Abstract
In 2000, a large water injection (over 23000 m3) has been conducted
in granite through a 5-km-deep borehole at Soultz-sous-Forêts,
in the Upper Rhine Graben (northeastern France). The microseismicity
induced by this hydraulic stimulation was monitored with a network
of 14 seismic stations deployed at ground surface. Some 7215 well-located
events have been used to conduct a 4-D tomography of P-wave velocities.
The method combines a double-difference tomography method with an
averaging post-processing that corrects for parameter dependence
effects. The total set of 7215 events has been divided into 14 subsets
that explore periods defined with respect to the injection scheme.
Particular attention is given to changes in injected flow rates,
periods of stationary injection conditions and post-injection periods.
Fast changes in VP velocities are identified in large rock mass volumes
precisely when the injection flow rate varies while little velocity
variation is detected during stationary injection periods. The VP
anomalies observed during stationary injection conditions are interpreted
as being caused by effective stress variations linked to fluid diffusion,
while the fast changes observed concomitantly to changes in flow
rate are considered to be caused by non-seismic motions.
Users
Please
log in to take part in the discussion (add own reviews or comments).