Faults In California Research Articles

Statistical tests of simple earthquake cycle models

AbstractA central goal of observing and modeling the earthquake cycle is to forecast when a particular fault may generate an earthquake: a fault late in its earthquake cycle may be more likely to generate an earthquake than a fault early in its earthquake cycle. Models that can explain geodetic observations throughout the entire earthquake cycle may be required to gain a more complete understanding of relevant physics and phenomenology. Previous efforts to develop unified earthquake models for strike‐slip faults have largely focused on explaining both preseismic and postseismic geodetic observations available across a few faults in California, Turkey, and Tibet. An alternative approach leverages the global distribution of geodetic and geologic slip rate estimates on strike‐slip faults worldwide. Here we use the Kolmogorov‐Smirnov test for similarity of distributions to infer, in a statistically rigorous manner, viscoelastic earthquake cycle models that are inconsistent with 15 sets of observations across major strike‐slip faults. We reject a large subset of two‐layer models incorporating Burgers rheologies at a significance level of α = 0.05 (those with long‐term Maxwell viscosities ηM <~ 4.0 × 1019 Pa s and ηM >~ 4.6 × 1020 Pa s) but cannot reject models on the basis of transient Kelvin viscosity ηK. Finally, we examine the implications of these results for the predicted earthquake cycle timing of the 15 faults considered and compare these predictions to the geologic and historical record.

Parametrizing Physics-Based Earthquake Simulations

Utilizing earthquake source parameter scaling relations, we formulate an extensible slip weakening friction law for quasi-static earthquake simulations. This algorithm is based on the method used to generate fault strengths for a recent earthquake simulator comparison study of the California fault system. Here we focus on the application of this algorithm in the Virtual Quake earthquake simulator. As a case study we probe the effects of the friction law’s parameters on simulated earthquake rates for the UCERF3 California fault model, and present the resulting conditional probabilities for California earthquake scenarios. The new friction model significantly extends the moment magnitude range over which simulated earthquake rates match observed rates in California, as well as substantially improving the agreement between simulated and observed scaling relations for mean slip and total rupture area.

Localized seismic deformation in the upper mantle revealed by dense seismic arrays.

Seismicity along continental transform faults is usually confined to the upper half of the crust, but the Newport-Inglewood fault (NIF), a major fault traversing the Los Angeles basin, is seismically active down to the upper mantle. We use seismic array analysis to illuminate the seismogenic root of the NIF beneath Long Beach, California, and identify seismicity in an actively deforming localized zone penetrating the lithospheric mantle. Deep earthquakes, which are spatially correlated with geochemical evidence of a fluid pathway from the mantle, as well as with a sharp vertical offset in the lithosphere-asthenosphere boundary, exhibit narrow size distribution and weak temporal clustering. We attribute these characteristics to a transition from strong to weak interaction regimes in a system of seismic asperities embedded in a ductile fault zone matrix.

Zipper junctions: A new approach to the intersections of conjugate strike-slip faults

Abstract Intersecting pairs of simultaneously active faults with opposing slip sense present geometrical and kinematic problems. Such faults rarely offset each other but usually merge into a single fault, even when they have displacements of many kilometers. The space problems involved are solved by lengthening the merged fault (zippering up the conjugate faults) or splitting it (unzippering). This process can operate in thrust, normal, and strike-slip fault settings. Examples of conjugate pairs of large-scale strike-slip faults that may have zippered up include the Garlock and San Andreas faults in California (USA), the North and East Anatolian faults (Turkey), the Karakoram and Altyn Tagh faults (Tibet), and the Tonale and Giudicarie faults (southern Alps). Intersecting conjugate ductile shear zones behave in the same way on outcrop and micro-scales. Zippering may produce complex and significant patterns of strain and rotation in the surrounding rocks, depending on the angle between the faults and the relative strength of the blocks they bound. A zippered fault will have a slip rate equal to the vector sum of the slip rates on the merging faults, unless that displacement is transferred into or out of the system by distributed strain in the surrounding rocks.

Radar Determination of Fault Slip and Location in Partially Decorrelated Images

Faced with the challenge of thousands of frames of radar interferometric images, automated feature extraction promises to spur data understanding and highlight geophysically active land regions for further study. We have developed techniques for automatically determining surface fault slip and location using deformation images from the NASA Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR), which is similar to satellitebased SAR but has more mission flexibility and higher resolution (pixels are approximately 7 m). This radar interferometry provides a highly sensitive method, clearly indicating faults slipping at levels of 10 mm or less. But interferometric images are subject to decorrelation between revisit times, creating spots of bad data in the image. Our method begins with freely available data products from the UAVSAR mission, chiefly unwrapped interferograms, coherence images, and flight metadata. The computer vision techniques we use assume no data gaps or holes; so a preliminary step detects and removes spots of bad data and fills these holes by interpolation and blurring. Detected and partially validated surface fractures from earthquake main shocks, aftershocks, and aseismicinduced slip are shown for faults in California, including El Mayor-Cucapah (M7.2, 2010), the Ocotillo aftershock (M5.7, 2010), and South Napa (M6.0, 2014). Aseismic slip is detected on the San Andreas Fault from the El Mayor-Cucapah earthquake, in regions of highly patterned partial decorrelation. Validation is performed by comparing slip estimates from two interferograms with published ground truth measurements.

Elevated time-dependent strengthening rates observed in San Andreas Fault drilling samples

The central San Andreas Fault in California is known as a creeping fault, however recent studies have shown that it may be accumulating a slip deficit and thus its seismogenic potential should be seriously considered. We conducted laboratory friction experiments measuring time-dependent frictional strengthening (healing) on fault zone and wall rock samples recovered during drilling at the San Andreas Fault Observatory at Depth (SAFOD), located near the southern edge of the creeping section and in the direct vicinity of three repeating microearthquake clusters. We find that for hold times of up to 3000 s, frictional healing follows a log-linear dependence on hold time and that the healing rate is very low for a sample of the actively shearing fault core, consistent with previous results. However, considering longer hold times up to ∼350,000 s, the healing rate accelerates such that the data for all samples are better described by a power law relation. In general, samples having a higher content of phyllosilicate minerals exhibit low log-linear healing rates, and the notably clay-rich fault zone sample also exhibits strong power-law healing when longer hold times are included. Our data suggest that weak faults, such as the creeping section of the San Andreas Fault, can accumulate interseismic shear stress more rapidly than expected from previous friction data. Using the power-law dependence of frictional healing on hold time, calculations of recurrence interval and stress drop based on our data accurately match observations of discrete creep events and repeating Mw=2 earthquakes on the San Andreas Fault.

Evolution and progressive geomorphic manifestation of surface faulting: A comparison of the Wairau and Awatere faults, South Island, New Zealand: COMMENT

Zinke et al. (2015) documented a flight of Saxton River terraces deformed by the Awatere fault, South Island, New Zealand. They concluded that the manifestation of off-fault deformation (OFD) is a cumulative effect, and therefore may not be immediately apparent on younger terraces, with implications for slip rate determinations. I argue that they did not adequately consider the detailed structural setting of elements of OFD within the Saxton River releasing bend on the Awatere fault. The surface deformation at Saxton River is probably controlled by the coincidence of terraces with structural zones within the bend, and I contest the authors’ conclusion that off-plane deformation is simply hidden in younger terraces. The influence of structural complexity on the width and geometry of OFD is well established (e.g., Milliner et al., 2015). Zinke et al.’s figure DR2 (in GSA Data Repository 2015341) shows that OFD along the Awetere fault occurs at bends or steps in the fault plane, including at Saxton River where the fault bends right relative to the main trace by 5–10°, ~50 m east of the T2-T3 riser. This gives rise to a graben (T1 tread) and two pressure ridges (T2 tread and bedrock spur) that define and bound an overall 150-m-wide releasing bend (Mason et al., 2006) (Figs. 1A and 1B). The paired pressure-ridge and graben geometry is predicted when the wall rocks on either side of a strike-slip fault approach and interact with a fault bend during an earthquake (Sibson, 1989), and has been reported on the Hope (Cowan, 1990) and Greendale (Duffy et al., 2013) faults in New Zealand, and the San Andreas fault in California (Ben-Zion et al., 2012), among others.

The Unintended Impacts of Anti-Redlining Legislation

Federal and state legislation intended to curb the practice of geographic loan discrimination or redlining may have the unintended and undesirable effect of preventing mortgage originators from using environmental characteristics as criteria in lending evaluations. Since the California fault rupture zones (special studies zones) do not contain systematic concentrations of poor, black, or elderly households, they should be targets for differential lending policies, such as mandatory earthquake insurance or structural reinforcements as conditions for mortgage loans. A clarification of the wording of the Housing Financial Discrimination Act is needed to alleviate some of the present potentially damaging effects of failing to discourage residential investment in surface fault rupture zones. is a form of geographic discrimination in lending. The term refers to the refusal to grant mortgage loans to otherwise qualified buyers for sound property in certain areas of the city [1]. Redlining was an early practice of the Home Owners Loan Corporation: maps were color coded to represent relative risks to lenders, with hazardous zones coded as red [2]. The decision that a neighborhood is a poor risk for mortgage loans has been based on several factors, including large numbers of renter-occupiers, changing racial composition, or visible signs of property deterioration [3]. This conservative lending policy has been justified on the argument that banks and saving and loans have a fiduciary responsibility to their investors to protect their savings from undue investment risk.

Bimaterial effects in an earthquake cycle model using rate‐and‐state friction

AbstractWe have developed a computational framework to study earthquake cycles in 2‐D plane strain and apply it to faults separating dissimilar material. We consider a planar, strike‐slip fault governed by rate‐and‐state friction where quasi‐dynamic events nucleate spontaneously due to remote, tectonic loading. We investigate the influence of material contrast over the course of many hundreds of years. For the parameters we consider, we find that the presence of bimaterial properties influences the earthquake nucleation site, such that rupture in the preferred direction (that is, in the direction of particle motion of the side of the fault with lower shear wave velocity) is favorable. For large values of the critical slip distance Dc, events propagating in the preferred rupture direction occur for a wide range of material contrasts. For smaller values of Dc, small events emerge, even in the monomaterial case. With material mismatch present, some of these small events propagate in the nonpreferred direction, made possible by a favorable stress distribution left on the fault from previous ruptures. These results may shed light on our understanding of rupture directivity on large strike‐slip faults (like the San Andreas Fault in California) which occasionally host events rupturing in the nonpreferred direction.

A case for historic joint rupture of the San Andreas and San Jacinto faults.

The San Andreas fault is considered to be the primary plate boundary fault in southern California and the most likely fault to produce a major earthquake. I use dynamic rupture modeling to show that the San Jacinto fault is capable of rupturing along with the San Andreas in a single earthquake, and interpret these results along with existing paleoseismic data and historic damage reports to suggest that this has likely occurred in the historic past. In particular, I find that paleoseismic data and historic observations for the ~M7.5 earthquake of 8 December 1812 are best explained by a rupture that begins on the San Jacinto fault and propagates onto the San Andreas fault. This precedent carries the implications that similar joint ruptures are possible in the future and that the San Jacinto fault plays a more significant role in seismic hazard in southern California than previously considered. My work also shows how physics-based modeling can be used for interpreting paleoseismic data sets and understanding prehistoric fault behavior.

Potential and limits of InSAR to characterize interseismic deformation independently of GPS data: Application to the southern San Andreas Fault system

AbstractThe evaluation of long‐wavelength deformation associated with interseismic strain accumulation traditionally relies on spatially sparse GPS measurements, or on high spatial‐resolution InSAR velocity fields aligned to a GPS‐based model. In this approach the InSAR contributes only short‐wavelength deformation and the two data sets are dependent, thereby challenging the evaluation of the InSAR uncertainties and the justification of atmospheric corrections. Here we present an analysis using 7 years of Envisat InSAR data to characterize interseismic deformation along the southern San Andreas Fault (SAF) and the San Jacinto Fault (SJF) in southern California, where the SAF bifurcates onto the Mission Creek (MCF) and the Banning (BF) fault strands. We outline the processing steps for using InSAR alone to characterize both the short‐ and long‐wavelength deformation, and evaluate the velocity field uncertainties with independent continuous GPS data. InSAR line‐of‐sight (LOS) and continuous GPS velocities agree within ∼1–2 mm/yr in the study area, suggesting that multiyear InSAR time series can be used to characterize interseismic deformation with a higher spatial resolution than GPS. We investigate with dislocation models the ability of this mean LOS velocity field to constrain fault slip rates and show that a single viewing geometry can help distinguish between different slip‐rate scenarios on the SAF and SJF (∼35 km apart) but multiple viewing geometries are needed to differentiate slip on the MCF and BF (<12 km apart). Our results demonstrate that interseismic models of strain accumulation used for seismic hazards assessment would benefit from the consideration of InSAR mean velocity maps.

Earthquake detection through computationally efficient similarity search.

Seismology is experiencing rapid growth in the quantity of data, which has outpaced the development of processing algorithms. Earthquake detection-identification of seismic events in continuous data-is a fundamental operation for observational seismology. We developed an efficient method to detect earthquakes using waveform similarity that overcomes the disadvantages of existing detection methods. Our method, called Fingerprint And Similarity Thresholding (FAST), can analyze a week of continuous seismic waveform data in less than 2 hours, or 140 times faster than autocorrelation. FAST adapts a data mining algorithm, originally designed to identify similar audio clips within large databases; it first creates compact "fingerprints" of waveforms by extracting key discriminative features, then groups similar fingerprints together within a database to facilitate fast, scalable search for similar fingerprint pairs, and finally generates a list of earthquake detections. FAST detected most (21 of 24) cataloged earthquakes and 68 uncataloged earthquakes in 1 week of continuous data from a station located near the Calaveras Fault in central California, achieving detection performance comparable to that of autocorrelation, with some additional false detections. FAST is expected to realize its full potential when applied to extremely long duration data sets over a distributed network of seismic stations. The widespread application of FAST has the potential to aid in the discovery of unexpected seismic signals, improve seismic monitoring, and promote a greater understanding of a variety of earthquake processes.

Dynamic Ridges and Valleys in a Strike‐Slip Environment

AbstractStrike‐slip faults have long been known for characteristic near‐fault landforms such as offset rivers and strike‐parallel valleys. In this study, we use a landscape evolution model to investigate the longer‐term, catchment‐wide landscape response to horizontal fault motion. Our results show that strike‐slip faulting induces a persistent state of disequilibrium in the modeled landscapes brought about by river lengthening along the fault alternating with abrupt shortening due to stream capture. The models also predict that, in some cases, ridges oriented perpendicular to the fault migrate laterally in conjunction with fault motion. We find that ridge migration happens when slip rate is slow enough and/or soil creep and river incision are efficient enough that the landscape can respond to the disequilibrium brought about by strike‐slip motion. Regional rock uplift relative to baselevel also plays a role, as topographic relief is required for ridge migration. In models with faster horizontal slip rates, stronger rocks, or less efficient hillslope transport, ridge mobility is limited or arrested despite the continuance of river lengthening and capture. In these cases, prominent steep, fault‐facing facets form along well‐developed fault valleys. Comparison of landscapes adjacent to fast‐slipping (>30 mm/yr) and slower‐slipping (≤1 mm/yr or less) strike‐slip faults in California, USA, reveals features that are consistent with model predictions. Our results highlight a potential suite of geomorphic signatures that can be used as indicators of horizontal crustal motion and geomorphic processes in strike‐slip settings even after river capture has diminished or erased apparent offset along the fault.

An Adaptive General Four-Component Scattering Power Decomposition With Unitary Transformation of Coherency Matrix (AG4U)

An adaptive general four-component scattering power decomposition method (AG4U) is proposed in this letter. The degree of polarization $m$ is used as a criterion for the adaptive nature of the proposed decomposition. In this method, one among the two complex special unitary transformation matrices is chosen to transform a real unitary rotated coherency matrix based on the largest value of $m$ . This transformed matrix is then utilized for the existing Yamaguchi et al. four-component decomposition scheme with an extended volume scattering model. The proposed decomposition is applied to Radarsat-2 full-polarimetic C-band data over San Francisco and Uninhabited Aerial Vehicle Synthetic Aperture Radar (UAVSAR) full-polarimetric L-band data over the Hayward Fault in California. The scattering powers estimated from the decomposition techniques of Yamaguchi et al. (Y4O), Singh et al. (G4U) , and AG4U are compared. AG4U shows appreciable improvements in the scattering powers, particularly in urban areas oriented about the radar line of sight compared with the Y4O and G4U decompositions. It also shows reduced percentage of pixels with negative powers considerably compared with the Y4O decomposition.

Reassessing the complexity of the rupture of the 2010 Jia-Shian earthquake (Mw 6.2) in Southwestern Taiwan by inverting jointly teleseismic, strong-motion and CGPS data

A new interpretation is proposed for the 2010 Jia-Shian earthquake (Mw 6.2) which occurred on March 4, 2010 in southwestern Taiwan. Among the few tens of CGPS stations which recorded the mainshock at a distance less than 50km we identified a few stations displaying horizontal displacements in disagreement with the dominant pattern of horizontal coseismic motion. Based on this observation, we find that a secondary rupture plane explains these peculiar motions and helps improving the modeling of the overall set of CGPS stations. We also model the teleseismic and strong-motion records to refine the rupture model. Finally, we obtain the slip distribution on a two-fault model by the joint inversion of all three datasets (CGPS, teleseismic, and strong-motion). The main fault plane is consistent with previous studies. It corresponds to a N311°E fault dipping 33° to the NE, with left-lateral–reverse oblique slip, likely related to the Chishan transfer fault zone (CTFZ). We find that the earthquake started with a sharp slip patch localized at the hypocenter. The secondary rupture plane is a N20°E thrust fault related to the Lungchuan anticline. Faulting on the main NW–SE plane during the Jia-Shian earthquake and stress inversions from previous studies both suggest a direction of maximum horizontal compression oriented NE–SW in the area of the earthquake, very different from the dominant NW–SE compression direction in Taiwan. We propose that such a variation can be related to a stress deviation in the immediate vicinity of the CTFZ, in a similar way as observed along the San Andreas Fault in California. On the other hand, rupture on the secondary fault plane associated with the Lungchuan anticline appears to be compatible with the dominant compression in Taiwan.

Rupture process for micro-earthquakes inferred from borehole seismic recordings

We investigate the spatial extent of rupture and variability in fault slip for micro-earthquakes by inverting seismic moment rate functions derived from empirical Green’s function deconvolution. By using waveforms from an array of borehole seismometers, we determine the spatial distributions of fault slip for M 3+ earthquakes that occurred along the Hayward fault in central California and identify a variety of slip behaviors including subevents, directivity, and high stress drop. The 2013 M w 3.2 Orinda earthquake exhibits a complex rupture process involving two subevents with northwest and up-dip directivity. The two subevents release 43 and 18 % of the total seismic moment (6.7 × 1013 N m), and their inferred peak stress drops are 18 and 8 MPa. The 2011 M w 4.0 Berkeley and 2012 M w 4.0 El Cerrito earthquakes are marked by high stress drop. The inferred peak and mean stress drops are about 100–130 and 40 MPa, respectively, which suggests that there are locally high levels of fault strength on the Hayward fault. Our finite-source modeling suggests that the radiation efficiency determined for these two earthquakes is very low (<0.1) and implies that most energy is dissipated during the earthquake rupture process.

Mesozoic gliding and Tertiary basin and range tectonics in eastern Sonora, Mexico

Complex deformational structures assigned to vertical movements, and gravitational activity, such as gliding, are well preserved in Upper Cretaceous rocks of the Arivechi region, Eastern Sonora, Mexico. These structures allow us to discriminate between synsedimentary and tectonic origins. The rocks at Arivechi have been divided into two units according to their lithology and degree of deformation, the Cañada de Tarachi (oldest) and El Potrero Grande units. The Upper Cretaceous volcano-sedimentary sequence is more than 6km thick and consists mainly of andesite, rhyolithic tuff, conglomerate sandstone, siltstone, and shale-type rocks that were deposited in the Arivechi back-arc basin. Within the Cañada de Tarachi Unit, monoliths and blocks were found; these include rocks dating from Proterozoic, Paleozoic and Mesozoic times. The Proterozoic monoliths are dominated by quartz sandstone and dolomite and the Paleozoic and Mesozoic monoliths consist of limestone, shale, limestone-shale-sandstone, conglomerate and igneous rocks. Two granitic monoliths have been dated, one at 76 Ma (U/Pb zircon date) and the other at 70 Ma (40Ar/39Ar whole rock step-heating). The sedimentary monoliths show gradational transitions from coherent beds to intense deformation along their edges and bases. The structure within the monoliths, blocks and surrounding host rocks was studied in detail. The new stratigraphic and structural data suggest that the monoliths and blocks may have originated at a positive land east of Arivechi, in western Chihuahua and eastern Sonora, Mexico, known as the Aldama Platform. The identified structures consist of folds and normal faults. Fold vergence suggests that the depocenter was located westward of the study area. A paleostress reconstruction documented the tectonic history of the eastern margin of the Arivechi back-arc basin. Stress inversion of fault slip data was achieved by an improved right-dihedra method followed by rotational optimization. The structural analysis and paleostress reconstructions showed that the Mesozoic-Cenozoic kinematic evolution was extensional and continued during Cretaceous times; it has also been identified in Miocene rocks. The rock sequence was modified by basin and range-type normal faulting, as suggested by our paleostress reconstruction. The last event is the result of the beginning of tectonic extension at the onset of the shift from a subduction regime toward a transform fault regime, i.e., from the subduction of the Fallaron plate to the San Andreas Gulf of California fault system during Miocene times. NE-SW and ENE-WSW extension was followed by E-W and WNW-ESE extension. The faulting sequence went from NE-SW to ENE-WSW and subsequently to WNW-ESE.

Spatial-temporal variation of low-frequency earthquake bursts near Parkfield, California

Tectonic tremor (TT) and low-frequency earthquakes (LFEs) have been found in the deeper crust of various tectonic environments globally in the last decade. The spatial-temporal behaviour of LFEs provides insight into deep fault zone processes. In this study, we examine recurrence times from a 12-yr catalogue of 88 LFE families with ∼730 000 LFEs in the vicinity of the Parkfield section of the San Andreas Fault (SAF) in central California. We apply an automatic burst detection algorithm to the LFE recurrence times to identify the clustering behaviour of LFEs (LFE bursts) in each family. We find that the burst behaviours in the northern and southern LFE groups differ. Generally, the northern group has longer burst duration but fewer LFEs per burst, while the southern group has shorter burst duration but more LFEs per burst. The southern group LFE bursts are generally more correlated than the northern group, suggesting more coherent deep fault slip and relatively simpler deep fault structure beneath the locked section of SAF. We also found that the 2004 Parkfield earthquake clearly increased the number of LFEs per burst and average burst duration for both the northern and the southern groups, with a relatively larger effect on the northern group. This could be due to the weakness of northern part of the fault, or the northwesterly rupture direction of the Parkfield earthquake.

Bloom

In Bloom, an Internet-based earth-art-work, minute movements of the Hayward Fault in California are detected by a seismograph, transmitted continuously via the Internet, and processed to generate an evolving field of circular blooms. The size and position of each bloom is based on real-time changes in Earth’s motion, measured as a vertical velocity continuously updated from the seismometer. The horizontal position of blooms is based on time, and their vertical position is based on magnitude of the second derivative or rate of change. Large movements create large blooms; small jitters create tiny buds. This essay presents several stills from the Bloom project, and an essay on the work and its creator, Ken Goldberg.

Introduction to the Special Issue on the 2012 Haida Gwaii and 2013 Craig Earthquakes at the Pacific-North America Plate Boundary (British Columbia and Alaska)

The 27 October 2012 M w 7.8 Haida Gwaii thrust earthquake and the 5 January 2013 M w 7.5 Craig strike‐slip earthquake are the focus of this special issue. They occurred along the transform boundary between the Pacific and North American plates (Fig. 1). The most identifiable feature of the plate boundary, the strike‐slip Queen Charlotte fault, might be viewed as typical of continent–ocean transform faults because it separates the continental crust of the North American plate from oceanic crust of the Pacific plate for most of its length. However, the current relative plate motion of about 5 cm/yr is highly oblique to the Queen Charlotte fault, causing a transpressive plate boundary in the region. Figure 1. Location map for the 2012 M w 7.8 Haida Gwaii and 2013 M w 7.5 Craig earthquakes, which occurred on the Pacific–North American plate boundary. Approximate rupture zones for the two events are shown in yellow, whereas other significant events on the plate boundary are depicted by their aftershock zones (red, 1949 Queen Charlotte fault; orange, 1970 Cape St. James; blue, 1972 Sitka). The epicentral positions are given by stars, and the focal mechanisms are plotted in a lower‐hemisphere projection. The black arrows indicate Pacific–North American relative motion. The Queen Charlotte terrace is located west of the Queen Charlotte fault and is interpreted to be a narrow accretionary prism. (Inset) The earthquakes occurred offshore the western margin of northern North America. Barbs indicate subduction of the Juan de Fuca (JDF) plate and smaller plates beneath North America at the Cascadia subduction zone (CSZ). Transform plate boundaries between continental and oceanic crust are often depicted as both simple and common. In reality, such continent–ocean transform boundaries are neither. For example, the San Andreas fault in California carries continental crustal fragments along with it on the seaward side …