Anelastic damping of acoustic and seismic pulses

The basic question of the linearity or non-linearity of the seismic attenuation mechanism has not yet been conclusively answered. Alternative linear theories are compared with recent measurements of the degeneration of pulse shape in attenuating media. Only a theory due to Azimiet al., (1968) survives the test. The assumption of linearity conflicts with observations of stress-strain hysteresis loops which always have cusped (pointed) ends, but no successful hysteretic model of pulse attenuation has emerged. Observations of the independence of attenuations of superimposed waves are critical to the linearity assumption, but all superposition observations so far are defective in that they have considered the superposition of phase-uncorrelated waves, whereas only the superposition of phase-correlated waves is a true test of superposition.