Fault reactivation and associated microseismicity induced by fluid injection into the subsurface pose a potential threat in geothermal power generation. In this research, the gradient of critical pore pressure change to trigger seismicity (Δpc/h), referred to as the fracture criticality, has been proposed to represent the critical state of subsurface fractures. The fracture criticality is subjected to variability due to heterogeneous fracture attributes and rock properties. The statistics of fracture criticality could be applied to the probabilistic evaluation of fluid injection-induced seismic risk, which considers the injection-driven pore pressure increase, the variability of fracture criticality, and local fracture density. The seismic risk evaluation based on statistics of fracture criticality was applied to the Hellisheiði geothermal site, where microseismic observations and reservoir simulation over a half-year fluid injection period were integrated to achieve the probabilistic distribution of fracture criticality and evaluate the injection-induced seismic risk in both fault and off-fault zones. It has been found that the fracture criticality within both fault and off-fault zones shows natural variability (mostly ranging between 0-1.0 bar/km), and the values estimated roughly follow Gaussian distributions. Relatively high probability of seismic event occurrence was estimated for fault zones around five geothermal fluid re-injection wells at the site, which were consistent with seismically-active areas over the microseismic monitoring period.
Induced seismicity is a longstanding issue experienced in geothermal production, where geothermal fluids are extracted from and the spent fluids re-injected into the subsurface and circulate in the reservoir system. Despite extensive research on the mechanism of induced seismicity, evaluating the risk of induced seismicity associated with fluid injection remains a significant challenge.
It is imperative to understand under which conditions seismic events would be induced by fluid injection. A rigorous analysis of the injection-induced seismic risk involves careful examination of three aspects: (1) a proper understanding of hydrological and geomechanical response of subsurface reservoirs under injection conditions; (2) knowledge of the heterogeneous distribution of pre-existing fractures in the subsurface, which are difficult to directly detect and characterise; and (3) consideration of natural variability of fracture attributes and rock properties, which can result in the variability of pore pressure perturbations required to trigger seismicity. Although either rock failure or fracture slippage in the subsurface is governed by the deterministic Mohr-Coulomb failure/slippage criterion, the intrinsic variability of rock properties and fracture attributes inevitably brings about uncertainty and stochasticity in the seismic-generation process.