Abstract:
Injection-induced seismicity poses a risk to public safety, infrastructure and
perception. The disposal of wastewater deep-underground presents an especially
large hazard in the central United States. Relations currently used in
forecasting earthquakes are based on injection volume and rate. However,
the triggering mechanism for induced seismicity is usually formulated as an
increase in
uid pressure that reduces frictional strength and destabilises a
fault. Laboratory experiments on Bentheim and Flechtingen sandstone were
performed to establish quantitative relations between acoustic emissions, the
elastic waves that accompany microscopic damage, and
uid pressure and
test the hypothesis of a linear relation between
uid pressure and seismicity.
Acoustic emissions are a micro-scale analogue for seismicity and obey empirical
relations such as Omori's and Gutenberg-Richter's laws. The acoustic
emission experiments consisted of four phases: saturation, con ning, loading
to a critically-stressed state, and undrained injection to macroscopic failure.
P-wave piezoelectric sensors actively measured wave velocity and passively
monitored acoustic emissions throughout the procedure. The injection pressure
was increased at the boundary in a series of steps to allow time for
uid
pressure to equilibrate and the acoustic emission rate to decay. It was concluded
that the number of acoustic emissions produced per unit increase of
uid pressure followed a power-law function of the total acoustic emissions
produced during the injection, once a critical pressure was exceeded. The observed
behaviour could be explained by a damage model where former weakening
by microscopic crack formation may have provided defects for subsequent
acoustic emissions.
