On the inversion of Love- and Rayleigh-wave dispersion and implications for Earth structure and anisotropy

Summary. Various factors can make it difficult to explain observations of Love- and Rayleigh-wave dispersion with the same relatively simple isotropic model. These factors include systematic errors which might occur in determinations of observed group and phase velocities, lateral variations in structure along the path of travel, and the attempt to explain observations with a model comprised of only a small number of thick layers. The last of these factors is illustrated by an inversion of dispersion data in the central United States where shear-wave anisotropy had previously been invoked as one way to explain incompatible Love- and Rayleigh-wave velocities. It is shown that the data can be satisfied equally well by an isotropic model consisting of several thin layers.
In cases where the incompatibility of Love- and Rayleigh-wave data might be produced by intrinsic anisotropy, it is necessary to invert those data using an anisotropic theory rather than by separate isotropic inversions of Love and Rayleigh waves. Inversions of fundamental-mode data for a region of the Pacific, assuming anisotropic media in which the layers are transversely isotropic with a vertical axis of symmetry, lead to models which are highly non-unique. Even if the inversions solve only for shear velocities in the litho-sphere and asthenosphere it is not possible, without supplementary information, to ascertain the depth interval over which anisotropy occurs or to determine the thickness of the lithosphere or asthenosphere with much precision.     

Rayleigh-wave dispersion function for a transversely isotropic layered spherical earth using the Thomson-Haskell matrix method

We consider the two coupled differential equations of the two radial functions appearing in the displacement components of spheroidal oscillations for a transversely isotropic (TI) medium in spherical coordinates. Elements of the layer matrix have been explicitly written—perhaps for the first time—to extend the use of the Thomson-Haskell matrix method to the derivation of the dispersion function of Rayleigh waves in a transversely isotropic spherical layered earth. Furthermore, an earth-flattening transformation (EFT) is found and effectively used for spheroidal oscillations. The exponential function solutions obtained for each layer give the dispersion function for TI spherical media the same form as that on a flat earth. This has been achieved by assuming that the five elastic parameters involved vary as r ^p and that the density varies as r ^p-2, where p is an arbitrary constant and r is the radial distance. A numerical illustration with p = - 2 shows that, in spite of the inhomogeneity assumed within layers, the results for spherical harmonic degree n , versus time period T , obtained here for the Primary Reference Earth Model (PREM), agree well with those obtained earlier by other authors using numerical integration or variational methods. The results for isotropic media derived here are also in agreement with previous results. The effect of transverse isotropy on phase velocity for the first two modes of Rayleigh waves in the period range 20 to 240 s is calculated and discussed for continental and oceanic models.     

A statistical method for detecting spatiotemporal co-occurrence patterns

Spatiotemporal co-occurrence patterns (STCOPs) are subsets of Boolean features whose instances frequently co-occur in both space and time. The detection of STCOPs is crucial to the investigation of the spatiotemporal interactions among different features. However, prevalent STCOPs reported by available methods do not necessarily indicate the statistically significant dependence among different features, which is likely to result in highly erroneous assessments in practice. To improve the reliability of results, this paper develops a statistical method to detect STCOPs and discern their statistical significance. The proposed method detects STCOPs against the null hypothesis that the spatiotemporal distributions of different features are independent of each other. To construct the null hypothesis, suitable spatiotemporal point-process models considering spatiotemporal autocorrelation are employed to model the distributions of different features. The performance of the proposed statistical method is assessed by synthetic experiments and a case study aimed at identifying crime patterns among multiple crime types in Portland City. The experimental results demonstrate that the proposed method is more effective for detecting meaningful STCOPs than the available alternative methods.     

A simple method of representing azimuthal anisotropy on a sphere

We describe a method of expressing azimuthally anisotropic surface wave velocities on the Earth using a local and smooth spherical-spline parametrization. Anisotropy in the Earth leads to azimuthally varying Love and Rayleigh wave velocities that can be expressed as (cos 2ζ, sin 2ζ) and (cos 4ζ, sin 4ζ) perturbations to the isotropic velocities, where ζ is the direction of surface-wave propagation. The strength of the perturbations varies laterally, and a current goal of seismic tomography is the detailed global mapping of these variations. Several parametrizations have previously been used to describe azimuthally varying velocities. The representation proposed here uses spherical splines and is designed to describe smooth variations in both the strength and geometry of azimuthal anisotropy. The method builds on a simple geometrical approximation for the local azimuth of propagation expressed at the defining spline knot points. It avoids the singularities at the poles that result when azimuthal variations are parametrized using traditional scalar spherical harmonics. Compared with a generalized spherical-harmonic expansion of the tensor fields that represent 2ζ and 4ζ azimuthal variations smoothly on a sphere, the new method offers the advantages of local geographical support and simplicity of implementation.     

Bias in the estimate of seismic moment tensor by the linear inversion method

Summary. We investigate the effects of various sources of error on the estimation of the seismic moment tensor using a linear least squares inversion on surface wave complex spectra. A series of numerical experiments involving synthetic data subjected to controlled error contamination are used to demonstrate the effects. Random errors are seen to enter additively or multiplicitively into the complex spectra. We show that random additive errors due to background recording noise do not pose difficulties for recovering reliable estimates of the moment tensor. On the other hand, multiplicative errors from a variety of sources, such as focusing, multipathing, or epicentre mislocation, may lead to significant overestimation or underestimation of the tensor elements and in general cause the estimates to be less reliable.