Potential CO2 storage sites need to perform risk assessments on the likelihood of anomalous events such as leakage. The intrinsic heterogeneity of the rock system with uncertain values for the capillary threshold pressures of the various rock elements is the most likely reason for unexpected vertical migration of CO2 within a storage complex. This study shows how the Invasion Percolation Markov Chain approach can be used to address this concern. We tested the approach using detailed 3D models of the multi-layer plume at Sleipner showing that even small variations in the threshold pressures of the shales can impact the flow of CO2 into multiple accumulations. Models with and without shale breaks reveal the importance of vertical feeders and/or faults, and the geometry of the shale layers is also crucial as the CO2 strongly conforms to topography. We demonstrate that the vertical migration of CO2 at Sleipner follows a Markovian model in which the probability of later migration events is highly dependent of the probability of preceding events. This case study illustrates how the initial migration events, which have the highest probability of occurring, should be the focus of CO2 storage risk assessments.

At GTC 2022, Nvidia announced a new product family that aims to cover from small enterprise workloads through exascale high performance computing (HPC) and trillion-parameter AI models. This column highlights the most interesting features of their new Hopper graphical processing unit (GPU) and Grace central processing unit (CPU) computer chips and the Hopper product family. We also discuss some of the history behind Nvidia technologies and their most useful features for computational scientists, such as the Hopper DPX dynamic programming (DP) instruction set, increased number of SMs, and FP 8 tensor core availability. Also included are descriptions of the new Hopper Clustered SMs architecture and updated NVSwitch technologies that integrate their new ARM-based Grace CPU.