Subsurface flow and transport take place in complex heterogeneous environments that exhibit a hierarchy of scales. More often than not, physical and bio-geochemical phenomena on one scale (e.g., a pore scale) affect, and are coupled to phenomena on a vastly different scale (e.g., a field scale). Such phenomena defy coarse-scale continuum descriptions, since they exhibit high localization (e.g., propagation of reactive fronts and biofilm growth) and/or strong nonlinear coupling between the processes involved (e.g., dynamic changes in porosity and permeability due to dissolution or precipitation). Hybrid numerical algorithms are to be used when coarse-scale continuum models fail to accurately describe a physical phenomenon in a small part of a computational domain. Accurate and efficient coupling of two (or more) models operating on vastly different spatial and/or temporal scales in a hybrid remains a major theoretical and computational challenge. A key to the success of a hybrid method is an efficient implementation of coupling conditions on the interface between its constitutive models. Additionally, uncertainty about pore geometry undermines the veracity of pore-scale simulations and, hence, of hybrid simulations of which they are a constitutive part. Overreaching goals of the proposed activity are to establish a theoretical foundation for hybrid modeling of subsurface flow and transport, to develop corresponding numerical algorithms, and to provide computational tools for robust uncertainty quantification in hybrid models. To achieve these goals, the investigators will develop hybrid algorithms for reactive flows in fractured and porous media, quantify uncertainty in pore-scale geometry in hybrid simulations, and experimentally validate hybrid simulations.
In terms of broader impact, this proposal will enhance sustainability of essential water resources. Groundwater is a principal source of domestic water supply in the United States, and it is a major source of fresh water for industrial, agricultural and public uses. By establishing a novel modeling framework, this proposal will provide a scientific basis for reliable predictions of impacts of land use change and climate change, and more accurate assessments of groundwater contamination risks. The methods and algorithms developed in the course of the proposed activity will lay a solid foundation for improved quantitative understanding of subsurface processes with tightly coupled pore- and field-scales. The proposal will support the efforts of graduate students. The investigators will actively promote STEM career paths of underrepresented groups, and establish an outreach K-12 program dedicated to providing an intensive college prep education for motivated low-income students.
