Approach for Determination of the Failure Probability of Fractures at the Soultz-sous-Forêts EGS Project

Abstract

Borehole stimulation in Enhanced Geothermal Systems (EGS) in order to improve the reservoir permeability is commonly accompanied with induced seismicity. These microearthquakes are caused by shearing of (pre-existing) fracture planes. In the present study an approach for estimation of the shearing probability of the fractures based on statistical analyses of fracture distribution, orientation and clusters, is proposed. Based on geophysical logs of five wells in Soultz-sous-Forêts, France, and with the help of statistical tools, the Mohr criterion, geological and mineralogical properties of the host rock and the fracture fillings, correlations between the geothermal wells are analyzed. This is achieved with the MATLAB-code Fracdens developed by us, which enables us to statistically analyze the datasets with depth. In a first step this is done with a focus on the orientational features of the fractures. By scanning the wells in depth intervals of 150 m, we can observe the evolution of fracture clusters with different orientation. The clusters occur at certain depths and are mainly subvertical fracture sets. The application of the Mohr Coulomb failure criterion enables us to estimate the distribution of the critical pressure of the fractures with depth. We show that the application of the Terzaghi correction for borehole orientation bias significantly influences the evaluation of the dataset, especially on the estimation of the critical pressure. Without applying the correction we can see much more critically stressed fractures than after correcting for orientation bias. In a further step we want to virtually apply a pore pressure perturbation on the reservoir to observe the evolution of the critical pressure on the fractures and thus get an idea about the mechanical response of the fractures to the pressure perturbation. A special focus is on the clay fillings of the fractures as they can reduce the frictional strength of the reservoir rock and thus determine its mechanical behavior.