Subsurface contamination by nonaqueous phase liquids (NAPLs) is a frequent occurrence. One of the most important process influencing risk assessment and environmental management decisions at NAPLcontaminated sites is the rate at which NAPLs dissolve into groundwater. NAPL dissolution controls fluxes of contaminants to down-gradient receptors, and is a significant process affecting the design of monitoring systems, the assessment of human and ecological risk, and the selection of remediation strategies.
Intellectual Merit: Numerous models for NAPL dissolution have been developed based upon and intended to describe centimeter-scale laboratory columns. While our understanding of NAPL dissolution at this scale is relatively mature, this understanding does not translate to reliable field-scale models for two reasons: (1) the computational problem is intractable with current technology if one endeavors to resolve centimeter-scale features known to be important for field-scale simulations; and (2) upscaled models of NAPL dissolution, which are computationally tractable, are in their developmental infancy. We have shown in previous work that the dissolution process leads to the formation of dissolution fingers, which have a profound effect on the rate of dissolution. Because these fingers can affect the rate of mass transfer by two or more orders of magnitude in some cases, it is essential that the effect of this mechanism be represented in field-scale simulators, and we know of no case in which this has been done. The goal of this proposed work is to extend a novel fractal-based upscaling approach to produce a nonlinear, multiscale mass transfer model suitable for more realistic field-scale simulation. Multiscale models are the holy grail of subsurface science, indeed much of science in general. Our efforts to date in this regard have met with considerable success, but several key unresolved issues that influence NAPL dissolution fingering remain: NAPL wettability, porous medium heterogeneity, the initial spatial distribution of NAPL contamination at the onset of dissolution, and the effect of realistic patterns of field-scale heterogeneity on the development and evolution of fingers and NAPL dissolution in general. A combination of creative theoretical, novel experimental, and advanced computational approaches will be used to address these issues.
Broader Impacts: In this work, a numerical model will be developed and used to investigate a nonlinear process and to extend the range of conditions beyond what can be easily studied in the laboratory to those typical of complex natural systems. The synergy between the experimental and computational approaches provides a unique opportunity to draw high school and undergraduate engineering and science students into the exciting world of subsurface hydrology and computational science. To that end, we will create two Problem-Based Learning (PBL) modules based on our work that will be tested with undergraduate students at the University of Delaware (UD) and the University of North Carolina (UNC) and used in a program intended to attract gifted high school students into engineering and science. PBL is a successful example of student-centered or active learning, the strengths of which have been firmly established by research in educational psychology. UD has developed an institutional focus on PBL and has become a national and international PBL leader, and they will disseminate these PBL educational modules to instructors throughout the US. In addition, this research will involve directly both undergraduate and graduate students from active research groups that have both a strong commitment to and representation of under-represented groups in science and engineering. 1
