Revisiting the Basel-1 Hydraulic Stimulation with a 3D Coupled Hydro-Mechanical Model

Abstract

On the 2nd of December 2006, the Basel-1 well was hydraulically stimulated with the aim of significantly increasing reservoir transmissivity and achieving economical geothermal energy fluxes. Over six days, 11,500 cubic meters of water were injected into a 371-meter-long openhole section of the well, reaching a depth of 5 kilometers. As seismic activity began to rise, peaking at up to 200 events per hour, the injection rate was reduced, and eventually, the well was temporarily shut down for five hours. Despite these measures, an earthquake of ML 3.4 occurred (and was felt in the city of Basel) just before the wellhead pressure was fully released. In light of this significant seismic event, the project was ultimately abandoned. Over the years, the Basel hydraulic stimulation and its associated seismicity have been extensively studied. Experts have examined various aspects, including stress levels and hydraulic and microseismic data. Although numerous numerical models have been developed, none have yet succeeded in accurately and comprehensively reproducing both the measured pressure and the spatiotemporal microseismicity migration. In this paper, we revisit the 2006 Basel-1 well stimulation using a fully coupled hydro-mechanical model to reproduce both pressure data and the spatio-temporal microseismicity distribution. In our model, microseismicity is assumed to be triggered by both the increase in pore pressure and the transfer of stress due to aseismic slip. Thus, the microseismicity cloud provides an indirect measure of the evolution of aseismic slip when this mechanism is dominant. Our primary model simulates the first activated group of events, which represents the largest stimulated structure, as a single slipping patch that propagates aseismically in pure shear mode. Key parameters, such as maximum horizontal stress, friction, dilatancy, and joint closure nonlinearity are varied to align with observed pressure and microseismic migration. We find that early pressure drops can be explained by a simple linear relation between dilatancy and slip, whereas the later pressure response requires incorporating the nonlinear effects of joint closure. Our results show that a coupled hydro-mechanical model, initially restricted to a single plane with linear slip-weakening friction and nonlinearly varying dilatancy and permeability, successfully reproduces the evolution of the well pressure data and the migration of the observed micro-seismicity of the first group.