LOVE-WAVE WAVEFORM INVERSION FOR SHALLOW SHEAR-WAVE VELOCITY USING A CONJUGATE GRADIENT ALGORITHM

2014 ◽  
Author(s):  
Yudi Pan ◽  
Jianghai Xia ◽  
Lingli Gao
Keyword(s):  
Shear Wave ◽  
Wave Velocity ◽  
Shear Wave Velocity ◽  
Conjugate Gradient ◽  
Love Wave ◽  
Waveform Inversion ◽  
Conjugate Gradient Algorithm ◽  
Gradient Algorithm

scholarly journals Preconditioned Conjugate Gradient Algorithm-Based 2D Waveform Inversion for Shallow-Surface Site Characterization

2021 ◽  
Vol 2021 ◽  
pp. 1-16
Author(s):  
Jianbo Guan ◽  
Yu Li ◽  
Guohua Liu
Keyword(s):  
Conjugate Gradient ◽  
Love Wave ◽  
Model Updating ◽  
Waveform Inversion ◽  
Site Characterization ◽  
Conjugate Gradient Algorithm ◽  
Gradient Algorithm ◽  
Explicit Calculation ◽  
Preconditioned Conjugate Gradient ◽  
Surface Site

The full-waveform inversion (FWI) of a Love wave has become a powerful tool for shallow-surface site characterization. In classic conjugate gradient algorithm- (CG) based FWI, the energy distribution of the gradient calculated with the adjoint state method does not scale with increasing depth, resulting in diminished illumination capability and insufficient model updating. The inverse Hessian matrix (HM) can be used as a preprocessing operator to balance, filter, and regularize the gradient to strengthen the model illumination capabilities at depth and improve the inversion accuracy. However, the explicit calculation of the HM is unacceptable due to its large dimension in FWI. In this paper, we present a new method for obtaining the inverse HM of the Love wave FWI by referring to HM determination in inverse scattering theory to achieve a preconditioned gradient, and the preconditioned CG (PCG) is developed. This method uses the Love wave wavefield stress components to construct a pseudo-HM to avoid the huge calculation cost. It can effectively alleviate the influence of nonuniform coverage from source to receiver, including double scattering, transmission, and geometric diffusion, thus improving the inversion result. The superiority of the proposed algorithm is verified with two synthetic tests. The inversion results indicate that the PCG significantly improves the imaging accuracy of deep media, accelerates the convergence rate, and has strong antinoise ability, which can be attributed to the use of the pseudo-HM.

Elastic SH- and Love-wave Full-Waveform Inversion for shallow shear wave velocity with a preconditioned technique

2020 ◽  
Vol 173 ◽  
pp. 103947
Author(s):  
Yingwei Yan ◽  
Zhejiang Wang ◽  
Jing Li ◽  
Nan Huai ◽  
Yuntao Liang ◽  
...  
Keyword(s):  
Shear Wave ◽  
Wave Velocity ◽  
Shear Wave Velocity ◽  
Love Wave ◽  
Waveform Inversion ◽  
Full Waveform Inversion ◽  
Full Waveform

Love-wave waveform inversion in time domain for shallow shear-wave velocity

Geophysics ◽  
2016 ◽  
Vol 81 (1) ◽  
pp. R1-R14 ◽  
Author(s):  
Yudi Pan ◽  
Jianghai Xia ◽  
Yixian Xu ◽  
Lingli Gao ◽  
Zongbo Xu
Keyword(s):  
Surface Wave ◽  
Wave Velocity ◽  
Time Domain ◽  
Love Wave ◽  
Waveform Inversion ◽  
Dispersion Curves ◽  
Gradient Algorithm ◽  
S Wave ◽  
Near Surface ◽  
Earth Models

High-frequency surface-wave techniques are widely used to estimate S-wave velocity of near-surface materials. Surface-wave methods based on inversions of dispersion curves are only suitable to laterally homogeneous or smoothly laterally varying heterogeneous earth models due to the layered-model assumption during calculation of dispersion curves. Waveform inversion directly fits the waveform of observed data, and it can be applied to any kinds of earth models. We have used the Love-wave waveform inversion in the time domain to estimate near-surface S-wave velocity. We used the finite-difference method as the forward modeling method. The source effect was removed by the deconvolution technique, which made our method independent of the source wavelet. We defined the difference between the deconvolved observed and calculated waveform as the misfit function. We divided the model into different sizes of blocks depending on the resolution of the Love waves, and we updated the S-wave velocity of each block via a conjugate gradient algorithm. We used two synthetic models to test the effectiveness of our method. A real-world case verified the validity of our method.

scholarly journals Surface and shear-wave velocity modelling of the Tongariro Volcanic Centre, New Zealand, using ambient noise cross-correlation

2021 ◽  
Author(s):  
◽  
Holly Joanne Godfrey
Keyword(s):  
Rayleigh Wave ◽  
Shear Wave ◽  
Wave Velocity ◽  
Shear Wave Velocity ◽  
Love Wave ◽  
Ambient Noise ◽  
Cross Correlation ◽  
Velocity Structure ◽  
Wave Velocities ◽  
Group Velocities

<p>We use continuous seismic data from permanent and temporary, broadband and short-period stations that were operating during 2001 and 2008 to investigate the subsurface velocity structure of the Tongariro Volcanic Centre (TgVC) of New Zealand, particularly the highly active but poorly understood Ruapehu and Tongariro Volcanoes.  Stacks of cross-correlation of two concurrent ambient noise seismograms can be used to estimate the interstation Green's Function, i.e., the impulse response of the earth between the two receivers. The Green's Functions are used to retrieve the dispersion relation (frequency-dependent velocity) of surface waves at different periods, which reflects the shear-wave velocity structure in the Fresnel volume of the propagating surface waves. Several studies have used dispersion measurements from ambient noise cross-correlations to investigate the shallow subsurface shear-wave velocity structure of active volcanoes around the world. Most use vertical components to retrieve the Rayleigh waves, but it is becoming increasingly common to use the horizontal seismogram components in addition to the vertical, giving further constraints to Rayleigh-wave measurements and introducing data relating to Love waves.  We compute 1,048,968 daily cross-correlations for 955 viable station pairs across the two periods, including all nine-components of the cross-correlation tensor where possible. These daily functions are then stacked into 7458 full-stacks, of which we make group velocity dispersion measurements for 2641 RR-, RZ-, TT-, ZR- and ZZ-component stacks. Cross-correlation quality varies across the networks, with some station pairs possibly contaminated with timing errors.  We observe both the fundamental and rst higher-order modes within our database of dispersion measurements. However, correctly identifying the mode of some measurements is challenging as the range of group velocities measured reflects both presence of multiple modes and heterogeneity of the local velocity structure. We assign modes to over 1900 measurements, of which we consider 1373 to be high quality.  We invert fundamental mode Rayleigh- and Love-wave dispersion curves independently and jointly for one dimensional shear-wave velocity profiles at Ruapehu and Tongariro Volcanoes, using dispersion measurements from two individual station pairs and average dispersion curves from measurements within specifi c areas on/around the volcanoes. Our Ruapehu profiles show little velocity variation with depth, suggesting that volcanic edifice is made of material that is compacting and being hydrothermally altered with depth. At Tongariro, we observe larger increases in velocity with depth, which we interpret as different layers within Tongariro's volcanic system. Slow shear-wave velocities, on the order of 1-2 km/s, are consistent with both P-wave velocities from existing velocity pro files of areas within the TgVC, and the observations of worldwide studies of shallow volcanic systems that used ambient noise cross-correlation.  A persistent observation across the majority of our dispersion measurements is that group velocities of the fundamental mode Love-wave group velocity measurements are slower than those of fundamental mode Rayleigh-waves, particularly in the frequency range of 0.25-1 Hz. Similarly, first higher-order mode Love-wave group velocities are slower than first higher-mode Rayleigh-wave velocities. This is inconsistent with the differences between synthetic dispersion curves that were calculated using isotropic, layered velocity models appropriate for Ruapehu and Tongariro. We think the Love-Rayleigh discrepancy is due to structures such as dykes or cracks in the vertical plane having greater influence than horizontal layering on surface-wave propagation. However, several measurements where Love-wave group velocities are faster than Rayleigh-wave group velocities suggests that in some places horizontal layering is the stronger influence.  We also observe that the differences between the Love- and Rayleigh-wave dispersion curves vary with the azimuth of the interstation path across Ruapehu and Tongariro Volcanoes. Some significant differences between Rayleigh-wave velocities of measurements with different interstation orientations are also observed, as are differences between Love-wave velocities. This suggests a component of azimuthal anisotropy within the volcanic structures, which coupled with the radial anistropy makes the shear-wave velocity structures of Ruapehu and Tongariro Volcanoes anisotropic with orthorhombic symmetry. We suggest that further work to determine three-dimensional structure should include provisions for anisotropy with orthorhombic or lower symmetry.</p>

scholarly journals Integration of SH seismic reflection and Love-wave dispersion data for shear wave velocity determination over quick clays

2017 ◽  
Vol 210 (3) ◽  
pp. 1922-1931 ◽  
Author(s):  
Cesare Comina ◽  
Charlotte M. Krawczyk ◽  
Ulrich Polom ◽  
Laura Valentina Socco
Keyword(s):  
Shear Wave ◽  
Wave Velocity ◽  
Shear Wave Velocity ◽  
Seismic Reflection ◽  
Love Wave ◽  
Wave Dispersion ◽  
Dispersion Data ◽  
Love Wave Dispersion

The complementarity of H/V and dispersion curves

Geophysics ◽  
2016 ◽  
Vol 81 (6) ◽  
pp. T323-T338 ◽  
Author(s):  
Silvia Castellaro
Keyword(s):  
Surface Waves ◽  
Shear Wave ◽  
Wave Velocity ◽  
Shear Wave Velocity ◽  
Love Wave ◽  
Velocity Profiles ◽  
Dispersion Curves ◽  
Source Surface ◽  
Diagnostic Features ◽  
Single Station

Noninvasive geophysical techniques based on the dispersion of surface waves in layered media are commonly used approaches for measuring shear-wave velocity profiles of the subsoil. Acquiring surface waves is a simple task, but the interpretation of their dispersion curves poses a number of challenges. In an increasing number of cases, shear-wave velocity profiles are derived from the inversion of dispersion curves of surface waves and single-station passive horizontal-to-vertical (H/V) spectral ratios, mostly using a blind joint fit of the two sets of curves. Here we emphasize the benefits of carrying out H/V surveys prior to any array acquisition. We propose to start by collecting at least two H/V recordings at a site to verify the 1D plane-parallel soil condition, as this is essential in dispersion curve inversion/modeling. Then, we look for the diagnostic features of velocity inversions in the H/V curves: when they occur, the interpretation of dispersion curves is made difficult by mode splitting/superposition and Love wave arrays will not be effective. Then we inspect the shape of the H/V curves: flat curves acquired on rock usually imply poor dispersion curves. Large receiver spacings are recommended in the arrays and Love wave arrays will not be efficient. Flat curves on soft material sites represent gently increasing [Formula: see text] gradients and Rayleigh wave arrays should be preferred. H/V curves with high frequency peaks indicate shallow impedance contrasts: this makes Love wave arrays efficient for the soft layer characterization, but provide little information at depth. H/V curves with low frequency peaks indicate deep bedrock and their inversion can provide approximate [Formula: see text] profiles down to greater depths than from an array. Equipped with the information coming from accurate H/V observations, practitioners could make better-informed decisions about array acquisition geometries, source/surface wave types, and inversion strategies.

Sign in / Sign up

Export Citation Format

Share Document