A classic assumption in geophysics is that shear wave polarization and splitting occurs due to stress-aligned structure, previously considered to be stress-aligned microcracks. This structure is now more often considered to be a desirable ‘open’ set of subvertical conducting fractures that are also assumed to be parallel or sub-parallel to the maximum horizontal stress. Geomechanics modeling unfortunately demonstrates that unless fractures are rather rough and wall strength rather high, or that there is overpressure, there are likely to be only very small hydraulic apertures at several kilometers depth. Deep-well measurements demonstrate that fractures that are under differential shear stress are more likely to be water conducting, and those that are principally under normal stress are less likely to be water conducting. In this paper, alternative interpretations of shear-wave polarization directions are examined, including the contribution of two, maybe unequal joint sets, intersected by the major stress, having different stiffnesses and dynamic compliances, and possibly with pre-peak non-linear shear strength and dilation contributions to their enhanced permeability. Shearing induced by reservoir production and compaction is also considered, both as a source of permeability maintenance, and as a potential source of temporal rotation of seismic anisotropy, as recently recorded in 4D seismic at the Ekofisk and Valhall reservoirs in the North Sea. The shear stresses, or the mobilized frictional strength assumed to be acting on sheared joints or minor faults in deep well analyses is very high, such as μ of 0.9, and the possibility of an error, due to application
The use of polarized shear waves, for indicating the presence of aligned and perhaps fluidconducting structure at depths in petroleum reservoirs has been a topic of interest for at least 20 years. The classic assumption has been that the aligned structure that causes frequencydependent shear-wave anisotropy is usually parallel or sub-parallel to the major stress.
In fact it has been shown in an extensive review of the literature [1], that deviation between the assumed major stress and apparent structure, or deviation between the axes of anisotropy and the assumed major stress, may each occur, each being more likely when one is no longer close to the surface. The reasons might be that more than one set of unequal fractures could be contributing to the anisotropy, and that if these are bisected by the major stress, permeability at depth could also be more easily explained, due to the shear stress actually causing slight, but desirable, dilation. The need for this alternative explanation is due to the difficulty of explaining ‘open’ fractures at depth: joints or fractures are likely to be held ‘closed’ by a minimum effective normal stress of ten's of MPa. However, mineralbridging, or joints with rough surfaces in hard rock are possible alternative explanations of ‘open’ features at depth.
