SBIR-STTR Award

Numerous types of data are collected when potential remediation sites are investigated for site-specific environmental problems or oil and gas wells are drilled. These data are processed to estimate physical, chemical, biological parameters of interest. Often these investigative records are processed more or less independently, even though one data type may be correlated with another. This project focuses on a rapid method for joint inversion and imaging of geophysical and geotechnical data. Drilling, fluid, core, seismic, and electromagnetic records will be incorporated. Innovative neural network and optimization methods will be developed to perform three- and four-dimensional data assimilation (sometimes called "data fusion") using personal computers. The Joint Environmental Data Imaging and Inversion (JEDII) software technology is devised for joint, fully three-dimensional data analysis and interpretation. During Phase I, the JEDII software technology will be breadboarded, a minimum of two differing types of geophysical or geotechnical signals will be jointly assimilated, and partners for Phase II, including Phase III customers, will be identified. Phase II will expand on the number and types of signals that are jointly processed, will "ruggedize" the software for commercial applications, and pursue demonstrations and potential customer sites to solidify financial support for Phase III.Commercial Applications and other Benefits as described by the awardee:Imaging and inversion play important roles in discovery, characterization, simulation, and performance evaluation for subsurface mineral, energy, and environmental remediation projects. The market includes (1) all DOE sites that have subsurface contamination; (2) all DOE sites having tank waste management responsibilities that include removal activities, and (3) all Superfund sites having significant subsurface contamination (approximately 80 percent of Superfund sites according to a recent Environmental Protection Agency report).

The accurate interpretation of multiple geophysical and geotechnical data types is required to develop a coherent understanding of the earth=s subsurface, needed for environmental cleanup and groundwater resource protection. Such data can be acquired with relatively little disturbance to contaminated sites, yet fully utilizing such data simultaneously has proven problematic. Rapid methods for inversion and imaging of geophysical and geotechnical data to characterize subsurface fluids and media will increase the value of collected data. This project will develop an innovative computational system based on a combination of artificial neural network and data inversion methods. The Joint Environmental Data Imaging and Inversion (JEDII) software technology is being devised for joint, fully three-dimensional data analysis and interpretation and will be deployed on PC-class computers. During Phase I, artificial neural networks for interpolation and inversion were developed and tested, and proof-of-concept was achieved. In Phase II, the inversion strategies based on artificial neural networks will be supplemented by a new, more classical, data inversion method which uses geostatistical constraints and estimates.

Commercial Applications and Other Benefits as described by the awardee:
Derived benefits from this project include improve performance monitoring of active and passive environmental cleanup systems, interpretation of geophysical data collected for contaminant identification, and improved design of groundwater resource protection and exploitation systems.