Dr. Paolo Primiero　(Schlumberger K.K.)
Waveform analyses for the Yufutsu hydraulic fracture
monitoring experiment, Hokkaido, Japan.

2007年11月28日

We conduct a waveform analysis of microseismic events
recorded during the Yufutsu hydraulic fracture
monitoring (HFM) experiment of October 2003. This
experiment poses an unusual challenge due to the
position of the receivers: a single vertical array was
placed at distance of over two kilometers from the
fractured zone above the injection point. Observed
waveforms show a strong effect of attenuation, which
reduced the high-frequency signal to below the noise level.
An average attenuation of Q=60 between the injection
zone and the receivers can explain the observed data. A
stabilized waveform inversion technique allows us to
determine source characteristics constrained by such
data for the observed microseismic events. The
constrained components of the inverted moment
tensors, characterizing the observed microseismic
events, indicate that down-dip mechanism (normal
faulting) probably dominate the induced shear
dislocation. We are able to reliably invert moment
magnitudes for events with good signal-to-noise ratios
and estimate the cumulative seismic moment (and
energy) released during the experiment. We compare
this energy with the injected energy and show that the
seismic energy released during this hydraulic fracturing
is only a fraction of the injected energy.

The inverted seismic moments are consistent with the dipping trend
of event locations, indicating normal faulting (but not
excluding a partial strike-slip faulting) in the vicinity of
the fractured well. We estimate seismic moments and
moment magnitudes including the uncertainty of these
measures. Using the magnitude-frequency relation we
infer the existence of a large number of undetected events. The
seismic moment rate released during the hydraulic
fracturing process progressively increases from
phase to phase. This may indicate that the hydraulic
fracturing encountered larger tectonically active
regions or larger tectonic obstacles to the fracture.
Furthermore, the released seismic moment is
increasingly delayed due to a memory effect similar to
the Kaiser effect. The absolute values of the moment
rate release are consistent with fracturing a very strong
material similar to hot dry rocks.