The mechanical integrity of caprocks overlying injection formations is one of the key factors for safe storage of carbon dioxide in geological formations. Undrained effects caused by CO₂ injection on strength and elastic parameters should be properly considered in the operational design to avoid fracture creation, fault reactivation and unwanted surface uplift. This study presents results from eleven undrained triaxial compression tests and one oedometer test on the Draupne shale, which is the main caprock of the Smeaheia site in the North Sea, to extract parameters relevant for seal integrity. Tests have been performed on samples oriented perpendicular to and parallel with the horizontal layering of the rock to study the effects of sample orientation relative to the loading direction. Results from undrained triaxial tests showed only minor effects of sample orientation on friction and cohesion. However, when loading during undrained shearing was parallel with layering (horizontal samples), measured Young’s modulus was roughly 1.4 times higher than for the vertical samples. Undrained shearing of vertical samples generated 30–50% more excess pore pressure than for horizontal samples with similar consolidation stress owing to more volume compaction of vertical samples. With apparent pre-consolidation stress determined from a high-stress oedometer test, the normalized undrained shear strength was found to correlate well with the overconsolidation ratio following the SHANSEP (Stress History and Normalized Soil Engineering Properties) procedure.

Recently, the road and railways infrastructure developments in Norway have led to renewed interests on the geotechnical challenges associated with slope stability under freezing-thawing cycles. Despite the amount of research available on the topic, there are few comprehensive studies involving laboratory testing, numerical analysis, and field monitoring of a slope during a freezing–thawing. In this case study, a critical slope was identified in a cold region based on field and laboratory tests, and a series of numerical simulations were carried out to evaluate the governing factor of slope stability using finite element methods. A remote monitoring system was installed on a real scale man-made slope to observe its behavior against the governing factors of slope stability. As a result, it was found that slope stability at the critical slope was significantly impacted by the freezing–thawing action, which was confirmed by the initial field observations from 2019 to 2020. Later, continuous monitoring data could be used to update soil parameters and to implement an early warning system for the high risky slope areas effected by freezing–thawing action in many cold regions.

Fractures and faults are critical elements affecting the geomechanical integrity of CO₂ storage sites. In particular, the slip of fractures and faults may affect reservoir integrity and increase potential for breach, may be monitored via the resulting seismicity. This paper presents an experimental study on shale samples from Draupne and Rurikfjellet formations from the North Sea and Svalbard, Norway, using a laboratory test procedure simulating the slip of fractures and faults under realistic stress conditions for North Sea CO₂ storage sites. The motivation of the study is to investigate whether the slip along the fractures within these shales may cause detectable seismic events, based on a slip stability criterion. Using a direct shear apparatus, frictional properties of the fractures were measured during shearing, as a function of the shear velocity and applied stress normal to the fracture. We calculated the friction coefficient of the fractures during the different stages of the shear tests and analysed its dependency on shear velocity. Information on velocity-dependent friction coefficient and its evolution with increasing slip were then used to assess whether slip was stable (velocity-strengthening) or unstable (velocity-weakening). Results showed that friction coefficient for both Draupne and Rurikfjellet shales increased when the shear velocity was increased from 10 to 50 µm/s, indicating a velocity-strengthening behaviour. Such a behaviour implies that slip on fractures and faults within these formations may be less prone to producing detectable seismicity during a slip event. These results will have implications for the type of techniques to be used for monitoring reservoir and caprock integrity, for instance, for CO₂ storage sites.

The thermal needle probe method, which is widely used for measuring the thermal conductivity λ of soils, deploys a long and thin metallic probe that houses a line heater and a temperature sensor. However, inserting such probes into consolidated or densely compacted soils or rocks is difficult, frequently causing buckling of the probe and severe disturbance to the surrounding ground, leading to unreliable measurements. We found that the use of a pre-drilled hole filled with thermally conductive grease for installing a thermal needle probe was feasible to overcome such challenges, and still yielded reliable measurements of thermal conductivity. The proposed method, i.e., the pre-drilling thermal needle probe method, was verified by finite element calculations and laboratory experiments by varying various parameters, such as the pre-drilled hole diameter, probe diameter, and thermal conductivity of thermal grease. It was observed that increases in the pre-drilled hole diameter and probe diameter and a decrease in the thermal conductivity of the thermal grease caused delays in temperature increase owing to the slowed heat transfer. Nevertheless, all the estimated λ values agreed well with the reference λ values with acceptable errors. Thus, the proposed method yields reliable measurements and can be applied for a wide range of soils from compacted soils to hard rocks.

Leak-off pressure (LOP) is a key parameter to determine the allowable weight of drilling mud in a well and the in situ horizontal stress. The LOP test is run in situ and is frequently used by the petroleum industry. If the well pressure exceeds the LOP, wellbore instability may occur, with hydraulic fracturing and large mud losses in the formation. A reliable prediction of LOP is required to ensure safe and economical drilling operations. The prediction of LOP is challenging because it is affected by the usually complex earlier geological loading history, and the values of LOP and their measurements can vary significantly geospatially. This paper investigates the ability of machine learning algorithms to predict leak-off pressure on the basis of geospatial information of LOP measurements. About 3000 LOP test data were collected from 1800 exploration wells offshore Norway. Three machine learning algorithms (the deep neural network (DNN), random forest (RF), and support vector machine (SVM) algorithms) optimized by three hyperparameter search methods (the grid search, randomized search and Bayesian search) were compared with multivariate regression analysis. The Bayesian search algorithm needed fewer iterations than the grid search algorithms to find an optimal combination of hyperparameters. The three machine learning algorithms showed better performance than the multivariate linear regression when the features of the geospatial inputs were properly scaled. The RF algorithm gave the most promising results regardless of data scaling. If the data were not scaled, the DNN and SVM algorithms, even with optimized parameters, did not provide significantly improved test scores compared to the multivariate regression analysis. The analyses also showed that when the number of data points in a geographical setting is much smaller than that of other geographical areas, the prediction accuracy reduces significantly.