TY - JOUR

T1 - Slant f-k transform of multichannel seismic surface wave data

AU - Serdyukov, Aleksander S.

AU - Yablokov, Aleksander V.

AU - Duchkov, Anton A.

AU - Azarov, Anton A.

AU - Baranov, Valery D.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We have addressed the problem of estimating surface-wave phase velocities through the spectral processing of seismic data. This is the key step of the well-known near-surface seismic exploration method, called multichannel analysis of surface waves. To increase the accuracy and ensure the unambiguity of the selection of dispersion curves, we have developed a new version of the frequency-wavenumber (f-k) transform based on the S-transform. We obtain the frequency-time representation of seismic data. We analyze the obtained S-transform frequency-time representation in a slant-stacking manner but use a spatial Fourier transform instead of amplitude stacking. Finally, we build the f-k image by analyzing the spatial spectra for different steering values of the surface-wave group velocities. The time localization of the surface-wave packet at each frequency increases the signal-to-noise ratio because of an exclusion of noise in other time steps (which does not fall in the effective width of the corresponding wavelet). The new f-k transform, i.e., the slant f-k (SFK) transform, renders a better spectral analysis than the conventional f-k transform and yields more accurate phase-velocity estimation, which is critical for the surface-wave analysis. The advantages of the SFK transform have been confirmed by synthetic- and field-data processing.

AB - We have addressed the problem of estimating surface-wave phase velocities through the spectral processing of seismic data. This is the key step of the well-known near-surface seismic exploration method, called multichannel analysis of surface waves. To increase the accuracy and ensure the unambiguity of the selection of dispersion curves, we have developed a new version of the frequency-wavenumber (f-k) transform based on the S-transform. We obtain the frequency-time representation of seismic data. We analyze the obtained S-transform frequency-time representation in a slant-stacking manner but use a spatial Fourier transform instead of amplitude stacking. Finally, we build the f-k image by analyzing the spatial spectra for different steering values of the surface-wave group velocities. The time localization of the surface-wave packet at each frequency increases the signal-to-noise ratio because of an exclusion of noise in other time steps (which does not fall in the effective width of the corresponding wavelet). The new f-k transform, i.e., the slant f-k (SFK) transform, renders a better spectral analysis than the conventional f-k transform and yields more accurate phase-velocity estimation, which is critical for the surface-wave analysis. The advantages of the SFK transform have been confirmed by synthetic- and field-data processing.

KW - S-TRANSFORM

KW - STATIC CORRECTIONS

KW - DISPERSIVE ENERGY

UR - http://www.scopus.com/inward/record.url?scp=85058646106&partnerID=8YFLogxK

U2 - 10.1190/geo2018-0430.1

DO - 10.1190/geo2018-0430.1

M3 - Article

AN - SCOPUS:85058646106

VL - 84

SP - A19-A24

JO - Geophysics

JF - Geophysics

SN - 0016-8033

IS - 1

ER -