Multiscale subsurface characterization for geo-energy applications

Geological Carbon Storage (GCS) is considered a promising technology to lower atmospheric emissions of CO2. Guaranteeing the sealing capacity of caprocks in this case becomes paramount as CO2 storage scales up to the gigaton scale. Many laboratory experiments have been performed with samples of intact rock, showing that low-permeability and high-entry pressure caprocks have excellent sealing capacities to contain CO2 deep underground. However, discontinuities, such as bedding planes, fractures, and faults, affect the rock properties at the field scale, being at the same time challenging to monitor in industrial-scale applications. The main objective of the thesis is to develop a methodology to characterize potential sites for low-carbon geo-energy applications at multiple scales in space and time, taking into account coupled processes. To achieve this objective, it is necessary to understand the complexity of the problem at multiple scales, starting from large-scale investigations, passing through in-situ underground rock laboratories, to laboratory investigations. This is why this thesis approaches characterizations that span a wide range of spatiotemporal scales from the order of millimeters and nanoseconds to the order of kilometers and decades. To achieve this general objective, the first part of the thesis explores four analytical solutions of pore pressure diffusion with periodic sources under relevant model geometries and boundary conditions to enable real-time interpretation of in-situ data. We compare the results with numerical solutions, assuming the same conditions as the analytical cases and incorporating reservoir deformation due to pressure waves. Encompassing three different cases to span a wide array of scenarios, we evaluate the attenuation of the signal at varying distances from the source. Numerical and analytical solutions fit when identical assumptions are upheld. Furthermore, the influence of effective mean stress variations yields errors of le