LAUSR

The decay of seismic amplitude is caused by a variety of physical phenomena that can be divided broadly into elastic transmission losses (including geometrical spreading, interface transmission losses and scattering attenuation) and intrinsic attenuation, where wave energy is converted into heat due to viscous friction. The so-called statistical averaging method is currently considered as the most advanced sonic wave attenuation estimation method, and there exist various implementations of this method. But the way elastic transmission losses – that mask the true intrinsic attenuation – are compensated for appears to be an issue and in some cases this correction has been overlooked. In this paper, we revisit the statistical averaging method for intrinsic attenuation estimation with particular focus on the role of elastic transmission losses. Through synthetic examples, we demonstrate the importance of compensating for elastic transmission losses even if the variation of velocity and density with depth is not notable. Our implementation of the method uses finite-difference simulations thereby providing a versatile and accurate way to generate synthetic seismograms. We use a combination of elastic and viscoelastic finite-difference simulations to demonstrate the significant error without accurate compensation of the elastic transmission losses.We apply our implementation of the method to sonic waveforms acquired in an exploration well from Browse basin, Australia. The resulting intrinsic attenuation estimates are indeed indicative of gas-saturated zones identified from petrophysical analysis in which viscous friction are thought to be of importance.