Tomographic and faulting studies represent two efficient ways to characterize the
behaviour of geothermal reservoirs during stimulation tests. Tomographic analysis of
induced microseismicity give two types of results: the 3D distribution of seismic
velocities within the medium and precise relocation of the microseismic events,
using the velocity model previously found. Analysis of faulting mechanisms,
especially through focal mechanisms studies, is able to give informations on the
main shearing structure and the type of movements that they support.
In the case of the stimulation of EGS-type reservoir, numerous microseismic events
are generally recorded; that allows performing reliable tomographic calculation.
Moreover, this huge quantity of data can be used to follow the temporal evolution of
the 3D distribution of the seismic velocities, which brings useful information on
the effect of the circulating water on the physical properties of the reservoir.
Using a velocity model that takes into account the temporal variation of the seismic
velocities allows getting a very precise relocation of the microseismic events.
This, combined with the analysis of the faulting mechanisms through the reservoir
gives a rather good view of the shearing processes in the reservoir.
In this study, we analyses the microseismic events induced during two tests: the
stimulation of GPK2 in 2000 and of GPK3 in 2003, both performed between 4.5 and 5 km
depth. The temporal evolution of the 3D distribution of seismic velocities during
the injections is calculated in both cases, leading to conclusions about the
variations of the properties of the reservoir. The analysis of the relocated
microseismic cloud demonstrates that some major structures, probably corresponding
to major faults, play a dominant role in the generation of seismic events, both of
small and higher magnitude.
From the focal mechanisms study, it appears that normal faulting, with a more or
less pronounced strike-slip component, represent the major regime, but quasi pure
strike-slip movements are also observed. From the recorded first-motion polarity
data, an estimate of the stress tensor has been calculated and then applied to the
nodal planes determined from focal mechanisms: orientation and dip of fractures
having sheared can thus be retrieved. Moreover, all determined focal mechanisms show
a double-couple solution, but from the analysis of the seismic moment tensor for
several 2003 events, we have been able to quantify the proportion of the non-double-
couple (NDC) component, which correspond to the proportion of opening in the
shearing process. It is interesting to observe that events in the vicinity of the
injection well GPK3 show a higher NDC component, than those far from the injection.
The combination of results given by all these methods, applied to two different
datasets gives thus very valuable information on the mechanical processes that occur
within the reservoir under stimulation conditions.