Title: Development and Application of a Sundance-Based Simulator to Model Non-Darcy Flow Through Porous Media

Speaker: Jill Reese, von Neumann Fellowship Candidate, North Carolina State University

Date/Time: Wednesday, February 1, 2006, 9:00 – 10:00 am

Location: Building 980, Room 95 (Sandia NM) Building 915, Room S145 (Sandia CA)

Brief Abstract: Problems involving flow through porous media arise routinely in nature, including applications of interest to DOE, such as groundwater flow near Yucca Mountain, the WIPP site, and legacy waste sites. Porous medium models are typically based on some combination of conservation principles and experimentally derived closure relations. Darcy's law, rooted in experimental work performed some 150 years ago, is a closure relation used almost universally in porous medium models as an approximation for momentum conservation.

The focus of this work is the development of computational tools needed to simulate flow through porous medium systems.  The underlying non-Darcy flow model, based upon an evolving theory, reduces to the traditional Darcy-based model under certain circumstances but includes more a faithful representation of the physics needed to accurately resolve certain systems.

The simulator developed is based upon Sundance, which is a powerful, general, CCA compliant problem solving framework that has been used to simulate a wide range of computational science problems. During an internship at Sandia, I extended the capabilities of Sundance by implementing an interface to the modern nonlinear solver capabilities in NOX. With this interface, accurate, efficient, and scalable simulators can be developed for models based on nonlinear partial differential equations.

In this presentation I focus on the computational science tools developed, the general applicability of these tools, and the application of Sundance to simulate the non-Darcy model formulation of interest.  Additionally, issues of convergence rates in heterogeneous systems are examined and scalable results are demonstrated.  Finally, a range of scientific issues that may be resolved using the simulation tools developed is summarized.

CSRI POC: Patty Hough, (925) 294-8962