Estimating quality factor and mean grain size of sediments from high-resolution marine seismic data

Abstract

Seismic quality factor has the potential to characterize sediment properties but seldom is used by the industry for offshore site investigations because of practical difficulties with reflection seismology ͑e.g., restricted bandwidth͒ and because of uncertainties in rock-physics models. A spectral-ratio analysis of highresolution marine seismic data can determine a quality factor to within a 95% confidence of ‫01ע‬ within the uppermost 30 m of unconsolidated marine sediments. Our spectral-ratio technique does not require assumptions on how attenuation scales with frequency. Emphasis is placed on interpretation of spectral signatures before applying an iteratively reweighted robust leastsquares regression to subdue the effects of noise and local heterogeneities when determining the quality factor of a sediment package. We combined data from boomer and chirp sources to examine attenuation over four octaves of frequency ͑0.5-8.0 kHz͒ and to demonstrate that expanding the frequency range improves the precision and accuracy of quality-factor fits. We obtain frequency-independent quality factors with 95% confidence intervals of 135 ‫;21ם͑‬ ‫͒01מ‬ and 107 ‫;6ם͑‬ ‫͒5מ‬ for silty clays with mean grain sizes of 7.7 and 6.9 phi, respectively, and 63 ‫;01ם͑‬ ‫͒7מ‬ for a modern sand deposit with mean grain size 2.5 phi, from the Solent ͑U. K.͒. Sediments with higher quality factors require more independent observations to achieve a desirable 95% confidence. We required only 45 traces over sands and 1250 traces over the lowest attenuating silty clays. By constructing an empirical model of quality factor against mean grain size from published sediment studies, the mean grain sizes of our Solent sediments can be located, and we find that quality factor can be used to distinguish between coarse grain-dominated and claydominated sediments.

Pre-stack full waveform inversion of ultra-high-frequency marine seismic reflection data

“…Malinkowski et al, 2011;Kamei & Pratt, 2013); furthermore, constraining Qs from marine seismic data would require isolating converted S-waves. In this case study, the sensitivity to Q is further limited by the large-scale high quality factor of the fine-grained, cohesive sediments that form the dipping beds (Pinson et al, 2008;Malinkowski et al, 2011), which limits the change in waveform due to attenuation. However, the large-scale, high precision, Qp model of Pinson et al (2008) accounts for the intrinsic attenuation of energy, enhancing the stability of the inversion and the reliability of the solution.…”

“…Previous geophysical data available in the area comprise acoustic quality factor Qp (Pinson et al, 2008), and long-wavelength P-wave velocity field from Migration Velocity Analysis (MVA) (Pinson, 2009), along with lithology, porosity and density log-measurements of a proximal sediment core. Seismic to log calibration identifies a 1-to-3 meters thick gravel layer, with low Qp (c. 50), high P-wave velocity and density; it overlays, separated by an erosive unconformity, an over-consolidated clay-dominated layered sediment sequence, with high Qp (c. 150) and acoustic velocity.…”

“…The latter is an unusually low value for shallow marine sediments, but is required due to the over-consolidation; tests run with higher average Poisson's ratio resulted in higher data misfit. Although we invert here for the elastic parameters, the forward model includes a long-wavelength Qp model determined using the spectral ratio method from near-offset Boomer and Chirp over the sediment sequence (Pinson et al, 2008); the Qs model is derived from Qp using, in the absence of other information, a high-Q empirical relationship between compressional and shear attenuation, i.e. Qs = 4/9Qp (Aki & Richards, 2002).…”

“…The Qp macro-model estimated using the spectral ratio method (Pinson et al, 2008) and the derived Qs were included in the forward model, in particular to account for the strong intrinsic attenuation within the top low-Qp gravel layer. Although a high-fidelity attenuation model would further improve the characterisation of the sediment column, a reliable and precise wavelengthscale Qp-model is difficult to achieve: in the high Qp limit there is little attenuation per wavelength, and in the low Qp limit the energy is unable to propagate.…”

“…Over the last few years, quantitative interpretation techniques have started to be applied also to near-surface seismic data in order to remotely derive decimetric resolution shallow sediment physical properties in terms of reflection coefficient and acoustic quality factor (Bull et al, 1998;Pinson et al, 2008;Vardy et al, 2012;Cevatoglu et al, 2015). Holland & Dettmer (2013) used the angle-dependent reflection amplitude as a function of frequency to derive physical properties layering and gradients within the shallow sediments.…”