Stress Estimation from Clay Content and Mineralogy-Eos Well in the Aurora CO2 Storage Site, Offshore Norway

Determination of in-situ stress is vital for many engineering applications, including Carbon Capture and Storage (CCS). A reliable estimation of minimum horizontal stress (σh) in a top seal is crucial for quantifying risk related to safe CO2 injection/storage. In this study, an alternative method is used to estimate the ratio between effective horizontal stress to vertical stress (K0, which directly relates to σh) from clay volume as indicated by gamma ray and spectra gamma ray logs. The method is based on correlations from laboratory uniaxial strain tests on natural and synthetic mixtures of clay with known mineralogy. The alternative method is predicting stress from volume clay (Vcl method) assuming uniaxial strain conditions; in addition, corrections accounting for smectite-rich high plasticity clays and uplift have been introduced. The method has been demonstrated for 5 wells and one geotechnical well in Aurora, Smeaheia, and Troll East, in the Horda Platform area. It is found that the sonic log method and the Vcl method are comparable within the Dunlin Group. For organic-rich shales like the Draupne Formation, the Thorium values from the spectral gamma log should be used for Vcl rather than the regular total gamma log. When other mineralogy data are present, this can be used for more precise quantification of Vcl and smectite and better estimation of K0. The Vcl method is a good alternative when sonic logs are unavailable to fill in the gaps in between field stress data, to understand the local lithological impact on stress, and reduce uncertainty. After correcting for uplift and high smectite content, the Vcl method were found to reproduce high K0 values as indicated from field stress tests in shallower units (Nordland, Hordaland, Shetland). For deeper main sealing units Draupne and Drake, the diagenetic impact may resulted in low K0 values although extensivly uplifted. The method presented improves quantification and reduces the uncertainty of σh which is one key input in all integrity evaluations