of the underlying lithosphere, but lithosphere rheology and distribution of applied forces are still poorly constrained. Geodynamic modelling is thus very uncertain, and distribution of lithospheric deformations highly controversial. Here, exploiting the property that deformation orients olivine crystals, we show that deep lithospheric deformations can be mapped directly with a new imaging approach of seismic anisotropy, with a resolution fine enough for detailed geodynamic interpretation. This method relies entirely on finite-frequency effects in the splitting of SKS waves. We build an extensive data set of more than 3400 SKS splitting measurements from the analysis of SKS waves recorded by all the broad-band stations in southern California. Resolution tests demonstrate that, using this data set, we are able to resolve anisotropic structures smaller than the size of the first Fresnel zone of SKS waves. The good vertical resolution in our tomographic images results from the short inter-station spacing in southern California. In our 3-D anisotropy model, we find no evidence for localized lithospheric shear deformation beneath the San Andreas Fault. Instead, the lithospheric plate boundary is localized along a broad shear zone beneath the East California Shear Zone. Therefore, surface and deep deformation patterns are poorly correlated and most likely decoupled. Active N–S convergence in the Transverse Ranges results in strongly coherent E–W alignment of olivine fast axes in the shallow lithosphere. At the top of the asthenosphere, the observed low level of anisotropy suggests weak lithospheric basal tractions and thus a strong decoupling between the lithosphere and the asthenosphere.

Pervasive lower-crustal seismic anisotropy in Southern California: Evidence for underplated schists and active tectonics

Understanding lower-crustal deformational processes and the related features that can be imaged by seismic waves is an important goal in active tectonics. We demonstrate that teleseismic receiver functions calculated for broadband seismic stations in Southern California reveal a signature of pervasive seismic anisotropy in the lower crust. The large amplitudes and small move-out of the diagnostic converted phases, as well as the broad similarity of data patterns from widely separated stations, support an origin primarily from a basal crustal layer of hexagonal anisotropy with a dipping symmetry axis. We conducted neighborhood algorithm searches for depth and thickness of the anisotropic layer and the trend and plunge of the anisotropy symmetry (slow) axis for 38 stations. The searches produced a wide range of results, but a dominant SW-NE trend of the symmetry axis emerged. When the results are divided into crustal blocks and restored to their pre–36 Ma locations, the regional-scale SW-NE trend becomes more consistent, although a small subset of the results can be attributed to NW-SE shearing related to San Andreas transform motion. We interpret this dominant trend as a fossilized fabric within schists, created from top-to-the-SW sense of shear that existed along the length of coastal California during pretransform, early Tertiary subduction or from shear that occurred during subsequent extrusion. Comparison of receiver-function common conversion point stacks to seismic models from the active Los Angeles Regional Seismic Experiment shows a strong correlation in the location of anisotropic layers with “bright” reflectors, further affirming these results.

Three-dimensional wavefield imaging of data from the USArray: New constraints on the geometry of the Farallon slab

This paper presents preliminary results from direct imaging of P to S conversion data from the Earthscope Transportable Array (TA). Input data are receiver function estimates from the Earthscope Automated Receiver Function Survey (EARS). The waveforms from EARS were first stacked to produce composite events from 60 different source regions defined by a radial grid with a center in the western United States (U.S.). The composite waveforms were imaged with a three-dimensional, prestack migration method using a plane wave decomposition and an inverse generalized Radon transform algorithm to yield estimates of radial and transverse component scattering potential from each composite event. Data deficiencies limited this processing to 50 of the composite events. The migrated results from these 50 events were stacked to produce a final three-dimensional image volume examined in this paper through three-dimensional visualization techniques. The crust-mantle boundary, the 410 km, and 660 km discontinuities are imaged. An erratic negative conversion at the depth that has been suggested as the seismic lithosphere-asthenosphere boundary is observed in these results, but I avoid interpreting this feature as it may be biased by interference with first-order, free-surface multiples from the Moho. The most significant new feature imaged by this technique is a series of east-dipping conversions directly above the 410 km discontinuity. This feature is imaged in the region immediately above the transition zone across the entire western U.S., although it becomes partly obscured in Arizona by interference with what I interpret as sediment reverberations that contaminate some stations in southern California and the Rio Grande Rift. Overlaying these results with published tomography models suggests that this feature is linked to the Juan de Fuca–Farallon slab. I follow this hypothesis to build a three-dimensional model of the top of the Farallon slab that honors the new data and results from published tomography models. The model clarifies that these dipping features are consistent with a bending plate model of the slab. Using a simple kinematic model of the geometry of the slab window linked to the widening of the San Andreas system, I argue that the same dipping features seen under Nevada are consistent with those seen to the north, suggesting that there is a continuous, relatively smooth slab-related boundary layer under most of the western U.S.

Rapid full-wave centroid moment tensor (CMT) inversion in a three-dimensional earth structure model for earthquakes in Southern California

SUMMARY A central problem of seismology is the inversion of regional waveform data for models of earthquake sources. In regions such as Southern California, preliminary 3-D earth structure models are already available, and efficient numerical methods have been developed for 3-D anelastic wave-propagation simulations. We describe an automated procedure that utilizes these capabilities to derive centroid moment tensors (CMTs). The procedure relies on the use of receiver-side Green’s tensors (RGTs), which comprise the spatial-temporal displacements produced by the three orthogonal unit impulsive point forces acting at the receivers. We have constructed a RGT database for 219 broad-band stations in Southern California using a tomographically improved version of the 3-D SCEC Community Velocity Model Version 4.0 (CVM4) and a staggered-grid finite-difference code. Finite-difference synthetic seismograms for any earthquake in our modelling volume can be simply calculated by extracting a small, source-centred volume from the RGT database and applying the reciprocity principle. The partial derivatives needed for the CMT inversion can be generated in the same way. We have developed an automated algorithm that combines a grid-search for suitable focal mechanisms and hypocentre locations with a Gauss–Newton optimization that further refines the grid-search results. Using this algorithm, we have determined CMT solutions for 165 small to medium-sized earthquakes in Southern California. Preliminary comparison with the CMT solutions provided by the Southern California Seismic Network (SCSN) shows that our solutions generally provide better fit to the observed waveforms. When applied to a large number of earthquakes, our algorithm may provide a more robust CMT catalogue for earthquakes in Southern California.

A year of microseisms in southern California

Microseisms are due to continuous harmonic forcing by ocean waves, whose sources vary in time, frequency, and azimuth. Using frequency‐domain array beamforming, this variation is studied using one‐year of continuous seismic data from 155 stations in southern California. Detailed analysis of data delineates spatiotemporal variations of sources for the primary and secondary microseisms. Both types of microseisms are generated near the coasts but the locations of excitation are different and change with season. Often sources are multiply located but can be spread out over wider areas, especially in the case of secondary microseisms. Distant storms can also be seen occasionally in the frequency range between the primary and secondary microseisms where spectral amplitudes from nearby sources are low.