Mid-crustal Discontinuity Research Articles

Crustal structure beneath broad-band seismic stations in the Mediterranean region

SUMMARY We have analysed receiver functions to derive simple models for crustal structure below 12 broad-band seismological stations from the MIDSEA project and 5 permanent broad-band stations in the Mediterranean region including northern Africa. To determine an accurate Moho depth we have reduced the trade-off between crustal velocities and discontinuity depth using a new grid search method, which is an extension of recently published methods to determine crustal thickness. In this method the best fitting synthetic receiver function, containing both the direct conversion and the reverberated phases, is identified on a model grid of varying Moho depth and varying Poisson's ratio. The values we found for Moho depth range from around 20 km for intra-oceanic islands and extended continental margins to near 45 km in regions where the Eurasian and African continents have collided. More detailed waveform modelling shows that all receiver functions can be well fit using a 2- or 3-layer model containing a sedimentary layer and/or a mid-crustal discontinuity. On comparing our results with Moho maps inferred from interpolated reflection and refraction data, we find that for some regions the agreement between our receiver function analysis and existing Moho maps is very good, while for other regions our observations deviate from the interpolated map values and extend beyond the geographic bounds of these maps.

Attenuation and source parameters of earthquakes in the Cascadia region

Abstract Digital ground-motion data recorded by the Western Canada Telemetered Network (WCTN) are used to examine the attenuation and source parameters of earthquakes in the Cascadia region of southwestern British Columbia and northwestern Washington. The data base is comprised of over 1000 vertical-component Fourier spectra, from earthquakes of magnitude 3 to 7 at distances from 10 to 500 km. Regression analyses determine the shape and coefficients of the regional attenuation curve and source spectra for 68 earthquakes. Seismic moment and stress drop are inferred from the amplitudes of the source spectra. Shallow (h < 10 km) earthquakes have an attenuation curve with a complex shape, exhibiting significant flattening (no apparent geometric spreading) in the distance range from 75 to 230 km; this shape is probably a result of strong reflected phases from mid-crustal discontinuities and the subducting slab. Events within the subducting slab and the lower crust exhibit a simple R−1 attenuation curve. The anelastic attenuation coefficient for the region as a whole is given by Q = 380f0.39. The duration of motion for each WCTN record is determined as the value that yields the observed relationship between time-domain and spectral-domain amplitudes, according to random process theory. These durations are approximately constant within 50 km of the source, then increase with distance as 0.07R. The high-frequency ground motions from Cascadia earthquakes are relatively weak. Cascadia source spectra are characterized by an average Brune stress drop of about 30 bars. This is significantly lower than the average California stress drop of 70 to 100 bars, and dramatically lower than the average eastern stress drop of 150 to 200 bars. It is concluded that there are significant regional variations in source parameters. Hazard estimates for the Cascadia region based on California groundmotion relations may be overly conservative.

The nature of crustal boundaries: combined interpretation of wide-angle and normal-incidence seismic data

After a few seconds two-way traveltime, normal-incidence seismic reflection sections are composed mainly of assemblages of short reflections. Very rarely are seen continuous reflections that might correspond to the Moho or a mid-crustal discontinuity. The inferred continuity of these boundaries has traditionally come from refraction seismology. There is now a body of high quality, coincident wide-angle and normal-incidence seismic data that have been recorded with 50–100 m shot spacing and with high frequency sources (e.g. Mobil, Babel). The complexity and characteristics of the wide-angle arrivals seen on these data suggest that they do not originate from continuous boundaries. It is suggested that these arrivals are reflections from the same assemblage of short length reflectors that are responsible for normal-incidence reflections. Seismic velocities below the middle crust may 1. (1) change corresponding to normal-incidence reflectivity, or 2. (2) generally increase with depth with localised sills or lens structures of different velocity accounting for the observed reflections. Wide-angle arrivals that have traditionally been identified as reflections from crustal boundaries (e.g. the mid-crust and Moho) and which were considered indicative of a sharp velocity discontinuity from continuous boundaries, may instead result from a concentration of lamellae.

Through thick and thin: A new model for the Colorado Plateau from seismic refraction data from Pacific to Arizona Crustal Experiment

We have constructed a model for the crust and uppermost mantle beneath the Colorado Plateau by combining newly acquired seismic refraction/wide‐angle reflection data from the 1989 Pacific to Arizona Crustal Experiment (PACE) cross profile with data collected by Roller (1965). A combination of forward modeling methods was used to model travel times and relative amplitudes of crustal and upper mantle phases. The salient features of the model for the Colorado Plateau (hereafter referred to as the plateau) are (1) an upper crust having an average velocity of 6.1 km/s with a low vertical velocity gradient, (2) midcrustal discontinuities at depths of 30 to 38 km beneath the plateau, (3) a lower crust that varies in thickness (5 to 16 km) and has an average velocity of 6.8 km/s, (4) a transitional lower crust/upper mantle boundary, and (5) a crustal thickness of 45 ± 3 km. A comparison of our new model for the Colorado Plateau with recent models for the Basin and Range shows that the upper crusts of the two provinces are similar but that the lower crust of the plateau is thicker, particularly in the plateau's central region. These observations are consistent with the idea that the Basin and Range could have formed from crust like that of the Colorado Plateau.

The Crustal Structure From Teleseismic P-Wave Coda-I. Method

The crustal structure beneath a broad-band station has a strong influence on the recorded body wave coda. Conversely, it is evident that crustal receiver effects can be employed to obtain information about the crustal structure at the receiver site. In this paper we present a method to estimate the crustal receiver response, and we invert the response to obtain a model of the most distinct features of the crustal structure beneath the station. The crustal receiver response of a station is determined from the P-wave coda of deep teleseisnic events by deconvolution with an empirically determined source wavelet. This source wavelet is assumed to incorporate source, attenutation and instrument effects. Responses obtained from clusters of events are stacked to enhance the signal-to-noise ratio of the data, and to obtain an estimate of the variance of the data. The crustal P- and S-velocity structure at the station is determined by non-linear waveform inversion of its crustal responses. An inversion strategy was developed to search efficiently for the global minimum in the misfit between data and synthetics. The inversion appears to be most sensitive to large velocity constrasts in the velocity structure which give rise to high-amplitude reflected and converted phases. The method is illustrated by the inversion of the crustal responses of NARS station NE05. It is shown that a thin layer (1.2 km) of unconsolidated sediments and a strong mid-crustal discontinuity at 11 km depth constitute the most prominent features of the crustal structure at this station. They cause an apparent delay of the first arrival on the radial component of the seismograms and produce high amplitude SV-multiples in the P-wave coda.

Waveform modeling of strong-motion data for the Saguenay earthquake of 25 November 1988

Abstract The magnitude 6.0 Saguenay earthquake of 25 November 1988, in Quebec, Canada, was one of the largest recorded in eastern North America (ENA) during the past 50 years, and it has provided important high-quality data for estimation of earthquake hazard in ENA. Ground motions for the event recorded by both Eastern Canadian Telemetered, and Strong Motion, Networks were anomalously large and unexpectedly rich in high-frequency energy, as compared with most other events having the same mb or moment magnitudes. Displacement waveforms derived from the eastern Canadian strong-motion acceleration data have been modeled in detail in an attempt to explain these anomalies. The modeling has provided strong evidence that the surface on which the faulting occurred is relatively long and narrow, with its long axis nearly horizontal and situated immediately above an internal crustal discontinuity that lies at a depth of about 30 km; the faulting process was essentially antiplane and unilateral, with the crack starting at the northwest end of the long axis and propagating in the negative strike direction towards the southeast; and slip durations at individual points on the fault were short (of the order of 0.2 sec) compared with the time taken for the crack to propagate from its starting point to the far end of the fault (about 2.5 sec in the model). The inferred short slip durations are consistent with the inferred fault geometry, and both are consistent with “normal” static and dynamic stress drops (i.e., within a factor of 2 of 100 bars). The narrow elongate fault geometry and associated short slip durations are responsible for the enhanced high-frequency radiation (as compared with “normal” events having the same seismic moment release). The associated spectra differ greatly from that of the Brune spectral model, with the former having at least two corner frequencies, whereas the latter has only one. The unilateral fault propagation resulted in a strong azimuthal variation of waveform characteristics, with the largest amplitudes and shortest durations concentrated towards the south and southeast and the smallest amplitudes and longest durations in the opposite directions. As the majority of strong-motion instruments and closer ECTN stations were located within the former azimuthal range, the anomalous high-frequency content and large observed amplitudes are accentuated by the station distribution with respect to the directivity characteristics of the source. Another significant factor likely to have contributed to the observed peak amplitudes, at least in some instances, is an indicated superposition of direct waves from the source with reflections from the inferred mid-crustal discontinuity. Elongate source geometry and associated effects such as those described above may not be uncommon for large thrust intraplate earthquakes and may have significant design implications for critical structures such as nuclear power plants.

An axial view of a metamorphic core complex: Crustal structure of the Whipple and Chemehuevi Mountains, southeastern California

A 135‐km‐long, NW‐SE trending, seismic refraction/wide‐angle reflection profile provides a unique along‐strike view of the crustal structure of a belt of metamorphic core complexes in southeastern California: the Whipple, Chemehuevi, and Sacramento mountains metamorphic core complexes. Interpretation of the seismic data was done by two‐dimensional forward modeling of travel times and amplitudes. The final model consists of (1) a thin (< 1.5 km) veneer of upper plate and fractured lower plate rocks (velocities of 1.5–5.3 km s−1) overlying a fairly homogeneous basement with velocities of 6.0 km s−1; (2) a localized, high‐velocity (6.4 km s−1) body, situated directly beneath the Whipple Mountains; (3) a 6.3–6.4 km s−1 middle crust that is thickest beneath the core complexes; (4) a 6.65±0.15 km s−1 lower crust; (5) crustal thickness of 27 km with a deeper crustal root (3 km) beneath the Whipple Mountains metamorphic core complex; and (6) a Pn velocity of 8.0±0.10 km s−1. The crustal structure that underlies the belt of metamorphic core complexes provides new insights into the processes that control extension in the deep crust. Upper crustal velocities are higher beneath the Whipple Mountains (where velocities increase to 6.4 km s−1 at ∼5 km depth) than beneath the Chemehuevi and Sacramento mountains. In addition, midcrustal discontinuities rise 2–5 km beneath the Whipple complex compared to the other complexes. These observations support greater uplift and a slightly deeper midcrustal origin for the rocks now exposed in the core of the Whipple Mountains compared to rocks in the Chemehuevi and Sacramento mountains. Despite the enhanced uplift and extension in the Whipple Mountains, the crust is thicker here (30 km) than anywhere else along the Colorado River extensional corridor. This may be in part a relic of compressional and magmatic thickening during the Mesozoic. However, we suggest that inflation of the crust during Tertiary extension was the dominant mechanism. Both mantle‐derived magmatism and lateral ductile inflow in the crust are proposed.

Seismic structure of the continental crust based on rock velocity measurements from the Kapuskasing Uplift

The Michipicoten greenstone belt (MGB), Wawa gneiss terrane (WGT), and Kapuskasing structural zone (KSZ) in Ontario, Canada, are regarded as a 25‐km‐thick partial cross section through the crust of the Superior Province. We measured compressional wave velocities (Vp) at confining pressures up to 600 MPa for representative rock samples from these terranes. Vp for MGB is variable (6.1–7 km/s) and depends on composition and anisotropy. Compressional wave velocities for low anisotropy quartzofeldspathic gneisses and intrusive rocks from the WGT and the KSZ show little variation over a wide range of silica content but mafic gneisses and anorthositic rocks show a wide velocity range over a narrower silica range, a trend corroborated by data for other high‐grade rocks. A model of the continental crust based on the laboratory data shows three megalayers with differing seismic characteristics. The shallowest levels (MGB) are seismically heterogeneous with high average velocities (6.6 km/s at 100 MPa) because of the dominance of metabasalts. The balance of the upper crust (WGT) is seismically homogeneous as it is dominated by a variety of isotropic quartzofeldspathic rocks with average velocities of 6.35 km/s at 100 MPa and 6.6 km/s at 600 MPa. The uppermost lower crust (KSZ) is layered, seismically heterogeneous and transversely isotropic. The average velocity at 600 MPa is 6.84 km/s. The boundary between WGT and KSZ is marked by a velocity increase of between 0.2 and 0.3 km/s, an increase similar to that commonly associated with midcrustal discontinuities (i.e., Conrad) in shield areas. Here, this boundary corresponds to the amphibolite‐granulite facies transition and to a change from a tonalitic middle crust to a lithologically heterogeneous uppermost lower crust. Refraction results show an anomalous high‐velocity zone in the upper crust under the KSZ that, on the basis of laboratory velocity data, can be correlated with KSZ lithologies. These data indicate that KSZ rocks can be traced to the west to depths as great as 20 km. Levels deeper than 20 km, not exposed at the surface in the KSZ, are characterized by velocities greater than 7.0 km/s and may be dominated by mafic and anorthositic lithologies or by more garnet‐rich rocks. The velocity model of the Superior Province crust based on geological reconstruction and laboratory velocity data is in general agreement with the velocity structure to depths of about 25 km determined for the crust below the MGB from refraction experiments. The velocity structure of this model is also generally similar to the velocity structure of shield areas.

Geochemistry across an exposed section of Archaean crust at Vredefort, South Africa: with implications for mid-crustal discontinuities

The central region of the Vredefort structure consists of a semi-circular multi-layered sequence of crystalline rocks which are nearly vertical in attitude, and which increase in metamorphic grade towards the core of the structure. Together with the overlying Precambrian strata, this sequence provides a cross-section through almost the entire crustal section of the Kaapvaal craton (36 km). The upper part of the Vredefort crystalline crust consists of a 3.0-Ga sequence of differentiated felsic rocks in amphibolite facies. The lower crust consists of a complex and heterogeneous (both chemically and isotopically) high-grade metamorphic terrain of charnockites, granulites (mafic and felsic) and supracrustal rocks. The upper crust is separated from the lower crust by the Vredefort discontinuity, a brittle-ductile shear zone characterised by a high concentration of pseudotachylite and brecciated rock. Petrographic, chemical and isotopic evidence suggest that the upper and lower crust have undergone very different styles of evolution. This is indicative of different geological environments prior to their present juxtaposition. We speculate that the upper and the lower parts of the Vredefort crystalline crust were juxtaposed during intracratonic thrusting ∼2.8 Ga ago, and that the entire 36-km section of crust, including the Vredefort discontinuity, were rotated and uplifted into their present vertical position during the 2.0-Ga Vredefort catastrophe.

Modeling ground-motion attenuation in eastern North America

In this study, we have utilized our understanding of the physics of elastic wave propagation to simulate ground motions within about 200 km from earthquakes in eastern North America. Detailed crustal velocity structure models were developed for sources located in the Grenville and Appalachian provinces. The former consists of a single gradient in the crust, while the latter contains a mid-crustal discontinuity and a deep Moho. Semi-empirical synthetic seismograms were computed by convolving the generalized ray response of the structures with an empirical source function obtained from a recording close to the source. Peak ground motions derived from the synthetic seismograms adequately interpolate the sparse empirical data, while providing information on the wave propagation phenomena that define ground-motion attenuation. They suggest that peak ground-motion amplitudes are controlled by (in order of increasing range) the upgoing, direct S-wave, the diving S-wave, post-critical reflections from the mid-crustal or Moho discontinuities, and finally, L g.

Crustal studies in the La Malbaie Region, Quebec

The results of three seismological investigations of the crust in the seismically active La Malbaie Region are reported. First arrivals from a reversed reflection–refraction profile within the Charlevoix structure indicate a uniform Pg velocity of 6.08 ± 0.04 km/s. Secondary arrivals forming discontinuous, sometimes arcuate segments on the record sections are interpreted as subcritical reflections from structural contortions related to impact of the Charlevoix meteorite. A coherent event at 14 s may signal reflection from the M-discontinuity near 45 km depth.Analysis of travel times from calibration shots recorded by a network of stations spanning the St. Lawrence River in terms of time–distance profiles, travel time residuals, and numerical models provides excellent support for a model of the Precambrian–Paleozoic contact striking along the north shore and dipping to the southeast about 20° beneath the wedge of 5.5 km/s sediments.Unreversed profiles obtained by recording timed Thetford Mines blasts across a similar network suggests a comparable deep structure beneath the north and south shores, with average crustal velocities of 6.8 and 6.7 km/s, respectively. A minor mid-crustal discontinuity is suggested, and Moho depth is estimated at 42–43 km.The 6.2 km/s upper crustal velocity found for the crater and immediately surrounding area is anomalously low compared to the 6.4–6.5 km/s reported for the eastern shield region. Structural lineations gleaned from satellite imagery suggest an elongated area of impact-related disruption much larger than previously recognized. This expanded area of crustal weakening encompasses and may partially explain the zone of continuing microseismic activity, as well as the low upper crustal velocity of the La Malbaie Region.