Sensitivity Analysis in Numerical Simulation of Multiphase Flow for Co2 Storage in Saline Aquifers Using the Probabilistic Collocation Approach

Injection of CO2 in geological formations has been proposed as an effective strategy to reduce the emissions of greenhouse gases. The understanding of interactions and the design of operations at an aquifer scale require the use computationally intensive numerical simulations. These simulations require the specification of flow and transport parameters of the formation that in general are poorly known leading to uncertainty in the calculated response. Therefore it is very important to determine which of these input variables have most influence on the predictions; this is called Sensitivity Analysis (SA). In general these methodologies require multiple runs of the original model becoming computationally intensive in even simple problems. In this paper, an efficient Sensitivity Analysis using polynomial chaos expansion along with the probabilistic collocation method is used in a simple problem of the numerical simulation of injection of CO2 in geological formations