Low‐frequency transfer of seismic energy by superficial soil deposits and soft rocks

Abstract

Recordings of recent strong earthquakes obtained on alluvial sites show that the maximum horizontal accelerations tend towards a limit of about 0.45 to 0.509, associated with large displacements. By contrast, vertical accelerations do not appear to be subject to such a limit (1.79 for the 1979 Imperial Valley earthquake). Theoretical linear elasticity models, when applied to superficial layers of low strength, seem to be inadequate for the prediction of near-field ground motions in alluvial deposits. A good approximation for the horizontal component of certain Imperial Valley records was, however, obtained through a non-linear approach, using local soil properties together with a reasonable hypothesis for motion at the base of the superficial layers in question that included large values of acceleration for high-frequency shear waves.

Attenuation Characteristics of Peak Ground Motions in the 1995 Hyogo-ken Nanbu Earthquake.

“…Namely, the horizontal component of seismic wave propagating as shear wave was deamplified in soil layers under liquefaction, while high-frequency vertical component propagating as compressional wave was amplified because of a large contrast in P-wave velocity between dry and water-saturated soft-soil layers. The larger peak vertical acceleration than horizontal acceleration at Array 6 from the 1979 Imperial Valley event was also explained by this non-linear behavior (Mohammadioun and Pecker, 1984). Their explanation is proved by the vertical array records at Port Island.…”

“…In this study, the relation between peak vertical and horizontal amplitudes of the observed records are examined and discussed in association with the rupture process and local ground conditions. A larger peak vertical acceleration than horizontal acceleration was observed at Array 6 in the 1979 Imperial Valley earthquake and explained by the non-linear behavior (Mohammadioun and Pecker, 1984). A clear non-linear behavior has also been found in the vertical array records at Port Island from this event, and vertical peak acceleration at the surface was larger than horizontal acceleration (Aguirre and Irikura, 1995).…”

Untitled

“…Generally, the effects of soil type on spectral shape of vertical motions were reported to be small [Mohammadioun and Pecker, 1984;Aguirre and Irikura, 1995;Ambraseys and Simpson, 1996;Abrahamson and Silva, 1997;Elnashai and Papazoglou, 1997;Ambraseys and Douglas, 2000;Campbell and Bozorgnia, 2003]. Soil type effect on spectral shape was examined in this study based on the large employed data set.…”

Untitled

“…Such a nonlinear ground response analysis can be conveniently carried out by an equivalent linear analysis; the algorithm PROSHAKE uses this approach, in which the shear modulus and damping values are adjusted to represent the appropriate strain level. However, it is important to note that the equivalent linear solutions at high frequencies are valid only up to certain amplitudes of the input motion: Joyner and Chen [1975], and Mohammadioun and Pecker [1984] showed significant differences between the nonlinear and equivalent linear solutions at PGA = 0.7 g and 1 g, respectively. At the same time, as noted in Mohammadioun and Pecker [1984], peak accelerations and fundamental frequencies can be reliably obtained.…”