A Geographic Information System (GIS)-Based Method for Estimating Nitrate in Drinking Water

Nitrate is a precursor in the endogenous formation of N-nitroso compounds, which are potent animal carcinogens. Elevated nitrate levels in drinking water have been associated with an increased risk of some cancers and other adverse health outcomes. Private wells are not regulated and often exceed the Maximum Contaminant Level of 10 mg/L nitrate-nitrogen in agricultural areas including our study area, the Platte River Basin in Colorado and Nebraska. Due to the lack of nitrate data for private wells, this population is often excluded from epidemiologic studies in spite of their higher exposure levels. We have developed an exposure metric for nitrates in drinking water supplies for use in a small area ecologic analysis of cancer incidence. The study area consists of 344 census block groups (CBG) located within the Platte River Basin (population = 368,803). We first assigned utility-based nitrate data to communities served by public water supplies in CBG. For the population on private well supplies, we developed a GIS-based linear regression model that predicts nitrate levels in the predominant groundwater aquifer used as a water supply. Variables in the model were derived from 12 GIS data layers, and include surface and pivot irrigation, corn agriculture, streams, soil permeability, cattle and swine feedlot locations, dairy operations, and depth to water table. Results from the model were used to assign nitrate exposure levels for private wells in each CBG. We then computed a population-weighted average of the private and public water supply exposure indices for each CBG for use in our epidemiological study. Preliminary results from the application of our model indicate that proximity of corn agriculture, irrigation, cattle feedlots, and surface streams can be used as explanatory variables in assigning a nitrate exposure index to populations using well water supplies in hydrological settings such as the Platte River Basin. Similar explanatory variables for nitrate in the same study region were reported by Dunliner et al. (1996). The results of the study also demonstrate the utility of a GIS-based approach in applying such exposure metrics to a subpopulation that is normally excluded from epidemiological studies of nitrates in drinking water supplies.

Riggs PD, Nuckols JR, Maxwell SM, Miller RS, Ward MH. A Geographic Information System (GIS)-Based Method for Estimating Nitrate in Drinking Water. International Society of Exposure Analysis 12th conference and International Society for Environmental Epidemiology 14th conference, August 11-15, 2002. Vancouver, British Columbia, Canada.

View the Project Poster (large-format) in .PNG graphic format ( 937 KB download )

(click here for help with .png files)

project image from EHASL