Sarah Morton

Revisiting levees in southern Texas using Love-wave multichannel analysis of surface waves with the high-resolution linear Radon transform

AbstractShear-wave velocities were estimated at a levee site by inverting Love waves using the multichannel analysis of surface waves (MASW) method augmented with the high-resolution linear Radon transform (HRLRT). The selected site was one of five levee sites in southern Texas chosen for the evaluation of several seismic data-analysis techniques readily available in 2004. The methods included P- and S-wave refraction tomography, Rayleigh- and Love-wave surface-wave analysis using MASW, and P- and S-wave cross-levee tomography. The results from the 2004 analysis revealed that although the P-wave methods provided reasonable and stable results, the S-wave methods produced surprisingly inconsistent shear-wave velocity VS estimates and trends compared with previous studies and borehole investigations. In addition, the Rayleigh-wave MASW method was nearly useless within the levee due to the sparsity of high frequencies in fundamental-mode surface waves and complexities associated with inverting higher modes. T...

DISPERSION-CURVE IMAGING CONSIDERATIONS WHEN USING MULTICHANNEL ANALYSIS OF SURFACE WAVE (MASW) METHOD

The multichannel analysis of surface wave (MASW) method can be an efficient tool for mapping the near-surface shear-wave velocity (Vs). Data acquisition, dispersion-curve imaging and estimations, inversion, and 2D visualization are distinct MASW components. Dispersion-curves can be estimated on images that can be obtained by various transforms including converting seismic data from the time-space domain (i.e., t-x domain) into frequency–wave-number (i.e., f-k domain) by applying the Fourier transform to both time and space, Phase-velocity–frequency domain (e.g., Cf-f or f-v domain), slowness–frequency domain (i.e, p-f domain), phase-velocity–wavelength domain, etc. It has been our observation that while the mathematical link between such transforms is well known, the relationship between the corresponding images can be visually clarified for better comprehension. In this work we show the visual relationship between the f-k and the Cf-f domain images. We also demonstrate the visual effects of using fewer geophones (i.e., data along the space axis) with different spread sizes on the phase-velocity–frequency dispersion-curve imaging using surface-wave data with dominant single- and multi-mode surface-wave forms of expressions. These examples could help better understand and make more efficient use of the MASW data acquisition, analysis, and interpretation of final results.

A modified F-K filter for removing the effects of higher-mode dispersion patterns from surface wave data

Multichannel analysis of surface waves (MASW) is widely accepted for quickly and efficiently estimating shear-wave velocity structure in a variety of environments (Xia et al., 1999; Park et al., 1999). In areas with vertical heterogeneities or shallow bedrock (less than 5 m), dispersion curves can be difficult to interpret due to higher mode contamination. By picking a dominant higher-mode trend and misinterpreting it and inverting it as a fundamental mode, the resulting velocity profile can give inconsistent or unrealistic results. Such misinterpretation can happen because some of the dispersion modes interfere or masked by this higher energy making it difficult to differentiate them. To account for this problem, a modified frequency-wavenumber (FK) filter was applied to surface-wave data to lessen or remove the impact of observed higher-mode energy. This dispersion-curve based-FK filter inputs velocity-frequency information from a picked dispersion curve tapered to define a curved cut zone, rather than the conventional pie-shape. This filter is applied to dispersion information in the FK domain through a series of Fast Fourier Transforms (FFT) resulting in dispersion patterns with more prominent fundamental energy that were more representative of the geologic structure. This technique has been successful on datasets collected in the Caribbean and the Western United States.