Basic Research Needs for Geosciences - Energetics Meetings and ...

GRAND CHALLENGE: SIMULATION OF MULTISCALE GEOLOGIC SYSTEMS FOR ULTRA-LONG TIMESRESEARCH APPROACHESResearch approaches to understanding processes relevant both to CO 2 geologic emplacement andstorage and to very long-term leakage from nuclear waste storage involve development of newconceptual and theoretical models of multiscale and multiphysics processes. Numericalimplementation ultimately requires high-performance computing and model validation usingboth laboratory and field experiments. Theoretical approaches range from pore-scale models thatcan be approached with existing methods, e.g., Navier-Stokes equations and their solutionthrough lattice Boltzmann (Kang et al. 2006) and smooth particle hydrodynamics techniques, topore network and Darcy-scale continuum models (e.g., Pruess et al. 1999). Accurate tracking ofinterfaces is critical to modeling pore-scale processes, a capability that can be provided by LevelSet methods (e.g., Prodanovic and Bryant 2006). Hybrid formulations, involving coupling ofpore-scale and continuum models or viscous and Darcy flow (e.g., Oldenburg and Spera 1992)across disparate scales may be required as well.Injection of supercritical CO 2 into deep underground reservoirs introduces unique computationalchallenges arising from the formation of multiphase buoyancy-driven flow, fingering instabilitiesas CO 2 dissolves into groundwater, and viscous fingering as supercritical CO 2 displaces water.Understanding these processes requires development of models that can resolve instabilities andfingering patterns that originate on the pore scale but grow to produce emergent effects possiblyextending to the basin scale.One possible approach to representing multiscale processes is the dual or multiple interactingcontinuum model. Multiple continuum formulations appear adequate for sufficiently slowreaction rates where concentration gradients within a single pore remain small (Lichtner 2000).As rates and concentration gradients within a single pore increase, dramatic dissolution processesare possible (e.g., Golfier et al. 2002), and hybrid approaches that resolve viscous boundarylayers and related transport processes (e.g., Taylor dispersion; see Taylor 1953) in the pore spaceare needed.A significant challenge to developing and implementing multiscale models is obtaining theappropriate properties and parameters for each scale. One approach is to perform simulations atthe pore scale and then upscale the results to the continuum scale. In so doing, the mostappropriate continuum formulation is determined along with values for its system parameters.Carrying out the pore-scale modeling requires an accurate representation of the pores and theirheterogeneity. Recent advances in pore-scale imaging (e.g., focused ion beam) can achievemicron to submicron resolution. Upscaling will be essential for obtaining effective parametersfor basin-scale applications.To model processes on a wide range of scales, observations and data are needed from laboratoryand field experiments. Experiments must be designed and carried out to aid in conceptual modeldevelopment and to validate resulting models. An iterative approach will be required to test,calibrate, and refine models against lab and field experiments. Innovative microfluidicexperiments will need to be designed for multiphase systems to investigate multiphase interfacialphenomena.100 Basic Research Needs for Geosciences: Facilitating 21 st Century Energy Systems