Nitrate occurs naturally in soil containing nitrogen-fixing bacteria, decaying plants, septic system effluent, and animal manure. Other sources of nitrate include nitrogenous fertilizers and airborne nitrogen compounds emitted by industry and automobiles [U.S. Geological Survey (USGS) 1996c]. Nitrate penetrates through soil and remains in groundwater for decades (Spalding and Exner 1993; USGS 1999). Groundwater is the source for > 50% of drinking water supplies, 96% of private water supplies, and an estimated 39% of public water supplies (USGS 1996a).
Factors affecting nitrate in drinking water. People who get water from shallow wells (
Drinking water regulations. The U.S. EPA sets the standards for water quality regulations written in the Safe Drinking Water Act (SDWA) and its subsequent amendments (SDWA 1974). The SDWA specifies MCL for contaminants, defined as the concentrations above which adverse human health effects may occur (U.S. EPA 2002b). The enforceable MCL applies only to public drinking water systems and government or privately run companies supplying water to at least 25 people or with ≥ 15 service connections (U.S. EPA 2003).
Drinking water standards are usually set at a fraction of the no observed adverse effect levels (NOAELs) because potential health risks are often unknown or hard to predict. The standards are based on data from experimental animal studies and available human health reports. A margin of safety is usually built into the standard to account for issues such as extrapolating from animals to humans. This safety factor (orders of magnitude) is usually higher if potential health effects are more uncertain [Risk Assessment Information System (RAIS) 1995]. The MCL for nitrate in drinking water was set at 10 mg/L nitrate-nitrogen (NO3-N) or 45 mg/L nitrate (NO3-), on the basis of 214 methemoglobinemia cases reported to the American Public Health Association for which nitrate concentration data were available (Walton 1951). Because the current MCL was nominally based on human exposure data, no uncertainty or modifying factors were used, so there is no safety factor built into the MCL for nitrates in drinking water (Johnson and Kross 1990; Walton 1951).
The appropriateness of the current MCL is questioned (Avery 1999). Complicating matters is the existence of methemoglobinemia in infants without exposure to water that contains elevated nitrate levels or any apparent exogenous nitrate or nitrite exposure. These reports suggested that bacterial infection and subsequent overproduction of nitric oxide, or other substances in drinking water, such as copper, cause methemoglobinemia in infants (Avery 1999; Felsot 1998; Hanukoglu and Danon 1996; Hegesh and Shiloah 1982). However, case reports persist of methemoglobinemia in infants related to well water containing nitrate levels above the MCL (Knobeloch et al. 2000). Private water systems are not regulated, and methemoglobinemia is not a reportable disease, so there is insufficient data on whether exposure to nitrate levels above the MCL is a problem among users of such systems.
Occurrence of nitrates in drinking water. The National Water Quality Assessment (NAWQA) program of the USGS assessed water quality of aquifer systems that cover the water resources of > 60% of the population in the contiguous United States. On the basis of the NAWQA findings, approximately 15% of shallow groundwater sampled beneath agricultural and urban areas had nitrate levels above the MCL. In comparison, 200 feet below the surface (USGS 1999). Other reports using the NAWQA data showed nitrate levels > 3 mg/L (report assumed levels of ≥ 3 mg/L because of contamination) in 28% of samples taken from public and private wells. More private wells sampled (11%) exceeded the MCL than did public wells (2%) (Squillace et al. 2002).
The U.S. EPA National Pesticide Survey (U.S. EPA 1992), which sampled private wells in 38 states and public water systems in 50 states, found 1.2% of public water systems and 2.4% of private wells exceeded the MCL for nitrate (Spalding and Exner 1993). From this survey, the U.S. EPA estimated that > 4 million people, including some 66,000 infants 5,500 private wells in nine midwestern states found nitrate levels above the MCL in 13.4% of wells sampled (CDC 1998). A survey of 3,351 domestic wells found that 9% had nitrate levels exceeding the MCL, compared with 1% of public wells (USGS 1995).
Although varied levels of nitrate in drinking water sources have been reported from state-based and national studies, a relationship between levels of nitrate and source of water is consistent. Specifically, higher levels of nitrate were found more often in groundwater than in surface water, in private wells than in public water supplies, in shallow wells than in deep wells, and in agricultural than in urban areas. The higher levels of nitrate in private water systems are of public health concern because users could be exposed to nitrate levels above the MCL.