Introduction Seismic wave attenuation (absorption or intrinsic attenuation) is underutilized by the petroleum industry mainly because of the difficulty of measuring it with sufficient accuracy using seismic reflection methods. However, the recent trend toward time-lapse 3D seismic monitoring of reservoirs means that absolute attenuation measurements are no longer so important. Instead, temporal changes in seismic amplitude and wavelet frequency content could hold the key to interpreting changes in reservoir fluid properties in response to certain production strategies. Seismic monitoring of heavy-oil reservoirs would be particularly appropriate because of the known link between seismic attenuation and pore fluid viscosity (Batzle et al., 2006a). Seismic wave attenuation in porous rocks is thought to be dominated by viscous fluid flow mechanisms such as those described by the unified Biot and Squirt (BISQ) model (Dvorkin and Nur, 1993; Dvorkin et al., 1994). By definition, the magnitude of attenuation is related to the viscosity of the pore fluid and the frequency of the elastic wave, as well as to other pore fluid parameters (density, bulk modulus) and pore geometry parameters (porosity, permeability, tortuosity, squirt flow length). This theoretical link between attenuation and pore fluid viscosity could be exploited if changes in heavy-oil viscosity caused by thermal stimulation, typically in the range 1–1000 cP ( 1 cP = 10−3 Pa·s ), give rise to sufficiently large changes in elastic wave attenuation for detection and monitoring by surface seismics. Also, an accurate model of frequency/viscosity-dependent attenuation as a function of lithology could help tie in sonic well logs at frequencies of 10–20 kHz to surface seismics at less than 200 Hz for the benefit of reservoir characterization.