Questa frase dell'introduzione nell'articolo postato da Mucciarelli mi sembra particolarmente significativa e la frase in neretto risponde ai nostri dubbi molto direttamente:
The latter issue was investigated in detail by De Martin et al. (2010) using data
collected from a borehole station during the 2005 Fukuoka prefecture Western
offshore earthquake. They showed qualitative evidence of nonlinearity during the
mainshock, but, when inverting along the stratigraphic column, they observed that
out of the eleven layers considered, six did not show any degradation of shear
modulus and two had a level of degradation lower than the one expected from laboratory
data. For two layers there was agreement between theoretical and observed
degradation and only one layer showed more degradation than expected.
Non-linearity appears to be a more subtle phenomenon than usually modeled by
the practitioner, and literature also reports cases of hardening non linearity in noncohesive,
partially saturated soils (Bonilla et al., 2005).
