SUMMARY Ambient noise tomography has been used to construct Rayleigh‐wave group velocity maps covering the northern (NMER), central (CMER) and southern (SMER) parts of the Main Ethiopian Rift (MER). In addition, dispersion curves, extracted from the group velocity maps, have been inverted to obtain a quasi‐3‐D model of crustal shear wave velocities. In comparison to crustal structure on the Ethiopian Plateau, we find (1) lower shear wave velocities at all crustal depths beneath the Yerer‐Tullu Wellel Volcanotectonic Lineament, (2) lower shear wave velocities throughout the MER at upper crustal depths (<10 km), (3) regions of lower shear wave velocities at mid‐ (10–20 km) crustal depths beneath the Wonji Fault Belt (WFB), in the transition between the NMER and CMER, and beneath the Silti‐Debre zeit Fault Zone (SDFZ) on the western side of the CMER, (4) an offset in the velocity pattern at mid‐crustal depths between the NMER and CMER coincident with the Boru‐Toru Structural High (BTSH) and (5) little evidence for lower shear wave velocities at mid‐ or lower‐crustal depths beneath the SMER. We attribute these findings primarily to along‐strike changes in crustal composition, melt content and thermal structure resulting from the Cenozoic to recent magmatism, and also, at upper crustal depths (<10 km), to basin structure and fill. Our findings corroborate a magmatic plumbing model for the MER that shows two major zones of magmatic activity, one beneath the WFB and the other beneath the SDFZ. The shear wave velocity patterns in our model show good correlation with the depth extent of seismicity, upper‐mantle seismic anomalies and seismic anisotropy, as would be expected if the along‐strike changes in shear wave velocity reflect the thermal and compositional structure of the crust.

[1] The crustally guided shear wave, Lg, is typically the most prominent phase of a nuclear explosion at regional distance. This Lg phase is analyzed often to discriminate a nuclear explosion from a natural earthquake. In addition, the Lg phase allows us to determine the size of the detonation. A nuclear explosion test in North Korea was conducted on 9 October 2006. The epicenter was located close to the eastern shore of the Korean Peninsula, resulting in raypaths that vary significantly according to the azimuths. In particular, rays radiated in the southern direction experience lateral variation of crustal structures at the continental margin. We examine the influence of raypaths on regional seismic phases by comparing the spectra and waveforms from different raypaths. Three natural earthquakes in North Korea are also examined to determine the raypath effect. We find that the Lg from the nuclear explosion dissipated significantly as result of energy leakage into the mantle resulting from variations in crustal thickness along the portion of the raypath traversing the western tip of the Sea of Japan (East Sea). Some of the leaked energy develops into mantle lid waves (Sn), causing a large energy increase to Sn. A similar feature is observed in the records of natural earthquakes. This feature is confirmed by seismic waveform modeling. The raypath effect also causes underestimation of magnitude. The Lg body wave magnitude, m b (Lg), is estimated to be 3.8-4.2 for records from pure continental paths and 2.6-3.4 for records from paths crossing continental margins. This result illustrates the need to consider raypath effects for the correct estimation of magnitudes of regional events, including a nuclear explosion.

No abstract

S U M M A R YWe present a seismic waveform inversion methodology based on the Gauss-Newton method from pre-stack seismic data. The inversion employs a staggered-grid finite difference solution of the 2-D elastic wave equation in the time domain, allowing accurate simulation of all possible waves in elastic media. The partial derivatives for the Gauss-Newton method are obtained from the differential equation of the wave equation in terms of model parameters. The resulting wave equation and virtual sources from the reciprocity principle allow us to apply the Gauss-Newton method to seismic waveform inversion. The partial derivative wavefields are explicitly computed by convolution of forward wavefields propagated from each source with reciprocal wavefields from each receiver. The Gauss-Newton method for seismic waveform inversion was proposed in the 1980s but has rarely been studied. Extensive computational and memory requirements have been principal difficulties which are addressed in this work. We used different sizes of grids for the inversion, temporal windowing, approximation of virtual sources, and parallelizing computations. With numerical experiments, we show that the Gauss-Newton method has significantly higher resolving power and convergence rate over the gradient method, and demonstrate potential applications to real seismic data.

The first crustal‐scale controlled source seismic refraction experiment in the southern Korean Peninsula, KCRUST2002, was carried out along a 300‐km long profile across this peninsula in December 2002. Iterative processing and modeling produced a laterally varying layered crustal velocity model. The crust is thickest (34 km) below the Okcheon fold belt in the middle of the transect and thinnest (28 km) at the eastern end where the Cretaceous Gyeongsang basin is characterized by 5 km of low velocity material that constitutes the upper crust. The P velocities in upper and lower crust range from 5.4 to 6.0 km/s and from 6.4 to 6.7 km/s, respectively. The average crustal Poisson's ratio is found to be 0.25–0.27 (Vp/Vs = 1.73−1.78) along the profile. A mid‐crustal velocity discontinuity is recognized in the northwestern part of the transect. The underlying mantle has velocities in the range of 7.9–8.1 km/s.