Page 144 - Reservoir Geomechanics

127 Rock failure in compression, tension and shear

The critically stressed crust

Three independent lines of evidence indicate that intraplate continental crust is gener-
ally in a state of incipient, albeit slow, frictional failure: (i) the widespread occurrence of
seismicity induced by either reservoir impoundment (or fluid injection; Healy, Rubey
et al. 1968; Raleigh, Healy et al. 1972; Pine, Jupe et al. 1990; Zoback and Harjes
1997); (ii) earthquakes triggered by small stress changes associated with other earth-
quakes (Stein, King et al. 1992); and (iii) in situ stress measurements in deep wells and
boreholes (see the review by Townend and Zoback 2000). The in situ stress measure-
ments further demonstrate that the stress magnitudes derived from Coulomb failure
theory utilizing laboratory-derived frictional coefficients of 0.6–1.0 are consistent with
measured stress magnitudes. This is well illustrated in Figure 4.24 by stress magnitude
data collected in the KTB borehole to ∼8km depth. Measured stresses are quite high
and consistent with the frictional faulting theory with a frictional coefficient of ∼0.7
(Zoback, Apel et al. 1993; Brudy, Zoback et al. 1997). Further evidence for such a
frictional failure stress state is provided by the fact that a series of earthquakes could be
triggered at ∼9km depth in rock surrounding the KTB borehole with extremely low per-
turbations of the ambient, approximately hydrostatic pore pressure (Zoback and Harjes
1997). In Chapter 9 we evaluate stress magnitude data from a variety of sedimentary
basins around the world that illustrate how stress magnitudes are in equilibrium with
frictional strength in normal, strike-slip and reverse faulting environments.
That the state of stress in the crust is generally in a state of incipient frictional failure
might seem surprising, especially for relatively stable intraplate areas. However, a
reason for this can be easily visualized in terms of a simple cartoon as shown in Figure
4.25 (after Zoback and Townend 2001). The lithosphere as a whole (shown simply in
Figure 4.25 as three distinct layers – the brittle upper crust, the ductile lower crust and
the ductile uppermost mantle) must support plate driving forces. The figure indicates
apower-law creep law (e.g. Brace and Kohlstedt 1980) typically used to characterize
the ductile deformation of the lower crust and upper mantle. Because the applied force
to the lithosphere will result in steady-state creep in the lower crust and upper mantle,
as long as the “three-layer” lithosphere is coupled, stress will build up in the upper
brittle layer due to the creep deformation in the layers below. Stress in the upper crust
builds over time, eventually to the point of failure. The fact that intraplate earthquakes
are relatively infrequent simply means that the ductile strain rate is low in the lower
crust and upper mantle (Zoback, Townend et al. 2002). Zoback and Townend (2001)
discuss the fact that at the relatively low strain rates characterizing intraplate regions,
sufficient plate-driving force is available to maintain a “strong” brittle crust in a state
of frictional failure equilibrium.
Stress measurements in many parts of the world indicate that earth’s crust is in a
state of frictional failure equilibrium as described by equation (4.39) and coefficients of