Elastic moduli evolution and accompanying stress changes with increasing crack damage: implications for stress changes around fault zones and volcanoes during deformation

P>The elastic moduli of rock in areas susceptible to crack damage, such as within fault zones or volcanic edifices, can be subject to large modifications. Knowledge of how elastic moduli may vary in such situations is important for both the reliable modelling of volcano deformation and stability and for linear and non-linear elastic crack models for earthquake rupture. Furthermore, it has previously been shown that changes in elastic moduli can induce changes in the stress field surrounding faults. Here we report both uniaxial experimental measurements of changes in elastic moduli during increasing-amplitude cyclic stressing experiments on a range of different rock types (basalts, sandstones and granite) and the results of modelled stress modifications. The trend in elastic moduli evolution with increasing damage was remarkably similar for each rock type, with the exception of essentially crack-free intrusive basalt that exhibited very minor changes. In general, Young's modulus decreased by between 11 and 32 per cent and Poisson's ratio increased by between 72 and 600 per cent over the total sequence of loading cycles. These changes are attributed to an increasing level of anisotropic crack damage within the samples. Our results also show that acoustic emission (AE) output during any loading cycle only commenced when new crack damage was generated. This corresponded to the level of stress where AE ceased during the unloading portion of the previous cycle. Using the multilayer elastic model of Faulkner et al. we demonstrate that the damage-induced changes in elastic moduli also result in significant decreases in differential stress, increases in mean stress and rotation of the applied greatest principal stress relative to the orientation of the mechanical layering. The similar trend in the evolution of the elastic moduli of all the rocks tested suggests that stress modification in the damage zone of faults might take the same form, regardless of the lithology through which the fault runs. These observations are discussed in terms of their applicability to both fault zones and deformation at volcanoes.