It is well known that the characterization of fluid-transport properties of rocks is one of the most important and difficult problems of the reservoir geophysics. Seismic methods have some fundamental difficulties in estimating such hydraulic properties of rocks like the fluid mobility or the permeability tensor. Microseismicity cloud for Soultz (France) borehole injection experiment We develop an approach that provides in-situ estimates of the permeability tensor characterizing a reservoir on the large spatial scale (of the order of 10^3m). This approach (we call it SBRC: Seismicity Based Reservoir Characterization) uses a spatio-temporal analysis of  fluid-injection induced microseismicity to reconstruct the permeability tensor. Small perturbations of the pore pressure caused by a fluid injection in a borehole can lead to the triggering of microearthquakes. On the Figure such a microseismicity cloud is shown for the Soultz (France) borehole injection experiment. The red line shows the borehole (GPK1). The source of injection was located approximately at the depth of 3km in crystalline rock. The vertical and horizontal sizes of the shown region are approximately 1km. The colored points are microseismic events. Colors denote the occurrence times. The SBRC approach uses such data to reconstruct the permeability tensor of rocks.

• Shapiro, S.A., Huenges, E., and Borm, G., 1997, Estimating the crust permeability from fluid-injection-induced seismic emission at the {KTB} site: Geophysical Journal International, v.131, F15-F18.
• Shapiro, S.A., Royer, J.-J., and Audigane, P., 1998, Estimating the Permeability from Fluid-Injection Induced Seismic Emission: in Thimus J.-F., Abousleiman Y., Cheng A.H.-D., Coussy O. and E. Detournay, Eds., Poromechanics, 301-305.
• Shapiro, S.A., Audigane, P.,  and Royer, J.-J., 1999, Large-scale in situ permeability tensor of rocks from induced microseismicity: Geophysical Journal International, v. 137, 207-213.

[ Seitenanfang ]

The attention to the microseismic monitoring during operation of geothermal or hydrocarbon reservoirs has grown considerably over the last several years. The observation of microseismicity occurring during borehole fluid injections or extractions has a large potential in characterizing reservoirs at locations as far as several kilometers from boreholes. Microseismic data could potentially be used to measure in-situ hydraulic properties of rocks at interwell scales,  providing information that could further guide operations to optimize field production.
Recently, an approach for the interpretation of microseismic data was proposed to provide in-situ estimates of the hydraulic diffusivity characterizing a geothermal or hydrocarbon reservoir on the large spatial scale (of the order of 103m). This approach is called "Seismicity Based Reservoir Characterization" (SBRC). It uses a spatio-temporal analysis of fluid-injection induced microseismicity to reconstruct the tensor of hydraulic diffusivity and to estimate the tensor of permeability.  Because the equations the SBRC is based on were were derived in a quasi-heuristical way, a quantitative approach is required to verify the inversion algorithms based on them. A possible way of verification is to apply the inversion algorithms to numerically simulated microseismic data. For this approach a numerical simulation of microseismicity during borehole fluid injections is required.
To model the triggering of microseismicity numerically, we simulate the process of the pore pressure relaxation in a medium with statistically distributed critical zones. We use a finite element algorithm (FE) to solve the time-dependent parabolic equation of diffusion for 2D and 3D hydraulically homogeneous, isotropic or heterogeneous, anisotropic background medium with arbitrary source point and input signal.  Comparing the solution of the pore pressure distribution with a given criticality value (failure criterion) allows to obtain synthetic microseismicity clouds. Now, the spatio-temporal evolution of the events due to pore pressure variation can be studied.

Results:  The main hypothesis of the method of Seismicity Based Reservoir Characterization (SBRC) is that fluid-induced microseismicity is triggered due to a diffusive process of pore pressure relaxation in subcritically stressed rocks. Using  this hypothesis we have developed a simple numerical model for simulating the space-time distribution of injection-induced microseismicity that depends on hydraulic properties and the statistics and spatial distributions of trigger criticality.  The forward model results show time-distance distributions of microseismicity similar to observed microseismic clouds. This similarity  supports the idea that the pore pressure relaxation is an important mechanism for triggering microearthquakes. We applied numerical simulations to test the inversion approaches of the SBRC method. We can show that if the hypothesis of the SBRC approach is valid, than the inversion method can be successfully used to reconstruct hydraulic properties of rocks from spatio-temporal evolutions of clouds of microseismic events.

• Rothert, E. and Shapiro, S. A, Microseismic Monitoring of Borehole Fluid Injections: Data Modeling and Inversion for Hydraulic Properties of Rocks. Short-note, accepted for publication Geophysics in vol. 68, no. 2, March/April 2003
• Serge A. Shapiro, Elmar Rothert and Axel Kaselow, Seismic propagating and diffusion waves to probe porous rocks under stress, 2nd Biot Conference 2002, Grenoble, France, 26.-26 August 2002, in expanded abstracts.
• Rothert, E., Shapiro S. A. and Rindschwentner, J., Beben aus dem heissen Stein, Der belebte Planet, Sonderheft der Berliner Geowissenschaftlichen Abhandlungen zum Jahr der Geowissenschaften 2002, ISBN 3-89582-098-9, S. 13-20

[ Seitenanfang ]

An approach for the interpretation of microseismic data was proposed to provide in-situ estimates of the hydraulic diffusivity characterizing a geothermal or  hydrocarbon reservoir on the large spatial scale (on the order of 103m). This approach is called "Seismicity Based Reservoir Characterization" (SBRC). The method is based on the hypothesis that the spatial propagation of hydraulically induced seismicity is caused mainly by the pore pressure relaxation process. According to this hypothesis the SBRC uses a spatio-temporal analysis of fluid-injection induced microseismicity to reconstruct the tensor of hydraulic diffusivity and to estimate the tensor of permeability in 3D. However, processes that can lead to triggering of microseismicity are not yet fully understood. A correlation of microseismic  hypocenters with structural images obtained from reflection seismics can help to better understand the physics of microseismicity triggering and thus to test the main assumption of the SBRC.
The SBRC approach was successfully applied to real data several times. Recently, fluid injection induced microseismicity at the German Continental Deep Drilling (KTB) site was analysed in terms of the SBRC method to reconstruct the tensor of permeability at the open hole section at 9.1 km depth. Using new data sets acquired in 2000 we are able to observe indications of the depth-dependency of hydraulic diffusivity at the KTB for the first time. The analysis of fluid-induced microseismicity leads to an estimation of the hydraulic diffusivity at the KTB at different depths. A lower value of hydraulic diffusivity was found in upper parts of the subsurface compared with the values at the open-hole section. Correlations with structural images were obtained. For example, we observe that rock volumes characterized by larger diffusivity also show larger reflectivity.

Acknowledgements
This work was kindly supported by the sponsors of the Wave Inversion Technology (WIT) Consortium, Karlsruhe, Germany and in a part from the Deutsche Forschungsgemeinschaft through grant SH 55/2-1 and SH 55/2-2. Data of KTB was provided courtesy of H.-P. Harjes (Bochum). We especially want to  acknowledge the cooperation with the geophysical research group of the Bochum university (Harjes, Bohnhoff, Baisch).

[ Seitenanfang ]