Four years of experience with a permanent seismic monitoring array at the Ketzin CO2 storage pilot site

CO2 was injected into a saline aquifer near the town of Ketzin in Germany from July 2008 to August 2013. To monitor CO2-
migration close to the injection well, TNO installed a fixed 2D seismic array of 120 meters length in 2009, with 3- component (3-
C) geophones at the surface, 4-component receivers at 50 meters depth and a central vertical array of 4-component receivers. The
test-bed is used both for the recording of high-quality active time-lapse seismic data as well as for continuous passive seismic
data recording. Here we focus on the latest results obtained from active source experiments, micro-seismic data analysis and
ambient noise seismic interferometry (ANSI) of passive seismic data.
Data from two experiments with a source driven by linear motors (permanent) and an impact hammer source (semipermanent)
were evaluated in order to assess the repeatability of the two source types.The repeatability of the first source was
relatively high compared to the second source. After the stop of injection in August 2013, active seismic shots were acquired
repeatedly on a daily basis. These data did not show significant impedance changes at reservoir level. Microseismic event
analysis of passive seismic data detected a few local events originating from reservoir depth. In order to quantify these events the
seismic array was successfully calibrated by using independent earthquake recording. This yielded estimates of the local
magnitude for local events within the range [-2.5,0.5]. Recent work on ANSI focused on the application of this method on both
synthetic and field data. The resulting reflection response shows a strong resemblance with outcomes from active shot reflection
profiles, and the key reflectors at reservoir level have comparable characteristics