Improving attenuation predictions in heterogeneous porous media using laboratory data

Over the past decade, significant efforts have been made to acurately model the dispersion and attenuation of seismic waves propagating in heterogeneous porous media. Different analytical models such as patchy saturation, squirt flow, or double porosity as well as some numerical approaches have been proposed. All these approaches account for losses due to wave induced flow occurring at the mesoscopic scale; i.e., much smaller than the wavelength but greater than the grain size. Some models, such as squirt flow, can explain attenuation levels observed in lab experiments while others, like the double porosity model, are better at explaining the attenuation measured in the seismic band of frequencies. Numerical methods are more general and can be used in any frequency range. However to correctly use the predictive power of these modeling methods, it is crucial to have a good knowledge of the nature of the heterogeneities present within the propagating medium. Among other parameters such as the porosity or the permeability, it's important to have a good idea of the spatial distribution of the elastic properties within the propagating medium since the elastic moduli of the frame of grains determines how much the fluid pressure changes and how much mesoscopic flow occurs. These fluctuations in the elastic moduli in rocks remain largely unknown at the mesoscopic scale. To fill this gap, we developed a micro-indenter able to map the elastic moduli of rock samples with a sub millimetric resolution. Various maps of the elastic properties obtained from different rocks samples will be presented as well as the attenuation level estimated from these data.