Impact of fault aperture variation on induced seismicity at the VITO geothermal site

Abstract Mitigating induced seismicity remains a major challenge for geothermal reservoir stimulation. At the VITO geothermal site in Belgium, operations have induced seismic events with moment magnitudes up to M W 2.0 within the fractured Dinantian limestone reservoir. Numerical models are commonly used to forecast and reassess seismic hazard during fluid injection, supporting operational strategies to reduce the likelihood of larger seismic events. In modeling studies, fault behavior is often simplified by assuming a constant aperture to reduce computational costs. This study investigates the influence of different stress- and slip-dependent fault aperture variation laws on the simulated seismic response and examines the mechanisms driving the resulting differences in seismicity by comparing three aperture formulations: (1) constant aperture, (2) aperture dependent on effective normal stress, and (3) aperture dependent on both effective normal stress and slip displacement. Our results show that the effective normal stress dependent aperture law best reproduces the observed b -value at the VITO geothermal site. However, the model with fracture stiffness characteristics including dependencies on effective normal stress and slip displacement provides a more conservative estimate of the maximum moment magnitude and is less likely to underestimate the occurrence of larger seismic events. At the same time, it maintains computational costs comparable to those of the constant aperture model and about half those of the model that depends solely on effective normal stress. These findings highlight the importance of incorporating dynamic aperture variations to improve the reliability of seismicity models for deterministic seismic hazard assessment.