Optimizing stress tensors and friction coefficient from stress inversion of microseismicity at the Decatur CO2 storage site

The Illinois Basin–Decatur Project represents a landmark in Carbon Capture and Storage, with approximately 1 million tonnes of supercritical CO₂ being injected into the Mt. Simon Sandstone. Nearly 20,000 microseismic events were recorded during injection, offering a unique opportunity to analyze the reservoir's geomechanical state. In this study, we use two datasets containing high-quality focal mechanisms to characterize the in-situ stress field and evaluate fault reactivation potential. We first apply a full stress inversion algorithm based on the Wallace-Bott hypothesis and stochastic optimization, constrained by vertical stress (S_v) and instantaneous shut-in pressure. The study area is dominated by strike-slip faulting regime (S_Hmax>S_v>S_hmin), initial results assuming a fault friction coefficient of 0.6 and zero cohesion yield minimum activation pressure between 23.3 and 26.0 MPa. These values notably exceed the average downhole injection pressure of ~23.0 MPa, implying that, under the assumptions, the observed seismicity should not have been triggered. However, our refined analyses show that reducing the friction coefficient to 0.4 lowers the activation pressure to values 16.7–21.9 MPa. Alternatively, a comparable consistency can also be achieved by assuming lower values of shut-in pressure. Building on these results, we extend the analysis to systematically identify the optimal combination of S_Hmax magnitude and fault friction coefficient required to induce slip. We employ slip tendency analysis with nodal plane selection to ensure physical consistency. Using S_hmin and S_v estimates derived from published stress gradients and density logs, we analyze the stability of the linked faults. We then perform a sensitivity-based search to identify the optimal range of friction coefficient and S_Hmax required to trigger fault slip. By varying S_Hmax and observing corresponding changes in the maximum mobilized friction coefficient, we  narrow the potential stress magnitude range. For example, assuming an initial friction coefficient range of 0.45 to 0.6, the S_Hmax is constrained to approximately 74–92 MPa. Our results highlight that fault friction, while remaining poorly constrained at the site scale, represents a first-order control on induced seismicity and stress interpretation. Integrating high-quality source mechanisms with sensitivity-based constraints on stress magnitudes and fault properties is essential for the reliable forecasting and long-term risk assessment of large-scale geological CO₂ storage.