Groundwater

Location:  The Floridian Aquifer in Florida’s Suwannee Rive Valley comprising a complex natural system, some 100 x 100 miles

Problem:  Interpreting megabytes of data collected from an extensive well monitoring network collected over 40 years and the high cost of maintaining the network.

Results:   ADMi successfully applied its data mining tools to the study area to show how a large data collection network could be improved while saving money by optimizing its information gathering performance.   The ANN-based groundwater model was found to be extraordinarily accurate and quickly developed when compared to traditional model being developed by another contractor.

Monitored Natural Attenuation (MNA)

Cleanup of contaminated groundwater is a high priority in the US and around the world.  The scope of the problem and technical difficulties involved in the cleanup is immense. Estimates of the number of hazardous waste sites where groundwater may be contaminated vary between 300,000 and 400,000 nationwide [NRC, 1994].  One alternative to engineered cleanup is a strategy known as monitored  natural attenuation (MNA).  The Environmental Protection Agency (EPA) defines natural attenuation as "a variety of physical, chemical, or biological processes that, under favorable conditions, act without human intervention to reduce the mass, toxicity, mobility, volume, or concentration of contaminants in soil or groundwater. These in situ processes include biodegradation; dispersion; dilution; sorption; volatilization; radioactive decay; and chemical or biological stabilization, transformation, or destruction of contaminants" [EPA, 1999].  The cost of the methods traditionally used to measure attenuation is high.  The tools of data mining can be used to determine if natural attenuation is proceeding satisfactorily with respect to containment and mitigation rate.  Data mining tools can also be used to correlate variables that are easily and less expensively measured with those traditionally used.