The main objective of the thesis is a further development of the seismicity based reservoir characterization approach (SBRC). In general, the SBRC method is applied to earthquakes resulting from fluid injections into the subsurface. This method allows firstly, to estimate the fluid-transport properties of hydraulically stimulated reservoir, secondly, to examine the fluid-rock interaction, and thirdly, to reconstruct critical pressures of the activated fractures within the reservoir. To extend the applicability of SBRC the thesis focus on the following topics.
The SBRC method so far assumes a constant injection source strength. This condition, however, is not always given, such as by the hydraulic stimulation of a geothermal reservoir in Basel (Switzerland). In the first part of the thesis, SBRC is extended in order to analyze seismicity resulting from fluid injections where the source strength is linearly increasing with time. For this purpose, an analytical solution of the diffusion equation is derived taking into account this special condition. The derived solution and the resulting expressions for the seismicity rate and the cumulative number of earthquakes are numerically verified using finite element modeling and synthetically generated seismicity. Afterwards, SBRC is applied to fluid-induced seismicity recorded in Basel providing consistent estimates of the permeability of the hydraulically stimulated reservoir and of the distribution of critical pressures.
In the second part of the thesis, a model is introduced in order to interpret induced seismicity of single-planar hydraulic fractures. The model considers the growth of fracture and seismicity as a combined geometry- and diffusion-controlled process. It is confirmed by observations from fracturing-induced seismicity in the Cotton Valley gas reservoir (USA). The space-time diagrams (r-t diagrams) of induced earthquakes show signatures of fracture volume growth, loss of treatment fluid, and diffusion of injection-induced pore pressure perturbations. Evaluation of envelopes of the spatio-temporal distribution of induced seismicity allows to determine geometrical parameters and hydraulical properties of the fracture. Considering a volume balance principle of the injected fluid permits to quantify the fluid loss from the fracture into the reservoir and to estimate the reservoir permeability. The proposed interpretational approach is applied to earthquakes induced during three fracturing stages in Cotton Valley. Although the three stages differ with respect to the treatment design parameters, it is found that all stages resulted in similar fracture geometries. Ratios of fracture volume and total injected fluid volume are nearly identical. It means that the fracture growth process is likely decoupled from the type of treatment design. Estimates of fluid loss and reservoir permeability are consistent for the investigated fracturing stages.
Fluid injections into the subsurface can sometimes induce earthquakes characterized by a significant magnitude. In particular, seismic events with larger magnitudes are reported from geothermal reservoirs. Understanding the scaling relations of magnitudes of fluid-induced seismicity is crucial for assessing the seismic risk by injection operations. In the last part of the thesis, a statistical model is introduced which describes the magnitude distribution of earthquakes induced during injections. It combines a Gutenberg-Richter statistics of magnitude probability with the cumulative number of induced earthquakes. Earthquake magnitudes presented in this thesis are in agreement with this model. Furthermore, the model allows to identify controlling parameters of the magnitude distribution. These include design parameters of a fluid injection, such as the injected fluid volume, and seismotectonic quantities like the probabilistic Gutenberg-Richter a- and b-value and the tectonic potential which is defined by statistical properties of pre-existing fractures.
Falls Ihr Browser eine Datei nicht öffnen kann, die Datei zuerst herunterladen und dann öffnen.