Inversion of controlled-source seismic tomography and gravity data with the self-adaptive wavelet parametrization of velocities and interfaces

A self-adaptive automated parametrization approach is suggested for the sequential inversion of controlled-source seismic tomography and gravity data. The velocities and interfaces are parametrized by their Haar wavelet expansion coefficients. Only those coefficients that are well constrained by the data, as measured by the number of rays that cross the corresponding wavelet function support area and their angular coverage, are inverted for, others are set to zero. This approach results in a reasonable distribution of resolution throughout the model even in cases of irregular ray coverage and does overcome the trade-off between different types of model parameters. A modified sequential inversion approach is suggested to join the traveltimes and gravity anomalies inversion. An algorithm is developed that inverts for smooth velocity and density variations inside the seismic layer, the position of its bottom interface as well as for optimal values of the velocity-to-density regression coefficients. The algorithm makes use of direct (diving), reflected and head (critically refracted) wave traveltimes. The algorithm workflow is demonstrated on a synthetic data example.     

Seismic amplitude tomography for crustal attenuation beneath China

Amplitude tomography reconstructs seismic attenuation directly from recorded wave amplitudes. We have applied the tomography to amplitude data reported in the 'Annual Bulletin of Chinese Earthquakes' and interpreted the regionally varying crustal attenuation in terms of tectonics. The seismic amplitudes were originally recorded for determining the M _L and M _S magnitudes. They generally correspond to the maximum amplitudes of the horizontal components of the short-period S waves and intermediate-period Rayleigh waves. Both sets of measurements are sensitive to crustal structure. The peak amplitudes from M _L amplitudes spread spherically with significant dispersion and scattering. M _S amplitudes show cylindrical spreading with little dispersion. Average crustal Q values for attenuation at 1 Hz are 737 and 505 for M _L and M _S, respectively, with substantial regional variations. Frequency dependence in the attenuation is also indicated. Regions with the lowest attenuation (high Q values) are beneath the south China Block, Sichuan Basin, Ordos Platform, the Daxinganling and the Korea Craton. These tend to be tectonically inactive regions, which are generally dominated by intrusive and cratonic rocks in the upper crust. Regions with the highest attenuation (low Q values) are beneath Bohai Basin, Yunnan, eastern Songpan-Ganzi Terrain, margins of the Ordos platform and the Qilian Shan. These are predominantly active basins, grabens and fold belts. The continental margin also highly attenuates both S and surface waves.     

Surface wave tomography of the western United States from ambient seismic noise: Rayleigh and Love wave phase velocity maps

We present the results of Rayleigh wave and Love wave phase velocity tomography in the western United States using ambient seismic noise observed at over 250 broad-band stations from the EarthScope/USArray Transportable Array and regional networks. All available three-component time-series for the 12-month span between 2005 November 1 and 2006 October 31 have been cross-correlated to yield estimated empirical Rayleigh and Love wave Green's functions. The Love wave signals were observed with higher average signal-to-noise ratio (SNR) than Rayleigh wave signals and hence cannot be fully explained by the scattering of Rayleigh waves. Phase velocity dispersion curves for both Rayleigh and Love waves between 5 and 40 speriod were measured for each interstation path by applying frequency–time analysis. The average uncertainty and systematic bias of the measurements are estimated using a method based on analysing thousands of nearly linearly aligned station-triplets. We find that empirical Green's functions can be estimated accurately from the negative time derivative of the symmetric component ambient noise cross-correlation without explicit knowledge of the source distribution. The average traveltime uncertainty is less than 1 s at periods shorter than 24 s. We present Rayleigh and Love wave phase speed maps at periods of 8, 12, 16,and 20 s. The maps show clear correlations with major geological structures and qualitative agreement with previous results based on Rayleigh wave group speeds.     

Palaeoseismicity of the Oquirrh fault, Utah from shallow seismic tomography

The magnitude and frequency of normal-fault palaeoearthquakes are usually determined by trenching studies that ascertain the size and number of colluvial wedges along the fault. Such information can be invaluable in predicting the seismic hazard and potential for a future earthquake in that region. Digging trenches across normal faults, however, is environmentally intrusive, expensive and limited in the penetration depth. To overcome these problems we propose the use of 3-D seismic tomography as a means to identify the shapes and sizes of colluvial wedges along normal faults. As an example,2-D and 3-D seismic surveys were conducted across the Oquirrh fault, Utah with the purpose of imaging the normal-fault structure to a depth of about 10  m. Results show that the 3-D tomogram clearly delineates the fault zone and a colluvial wedge, both of which correlate extremely well with the geological cross-section interpreted from an adjacent trench. The thickness of the colluvial wedge image is used in conjunction with a seismic section to compute an estimate of a 6.8 moment magnitude earthquake for the most recent event on this fault, which is in close agreement with the 7.0 estimate based on a nearby trenching study. These tomographic results demonstrate, for the first time, that seismic imaging methods can be used in some cases to estimate unambiguously the shapes of colluvial wedges and the sizes of prehistoric earthquakes. Thus, seismic tomography has the possibility of providing cheaper, deeper and wider, but less resolved, images of fault systems than the intrusive excavation of trenches across faults.     

A practical regularization for seismic tomography

Regularization is usually necessary in solving seismic tomographic inversion problems. In general the equation system of seismic tomography is very large, often making a suitable choice of the regularization parameter difficult. In this paper, we propose an algorithm for the practical choice of the regularization parameter in linear tomographic inversion. The algorithm is based on the types of statistical assumptions most commonly used in seismic tomography. We first transfer the system of equations into a Krylov subspace by using Lanczos bidiagonalization. In the transformed subspace, the system of equations is then changed into the form of a standard damped least squares normal equation. The solution to this normal equation can be written as an explicit function of the regularization parameter, which makes the choice of the regularization computationally convenient. Two criteria for the choice of the regularization parameter are investigated with the numerical simulations. If the dimensions of the transformed space are much less than that of the original model space, the algorithm can be very computationally efficient, which is practically useful in large seismic tomography problems.     

Surface wave array tomography in SE Tibet from ambient seismic noise and two-station analysis – II. Crustal and upper-mantle structure

We determine the 3-D shear wave speed variations in the crust and upper mantle in the southeastern borderland of the Tibetan Plateau, SW China, with data from 25 temporary broad-band stations and one permanent station. Interstation Rayleigh wave (phase velocity) dispersion curves were obtained at periods from 10 to 50 s from empirical Green's function (EGF) derived from (ambient noise) interferometry and from 20 to 150 s from traditional two-station (TS) analysis. Here, we use these measurements to construct phase velocity maps (from 10 to 150 s, using the average interstation dispersion from the EGF and TS methods between 20 and 50 s) and estimate from them (with the Neighbourhood Algorithm) the 3-D wave speed variations and their uncertainty. The crust structure, parametrized in three layers, can be well resolved with a horizontal resolution about of 100 km or less. Because of the possible effect of mechanically weak layers on regional deformation, of particular interest is the existence and geometry of low (shear) velocity layers (LVLs). In some regions prominent LVLs occur in the middle crust, in others they may appear in the lower crust. In some cases the lateral transition of shear wave speed coincides with major fault zones. The spatial variation in strength and depth of crustal LVLs suggests that the 3-D geometry of weak layers is complex and that unhindered crustal flow over large regions may not occur. Consideration of such complexity may be the key to a better understanding of relative block motion and patterns of seismicity.     

Non-linear optimization for seismic traveltime tomography

This paper presents a non-linear algorithmic approach for seismic traveltime. It is based on large-scale optimization using non-linear least-squares and trust-region methods. These methods provide a natural way to stabilize algorithms based on Newton's iteration for non-linear minimization. They also correspond to an alternative (and often more efficient) view of the Levenberg-Marquardt method. Numerical experience on synthetic data and on real borehole-to-borehole problems are presented. In particular, results produced by the new algorithm are compared with those of Ivansson (1985) for the Kråkemåla experiment.