ALBERT

Description: The seismic wavefield mainly contains reflected, refracted and direct waves but energy related to elastic scattering can also be identified at frequencies of 1 Hz and higher. The scattered, high-frequency seismic wavefield contains information on the small-scale structure of the Earth's crust, mantle and core. Due to the high thermal conductivity of mantle materials causing rapid dissipation of thermal anomalies, the Earth's small-scale structure most likely reveals details of the composition of the interior, and, is therefore essential for our understanding of the dynamics and evolution of the Earth. Using specific ray configurations we can identify scattered energy originating in the lower mantle and under certain circumstances locate its point of origin in the Earth allowing further insight into the structure of the lowermost mantle. Here we present evidence, from scattered PKP waves, for a heterogeneous structure at the core–mantle boundary (CMB) beneath southern Africa. The structure rises approximately 80 km above the CMB and is located at the eastern edge of the African LLSVP. Mining-related and tectonic seismic events in South Africa, with m b from 3.2 to 6.0 recorded at epicentral distances of 119.3° to 138.8° from Yellowknife Array (YKA) (Canada), show large amplitude precursors to PKP df arriving 3–15 s prior to the main phase. We use array processing to measure slowness and backazimuth of the scattered energy and determine the scatterer location in the deep Earth. To improve the resolution of the slowness vector at the medium aperture YKA we present a new application of the F -statistic. The high-resolution slowness and backazimuth measurements indicate scattering from a structure up to 80 km tall at the CMB with lateral dimensions of at least 1200 km by 300 km, at the edge of the African Large Low Shear Velocity Province. The forward scattering nature of the PKP probe indicates that this is velocity-type scattering resulting primarily from changes in elastic parameters. The PKP scattering data are in agreement with dynamically supported dense material related to the Large Low Shear Velocity Province.

Description: The generalized F method, developed for teleseismic signal detection at small-aperture arrays, is extended to medium-aperture arrays and regional-distance signals using a multiple-filter technique. The technique allows the continuous estimate of the instantaneous amplitude and phase of the array seismograms, while at the same time allowing time-domain beamforming, making the method applicable to situations where the transit time of the signal across an array is much greater than the signal duration. The method is tested by application to waveform data from 22 seismometer arrays of the International Monitoring System, being set up to monitor compliance with the Comprehensive Nuclear-Test-Ban Treaty. A comparison with the results of the traditional signal detection method used by the International Data Centre (IDC) shows that the F detector increases candidate first P associations with IDC Reviewed Event Bulletin events, whereas at the same time halving the overall number of detections. The F method increases the number of associations at 21 of the 22 arrays, and for all signal slownesses.

Description: The ability to confidently estimate the depths of small-to-medium sized (3.5 ≤ m b  ≤ 5.5) seismic disturbances is important both in plate tectonics studies, and when monitoring compliance with the Comprehensive Nuclear-Test-Ban Treaty (CTBT). Seismic source depths can be determined by identification of the teleseismic depth phases pP and sP , and also by modelling surface wave amplitude spectra. The radiation pattern of the teleseismic depth phase pP and fundamental mode Rayleigh amplitudes show that the effectiveness of these methods of earthquake depth estimation is dependent on the orientation of the focal mechanism and the station locations. For some focal mechanisms, the predicted amplitude of the teleseismic depth phase pP will only be large for stations in certain locations, and the Rayleigh wave spectral nulls that tightly constrain the seismic source depth when modelling surface wave amplitude spectra often only occur for a limited range of azimuths. In this study, we show that for sources where Rayleigh wave spectral nulls are not observed and the source depth cannot be constrained using the surface wave amplitude spectra, the focal mechanism obtained by modelling Rayleigh and Love wave amplitude spectra can be used to identify the locations of stations where pP should have a large amplitude and hence be easiest for an analyst to identify. The increased global coverage of seismometer stations means that there is an increased likelihood that stations exist in the locations where the predicted amplitude of pP is large. As the identified depth phases are consistent with the focal mechanism this approach allows increased confidence to be placed in the identified depth phases and hence the estimated source depth. This approach could potentially be used with other methods of focal mechanism estimation provided that the method used to estimate the focal mechanism is independent of the source depth.

Description: Seismometer arrays, first designed and still used to monitor underground nuclear tests, are generally the most sensitive seismological stations, and consequently their reports make a major contribution to earthquake bulletins. The detection of signals at arrays is therefore a foundational field of study in seismology. One of the advantages of arrays over single seismometer stations is that robust statistical criteria can be used to discriminate between real and false detections. Essentially this is because arrays allow the simultaneous estimation of signal and noise power. The F detector, first successfully applied to seismology by Blandford (1974), makes use of this characteristic of arrays. As originally formulated, F requires that the signal be strongly coherent across the array and so is found to be more useful for teleseismic signals than for regional signals. Also the F detector, as originally formulated, requires that seismic noise is uncorrelated between the seismometers in the array. Consequently, at small-aperture (〈8 km) arrays, F is often found to be no more useful than traditional detectors based on the ratio of short-term and long-term power averages (STA/LTA) on a beam. Here, the F detector is generalized to make use of a priori information about the signal and noise. This is accomplished by making use of standard least-squares inverse theory, which has been applied to seismometer arrays for over 40 yr. The generalized F detector is then applied to seismograms recorded at the small-aperture ( approximately 3 km) ARCES array in Norway. The enhanced detection capability compared to the traditional F and STA/LTA detectors is demonstrated.

Description: The Burma arc links the Himalaya to the Andaman-Sumatra trench, and comprehension of the geological features in the region is vital to understand the active tectonic processes in the area. Subduction on the Sumatra trench produced the devastating 2004 Sumatra earthquake (M (sub w) approximately 9.0), but the mechanisms accommodating relative India-Sundaland motion in Burma are still unknown. Previous seismological studies of the area use only earthquake catalog hypocenters, so structural details of subducted material remain poorly understood. Here, accurate relative hypocenters for 81 earthquakes are estimated using first arrivals picked from regional and teleseismic recordings. Depth determination is improved using measured travel-time differences between P onsets and depth phases (pP and sP). The results clearly illustrate a slab 21 degrees -25 degrees N dipping approximately 25 degrees at 40-80 km deep, approximately 40 degrees at 80-120 km deep, and approximately 60 degrees at 120-160 km deep, and its strike follows the Indo-Burman ranges. At latitudes 〉25 degrees N earthquakes become more shallow, and a diffuse picture of hypocenters with an east-west lateral discontinuity at depth is observed, indicating lateral deformation of the slab. Previous studies suggest there is a transition at approximately 90 km deep from shallower strike-slip earthquakes to deeper thrust-type events. Here, accurate depth estimates, combined with confirmation of two Global Centroid Moment Tensor solutions using body-wave modeling, shows that thrust and strike-slip earthquakes both occur deeper than 100 km in the Burma arc.

Description: Waveform data recorded at seven of the International Monitoring System (IMS) seismometer arrays are used to measure the relative times of teleseismic P signals with three methods: analyst picks, and two methods that make full use of the capabilities of seismometer arrays, the cross correlation of array beams and the average of channel-by-channel cross correlations. These times are used to estimate the relative locations of the 9 October 2006 and 25 May 2009 Democratic People's Republic of Korea announced underground nuclear tests. The 2009 test is found to have occurred about 1.8+ or -0.8 km to the west and 0.4+ or -0.6 km to the north of the 2006 test. The use of cross correlation reduces the standard deviation of the travel-time residuals from approximately 0.05 to approximately 0.01 sec, enabling the two epicenters to be statistically distinguished with high confidence. This result demonstrates the power of a small number of IMS seismometer arrays at teleseismic distances to detect and relatively locate small explosions with high precision.

Description: Anderson et al. (2003) interpret a set of unassociated seismic arrivals observed on 24 November 1993 as a line source, which they claim is consistent with the passage of a “strange quark nugget” through the Earth. In fact, these arrivals can be convincingly interpreted as an earthquake source on the Pacific-Antarctic ridge. We attempt to form an event using the arrivals listed in table 6 of Anderson et al. (2003) and find, as the authors do, that this is not possible without a large residual at station stk. We then use the bulletin of the ISC (International Seismological Centre, 2001) to investigate residuals at stk during November 1993. The results are shown in Figure 1. We find that the residuals show a clear trend with values around −9 sec on 24 November. Consequently we believe …

Description: The defining characteristic of a fundamental mode Rayleigh wave in a spherical, isotropic Earth is its elliptical polarization in the plane of propagation. Measurement of this polarization is thus critical both in identification and in determining the backazimuth of a Rayleigh wave from receiver to source and hence the association with a particular seismic disturbance. Currently, the prototype International Data Center (pIDC) and the International Data Center (IDC) measure backazimuth but do not utilize this in association because it is believed that measurement of backazimuth is unreliable. In this article, we demonstrate that an accurate measurement of backazimuth can be obtained for the large majority of Rayleigh waves reported in the bulletins of the pIDC/IDC. A reliable backazimuth measurement confirms the association of a waveform with a seismic disturbance, whereas the inclusion of large backazimuth errors in the bulletin can cast doubt on the association. Because the object of event screening is to positively identify earthquakes, it is vital that surface-wave magnitudes are only measured from positively identified and associated Rayleigh waves.

Description: Earthquakes that occur near known nuclear test sites are invaluable for evaluating the capabilities of the International Monitoring System (IMS), currently being established to monitor compliance with the Comprehensive Nuclear-Test-Ban Treaty (CTBT). On 13 March 2003 a seismic disturbance with magnitude m (sub b) (National Earthquake Information Center [NEIC]) 4.8 occurred near the Chinese nuclear test site at Lop Nor, southern Xinjiang. Here we attempt to identify this disturbance as an earthquake by using three types of seismic data: (1) teleseismic P waves recorded at IMS stations; (2) long-period surface waves recorded at a network of stations in Eurasia that simulates the proposed IMS seismic network in this region, and (3) long-period full waveform (P, S, and surface wave) modeling at station WMQ, which is approximately 250 km from the reported epicenter of the 13 March 2003 disturbance. We find that all impulsive teleseismic P-wave onsets show compressive first motion and that discrimination using the m (sub b) :M (sub s) criterion requires assumptions that cannot be justified on theoretical grounds. However, by combining the three data types, we conclude that the observations are consistent with a double-couple source with strike phi = 125+ or -10 degrees , dip delta = 40+ or -10 degrees , rake lambda = 90+ or -10 degrees , and moment M (sub 0) 5.5+ or -1X10 (super 15) N m. That said, consistency of the seismic wave field with a double couple does not immediately rule out an explosion source, because tectonic release accompanying an explosion can make it impossible to resolve the isotropic moment using long-period data. The strongest argument for an earthquake is that both the regional surface waves, and waveform modeling at WMQ, give an estimated focal depth of 6+ or -1 km. This result reinforces the importance of focal depth determination to CTBT monitoring, in particular, for shallow earthquakes with mechanisms that are close to perfect 45 degrees reverse-dip-slip, which are difficult to discriminate by using other methods. Fortunately, depth determination using surface waves is especially favorable for this type of earthquake, if a sufficient number of stations can be used. We therefore recommend that long-period data recorded at auxiliary seismic stations of the IMS be utilized by the International Data Centre to monitor compliance with the CTBT.