Grand Challenges in Groundwater Remediation

Top view of a groundwater contaminant simulation. Groundwater flow is from the top edge of the figure, and the green surface indicates a very low level of contaminant.

Significance

Over half of the U.S. population depends on groundwater for its water supply. Groundwater is also an important source of irrigation and industrial process water. In many regions, available sources of groundwater are a fundamental constraint on development and economic activity. Groundwater supplies are increasingly threatened by organic, inorganic and radioactive contaminants introduced to the environment by improper disposal or accidental release. Estimates of remediation costs at U.S. government sites alone range into the hundreds of billions of dollars. Protecting the quality of groundwater supplies is a problem of broad societal importance.

Remediation methods remain extremely (and potentially prohibitively) expensive and unpredictable in their success. The codes developed under HPCC will be critical in developing effective remediation strategies. Numerical modeling of groundwater transport, at the Grand Challenge level of massively parallel computing, will improve the competitiveness of the U.S. economy in three ways: by direct application of groundwater technologies to groundwater problems, by application of these technologies to related industrial processes, and by the application of generic massively parallel computational methods to industrial processes.

Accomplishments

The major research effort under this initiative is the continuing development of a state of the art parallel computer code, the Groundwater Contaminant Transport simulator, GCT, that models the flow and reactive transport of subsurface fluids through a heterogeneous porous medium of irregular geometry in three spatial dimensions. The code is designed to run on a massively parallel, distributed memory computer such as the Intel Paragon, or even on a collection of workstations.

As detailed in the 1995 PICS Research Proposal, specific research accomplishments include:

GCT 1.3: A two phase flow model was developed for the modeling of groundwater hydrology in three spatial dimensions. A parallel code for the Intel Paragon, this code is currently undergoing performance testing.
The PRE Input Preprocessor: A prototype Graphical User Interface (GUI) for the entry of hydrologic data was developed.
The EYE Output Visualizalizer: Eye offers an assortment of output viewing capabilities for PostScript, X Windows, or Silicon Graphics GL displays.

HPCC Funding Implications

Modeling is essential in understanding the extent of contamination and in developing effective remediation strategies. Modeling efforts are hampered by three issues: the proper mathematical description of the physics and chemistry governing flow and transport, the sparsity of data characterizing any one site, and the computational power required to apply sophisticated models to realistic situations. The Grand Challenge addresses the issues of data sparsity and computational power. Advances in these two areas will provide valuable information and tools to improve the third: mathematical descriptions of the problem.

The only foreseeable technology to address the computational power requirement is massively parallel computing. Both in its hardware and software aspects, this is an emerging technology. The purpose of this project is to develop this technology in place, in a form powerful enough to create a broad new capability for the prediction of groundwater remediation efforts, and thus to enable improved, cost effective and reliable remediation strategies.

This project is developing a new, state-of-the-art, parallelized flow and transport code to model groundwater transport as well as supporting technologies for parameter estimation. It will validate these codes, including sample applications to contamination sites.

HPCC funding also supports the accquistion of parallel computing equipment. The groundwater remediation codes makes particularly heavy use of the Intel Paragon, a distributed memory parallel supercomputer. PICS consortium machines include:

ORNL
XPS/5 (66 compute processors), XPS/35 (512 compute processors), and the XPS/150 (2048 compute processors)
Brookhaven National Labs and SUNY-Stony Brook
XPS/5 (56 compute processors)
Rice University
XPS/5 (56 compute processors)
Texas A&M Universiry
Sigma S2 (28 compute processors)
University of South Carolina
XPS/5 (56 compute processors)

Contacts

PICS Grand Challenge Coordinator/Technical Contact: Richard E. Ewing, ewing@isc.tamu.edu

For more information, see the PICS Groundwater home page.

Please send comments to NKC.
Last updated September 13, 2007 by abnersg@math.