IMPROVING THE ASSESSMENT OF CARBON CAPTURE AND STORAGE RISK ANALYSIS BY PROPER REPRESENTATION OF HYDRAULIC CONDUCTIVITY AND DISPERSIVE PROPERTIES

Carbon Capture and Storage (CCS) in deep saline aquifers provides a means to reduce CO₂ emissions into the atmosphere, but may have adverse impacts on overlying aquifer drinking-water quality if leakage occurs. A probabilistic risk assessment that quantifies the human health risks associated with CO₂ leakage by explicitly examining solute transport from the leakage source to the individual is presented here. This type of transferrable methodology is necessary to compare the risk associated with potential CCS sites and is a crucial component to large-scale CCS evaluation and implementation. Risk assessment in contaminant hydrology involves a combination of uncertainties in toxicity and hydrologic constructs. Uncertainties associated with flow and transport parameters greatly affect the uncertainty associated with overall health risk. Limiting hydrologic uncertainties and understanding how these hydrologic parameters can influence model outcomes can greatly improve risk predictions. The level of heterogeneity and spatial correlation of hydraulic conductivity determines plume characteristics and creates complex interactions with dispersive mechanisms. Hydrodynamic dispersion, both in the longitudinal and transverse directions, can potentially play important roles in plume migration, concentration, and extent. In this work we investigate the interaction between dispersion mechanisms and patterns of hydraulic conductivity heterogeneities, specifically aquifer stratification and hydraulic conductivity anisotropy. Contaminant plume characteristics attributed to dispersion and aquifer heterogeneity are discussed in detail. A comparison of particle tracking and streamline model approaches in terms of accuracy and computational efficiency are also discussed. Lastly, we provide examples and applications of how accounting for these hydrologic mechanisms may reduce risk uncertainties, specifically in CO₂ leakage scenarios.