NERSCPowering Scientific Discovery Since 1974

Multiscale Subsurface Biogeochemical Modeling


Simulation of flow inside an experimental packed bed, performed on Franklin

Key Challenges: Predictive modeling of subsurface reactive flows is a daunting task because of the wide range of spatial scales involved, ranging over more than six orders of magnitude in length (from the pore to the field), and the wide range of time scales involved, ranging from seconds or less to millions of years.  There is additional challenge is in addressing spatial heterogeneity in the subsurface materials, the effect of multi-phase, multi-domain, coupled processes involving water, air/gas, non-aqueous phase liquids (oils, solvents), supercritical fluids (CO2), and mineral precipitation, and uncertainty quantification in the results.


Why it matters: Important applications requiring these complex multiscale, extensive time period simulations (100s-1000s years) include the fate and transport of microbial and other contaminants in aquifers, bioremediation of metal and radionuclides, and geologic carbon sequestration.

Accomplishments: The goal of this SciDAC Science Application is to characterize and model natural subsurface heterogeneity and its impact on biogeochemically reactive transport in groundwater systems. The outcome will be an integrated multiscale modeling framework that can directly link different subsurface flow, transport, and reaction process models at continuum, pore, and sub-pore scales.

Investigators: Timothy Scheibe, Bruce Palmer, Alexandre Tartakovsky, and Yilin Fang, Pacific Northwest National Laboratory; Paul Meakin, Idaho National Laboratory