Abstract

Three-component records of shear waves from small earthquakes recorded within the shear-wave window at the surface and shear waves recorded below the surface in three-component VSPs almost routinely display shear-wave splitting indicating some form of effective anisotropy. This splitting appears to be the result of liquid-filled cracks throughout at least the brittle, top 10-20 km of the crust, where the cracks are aligned with respect to the direction of the principal axes of stress. Liquid-filled cracks with small aspect ratios have very little effect on P-wave propagation, but may have major effects on the polarisations of shear-waves. Fluids are a very common constituent of the crustal rock mass (Fyfe, Price and Thompson 1978). Water is contained in the pore space of sedimentary rocks when they are first deposited, and every prograde metamorphic process releases chemically-bound water into microcracks of the otherwise intact rock mass. Deep wells (Kozlovsky 1984) find water-filled fractures deep in the crust, and even upper mantle xenoliths are pervaded by fluid-filled microcracks, where the fluid is largely CO2 (Andersen, O'Reilly and Griffin 1984). We shall usually refer to such cracks in the top half of the crust as 'waterfilled' but recognise that they may be oil-filled in hydrocarbon reservoirs. The major difficulty in accepting the proposition that such cracks are common is that seismic investigations, which are the most direct way we have of evaluating the properties of rocks in situ, have previously displayed little evidence for widespread cracking. The reason most seismic investigations in the past, including the detailed surveys by the hydrocarbon industry, have not found evidence for widespread cracking is because these studies have usually been confined to P-waves, and water-filled cracks have little effect on P-wave propagation. In contrast, the behaviour of shear-wave particle displacements (polarisations) is very sensitive to crack distributions whenever the cracks display preferred orientations. This paper reports recent observations of shear waves in seismic zones by my colleagues and myself at the British Geological Survey (Crampin et al. 1980; Booth et al. 1985; Crampin and Booth 1985; Crampin et al. 1985b) and in stable continental areas by the hydrocarbon industry, that display a number of phenomena which suggest that distributions of aligned water-filled cracks are widespread in (at least) the brittle, top 10 to 20 km of the crust. Monitoring the behaviour of such extensive dilatancy anisotropy (EDA) by the analysis of shearwave polarisations has been suggested as a way of monitoring the build-up of stress before earthquakes (Crampin et al. 1984).

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call