Seismic attenuation is here defined as the energy of seismic waves that is lost due to its conversion into heat. In fluid-saturated rocks, attenuation is controlled by the properties of the pore fluids and of the rock. Particularly, it is often controlled by a property referred to as fluid mobility which is related to the ability of the rock to let fluid to flow (permeability) and of the fluid itself to easily flow (viscosity). Therefore, attenuation recovered from seismic field data can, potentially, be inverted into information about such properties of a subsurface rock. Based on empirical observations, qualitative interpretation of seismic data to infer about, for example, permeability variations has become a realistic target that, very recently, the industry has started to explore. This proposal aims to provide a solid theoretical basis for that by identifying in laboratory data, through integration with numerical simulations, the physical mechanisms that dominate seismic attenuation at subsurface stress and fluid pressure conditions. Identifying and understanding these mechanisms will help develop the mentioned qualitative procedures into quantitative interpretation tools and inversion workflows.
So far, neither laboratory nor field data for seismic attenuation in fluid-saturated rocks at subsurface conditions have been conclusively explained by theoretical models. Nevertheless, a recently proposed numerical methodology for a mechanism associated to fluid mobility and a newly proposed attenuation mechanism, together with recent advances in laboratory techniques, are all available now and can be combined with that purpose. The idea is that, while in the laboratory various physical mechanisms can simultaneously affect seismic attenuation, with numerical modeling it is possible to simulate the effect of a single mechanism. I thus propose the integration of systematic laboratory measurements planned to maximize the effects of a target mechanism, and minimize the effects of others, with numerical computations that account only for the target mechanism. The focus is on two mechanisms that can cause significant seismic attenuation, at least in relatively idealized situations:
- mesoscopic wave-induced fluid flow, controlled by fluid mobility; and
- gas dissolution and exsolution, controlled by physiochemical properties of the fluids.
The first strongly depends on spatial distributions of rock-fluid properties and, therefore, 3D numerical models that can describe the rock samples tested in the laboratory will be used for quasi-static relaxation tests based on Biot’s equations of poroelasticity. The laboratory experiments will use samples containing controlled mesoscopic heterogeneities in: (i) fluid properties (patchy saturation, homogeneous solid frame), in one series of experiments, or (ii) in the solid frame (fractures, full water saturation), in the other series. The second mechanism, newly proposed by the Applicant and collaborators, is expected to be associated to a reasonably uniform distribution of microscopic bubbles within the pore-fluid system, at least in the cases considered in this proposal. A 1D numerical solution of the related diffusion equations will be compared to the corresponding laboratory measurements on samples saturated with ~99% aqueous solution and: (i) ~1% air, in one series of experiments, or (ii) CO2, in the other series.