2006 Philadelphia Annual Meeting (22–25 October 2006)

Paper No. 8
Presentation Time: 10:20 AM

CAN WE PREDICT PROBABILITY OF NITRATE CONCENTRATIONS IN PRIVATE WELLS IN THE GLACIAL AQUIFER SYSTEM?


WARNER, Kelly L. and ARNOLD, Terri L., U.S. Geol Survey, 1201 W. University, Urbana, IL 61821, klwarner@usgs.gov

The unconsolidated material within the glaciated area of the United States is referred to as the glacial aquifer system. It is the largest principal aquifer in area and is the most heavily pumped aquifer for drinking-water supply in the United States. Approximately 41 million people relied on the glacial aquifer system for private and public supply in 2000. Drinking-water in rural agricultural areas, of which an estimated 90 percent is from private wells, is most vulnerable to nitrate contamination because of the surrounding agricultural land use. The U.S. Geological Survey National Water-Quality Assessment (NAWQA) program began assessing water-quality conditions in the glacial aquifer system in 1991. Nitrate was analyzed from 469 private wells.

The occurrence and characterization of nitrate concentrations in the glacial aquifer system varies in areas of different intrinsic susceptibility and vulnerability. Non-parametric survival analysis is being used to compare water quality in private wells to public-supplies sampled by NAWQA. One focus of this study is developing a model to predict the probability of detecting nitrate above background concentrations (4 mg/L) in private wells where nitrate has not been analyzed. Areas of similar water chemistry, based on major ion data collected from 1991 to 2004, have been delineated within the glacial aquifer system using cluster analysis and maximum likelihood classification and are considered in predicting the probability of nitrate detections in private wells. The impact of redox on predicting the probability of nitrate concentration is being investigated. A logistic regression model to estimate the probability that nitrate concentration will be greater than background concentrations fits well when depth, casing diameter, soils, ground-water age, and nitrogen source are considered.