NITRATE IN GROUND
WATERS OF THE UNITED STATES
By Bernard T. Nolan, Barbara C. Ruddy, Kerie J. Hitt, and Dennis R. Helsel
A national map indicates that the risk of nitrate contamination of shallow ground water is greatest in the Midwest and in portions of the western and northeastern United States. Contamination risk in the Southeast is comparatively low, despite high nitrogen input and, in some cases, well-drained soils. During map calibration the extent of woodland acreage was used to represent potential nitrate attenuation processes in the Southeast, including denitrification, dilution, and plant uptake. Data collected during 1993-1995 indicate that areas with high nitrogen input, well-drained soils, and less extensive woodland relative to cropland have the highest risk of nitrate contamination of ground water less than 100 feet deep. Median nitrate concentration was 4.3 mg/L in wells representing the high-risk group, and the U.S. Environmental Protection Agency's maximum contaminant level (MCL) of 10 mg/L nitrate as nitrogen was exceeded 24% of the time. In contrast, median nitrate concentration was 0.34 mg/L in wells representing the low-risk group, and the MCL was exceeded only 5% of the time.
A principal components analysis (PCA) was performed to explore potential nitrate attenuation processes in the Southeast. A "nitrate-reduction" component explains 23% of the total variance and indicates that dissolved oxygen and nitrate are inversely related to ammonium, iron, manganese, and dissolved organic carbon. Additional components extracted by PCA include "calcite-dissolution" (18% of variance explained) and "phosphate-dissolution" (9% of variance explained). Together, the three principal components explain 50% of the total variation in the data subset representing the Southeast.
Ground water is an
important national resource, providing drinking water for more than one-half of the people
in the United States (Solley and others, 1993). Additionally,
ground water accounted for 39 percent of water withdrawn to supply cities and towns and 96
percent of water withdrawn by private users in 1990.
Ground water is vulnerable to contamination by chemicals, including nitrate, that can pass through soil to the water table. Nitrate comes from nitrogen supplied primarily by inorganic fertilizer and animal manure. Additionally, airborne nitrogen compounds emitted by industry and automobiles are deposited on the land in precipitation, gases, and dry particles (Puckett, 1994). Nitrate is soluble in water, can easily leach through soil, and can persist in shallow ground water for decades.
Ingestion of nitrate in drinking water by infants can cause low oxygen levels in the blood, a potentially fatal condition (Spalding and Exner, 1993). For this reason, the U.S. Environmental Protection Agency (USEPA) has established a maximum contaminant level (MCL) of 10 milligrams per liter (mg/L) nitrate as nitrogen (U.S. Environmental Protection Agency, 1995). Additional adverse health effects have been implicated in recent studies of contaminated ground water. A case study in Indiana indicated that nitrate concentrations of 19-29 mg/L in rural, domestic wells might have caused eight spontaneous abortions among four women during 1991-1994 (Centers for Disease Control and Prevention, 1996). Nitrate concentrations of 4 mg/L or more in water from community wells in Nebraska have been associated with increased risk of non-Hodgkin's lymphoma (Ward and others, 1996). Nitrate concentrations in natural ground waters commonly are 2 mg/L or less (Mueller and others, 1995).
Knowing where and what type of risks to ground water exist can alert water-resource managers and private users of the need to protect water supplies. A national map describing the risk of nitrate contamination of shallow ground water was compiled by Nolan and others (1997). (Please see http://wwwrvares.er.usgs.gov/nawqa/wcp/index.html for a color version of the map.) The national map shows four levels of contamination risk to shallow ground water, based on nitrogen input and aquifer vulnerability: (1) low nitrogen input and low aquifer vulnerability (green area on the map); (2) low nitrogen input and high aquifer vulnerability (yellow); (3) high nitrogen input and low aquifer vulnerability (orange); and (4) high nitrogen input and high aquifer vulnerability (red). Nitrogen input refers to nitrogen deposited on the land surface, and aquifer vulnerability indicates the likelihood that nitrate from a nitrogen source at the land surface will reach the water table. High-risk areas generally have high nitrogen input, well-drained soils, and less extensive woodland relative to cropland. The national risk map was calibrated to historical data collected during 1969-1992 and compiled by Mueller at al. (1995). Although the historical data generally support the national risk map, it had not been verified with an independent data set.
The national map shows that the Midwest has a high risk of ground-water contamination by nitrate, but parts of the western and northeastern United States also are high-risk. Contamination risk in the Southeast was lower than expected, given the high nitrogen input in the region and, in some cases, well-drained soils. Kellogg and others (1992) predicted high vulnerability of ground water to contamination by nitrogen fertilizer for the Southeast. Their result was based on soil leaching potential, precipitation, and chemical use, but did not consider the potential for nitrate attenuation by natural processes. Nolan and others (1997, 1998) used the extent of cropland versus woodland in agricultural areas to represent a combination of nitrate attenuation processes in the Southeast, including denitrification, dilution, and plant uptake. Precipitation seeping through forest soils to ground water contains less nitrogen than seepage beneath an agricultural field, and Lowrance (1992) discussed denitrification and plant uptake beneath forests bordering streams near cropland in the Coastal Plain of Georgia. Nolan and others (1997, 1998), however, did not evaluate chemical compounds other than nitrate, so processes influencing ground-water quality in the Southeast were not well understood.
The objectives of the current paper are
l to statistically verify the national risk map with ground-water nitrate data collected by the National Water-Quality Assessment (NAWQA) Program during 1993-1995; and
l to infer mechanisms by which nitrate concentration in ground-waters of the southeastern Unites States is attenuated.
Results are described below, after discussion of methods.
Data on shallow ground-water (less than 100 feet deep) collected by the NAWQA program were stratified by the risk groups delineated on the national risk map, to statistically verify map predictions. These data represent more than 1,400 wells sampled during 1993-1995. Median nitrate concentration was used to represent the central tendency of risk groups shown on the national map. The median is resistant to outliers, which commonly cause water-quality data to be highly skewed (Helsel and Hirsch, 1992). Only the most recent observation from a site was used to preclude undue influence by sites having more than one nitrate observation. Censored values were converted to one-half the detection limit before computing medians.
Principal components analysis (PCA) was used to infer processes affecting ground-water nitrate concentration in the southeastern U.S. PCA reduces a large number of variables to a few independent, composite variables (principal components) that explain much of the variance of the original data (Puckett and Bricker, 1992). Component loadings are correlations between the original variables and each principal component. Variables that are highly correlated with a principal component can indicate underlying processes influencing the data. Interpreting principal components, however, is a subjective process.
The data subset used in PCA consisted of values of nutrients, major ions, and field-measured water-quality properties collected during 1993-1995 as part of agricultural land-use studies conducted by southeastern study units of the NAWQA Program: Apalachicola-Chattahoochee-Flint River Basin (ACFB); Albemarle-Pamlico Drainage Basin (ALBE); Georgia-Florida Coastal Plain (GAFL); and Potomac River Basin (POTO). NAWQA "land-use studies" focus on the co-occurrence of specific land-use and hydrogeologic conditions.
Results and Discussion
Statistical verification of national risk map
The data collected during 1993-1995 indicate that nitrate concentration in ground water generally increases with increasing risk (figure 1). Each bar in figure 1 represents a risk group of the same color on the national map. Median nitrate concentration was 4.3 mg/L in wells representing the high-risk group, and the USEPA's MCL of 10 mg/L nitrate as nitrogen was exceeded 24% of the time (figure 1). (See http://wwwrvares.er.usgs.gov/nawqa/wcp/index.html for a color version of the bar chart.) In contrast, median nitrate concentration was 0.34 mg/L in wells representing the low-risk group, and the MCL was exceeded only 5% of the time. The statistical analysis focused on shallow ground water (less than 100 feet deep), which is closer to the land surface and to potential sources of contamination. Nitrate contamination of ground water greater than 200 feet deep is unlikely, even in high-risk areas (Nolan and others, 1998).
Data for specific locations give examples of the difference in risk. Ground water in areas shown in red in southeastern Washington State on the national risk map has a median nitrate concentration of 9.3 mg/L. In contrast, ground water in areas shown in green and yellow in western New Mexico, where nitrogen input is low, has a median nitrate concentration of 0.17 mg/L.
Poorly drained soils can reduce the risk of ground-water contamination, even in areas with high nitrogen input. For example, ground water in areas shown in orange in southern Indiana on the national map has a median nitrate concentration of < 0.05 mg/L. Although nitrogen input is high, most soils in the area are poorly drained, which restricts the movement of nitrate to the water table (Mueller and others, 1995). Additionally, drains and ditches carry water off to streams rather than letting it seep to ground water.
Large amounts of woodland interspersed among cropland can decrease the likelihood of ground-water contamination, even in areas with high nitrogen input and, in some cases, well-drained soils. The median concentration of nitrate in ground-water samples from agricultural land-use studies in the southeastern U.S. (predominantly green and orange on the national map) is 1.5 mg/L. (Paired wells were excluded when computing the regional median to preclude undue influence by dense sampling networks in a given area.) Median nitrate concentration in ground-water is less than 2.0 mg/L in four out of seven studies in agricultural areas of the Southeast, and a fifth study has only slightly elevated nitrate (2.2 mg/L) (table 1).
Ground-water nitrate data in some areas did not conform to risk patterns shown on the national map. For example, median nitrate concentration in ground-water samples from eastern North Dakota (shown in red and orange on the national map) is < 0.05 mg/L, even though the map indicates high contamination potential. The undulating, hilly landscape might be a factor. Although soils in the area are fine-textured, they are classified as well-drained because of their position and slope on the landscape. Water quickly runs off the hills and collects in low-lying areas, where denitrification can occur. Other factors not used to create the national map but that can affect nitrate concentration in ground water include land use, aquifer type, and rainfall and irrigation amounts.
Nitrate attenuation and
other processes in the southeastern United States
PCA results indicate that the first three components explain 50% of the variance in the data (table 2). Variables with the highest loadings are shown in bold type in table 2 for each principal component. Component 1, designated "nitrate reduction," indicates that dissolved oxygen (DO) and nitrite-plus-nitrate (NO2- plus NO3-, referred to as "nitrate" in this paper) are inversely related to iron (Fe), manganese (Mn), ammonium (NH4+), and dissolved organic carbon (DOC). (Although ammonium concentration is reported as ammonia by the USGS's National Water Quality Laboratory, it will be referred to as "ammonium" in this paper. Ammonia in natural ground water typically exists as ammonium ion.)The strong relation between nitrate and DO suggests bacterially mediated processes that reduce nitrate under anaerobic conditions, such as denitrification. When DO levels are insufficient for aerobic oxidation of organic matter, bacteria use nitrate as an electron acceptor (Korom, 1992; Speiran, 1996). DOC is inversely related to nitrate, further suggesting denitrification. Higher organic carbon concentrations support more heterotrophs (Tiedje and others, 1982), which use the organic carbon as an energy source to obtain cellular carbon (Korom, 1992).
PCA results indicate additional processes not related to nitrate reduction. Component 2, designated "carbonate dissolution," indicates positive relations between calcium (Ca), alkalinity as CaCO3, pH, specific conductivity, and dissolved solids (DS) (table 2). The high Ca loading suggests weathering of calcite (CaCO3). This is consistent with the hydrogeology of the Potomac River Basin, where the Great Valley is underlain by carbonate bedrock (Ator and others, 1998). Additionally, high loadings of bicarbonate alkalinity, pH, and Ca were associated with calcite dissolution in prior research involving PCA analysis of stream-water quality in Virginia and Maryland (Puckett and Bricker, 1992).
Component 3, designated "phosphate dissolution," indicates positive relations between phosphorus (P), ortho phosphorus (PO43-), and fluoride (F) (table 2). In the ALBE study unit, phosphorus concentration was significantly higher in deeper ground water, compared with shallow ground water. The comparatively high phosphorus concentration at depth suggests a mineral source such as fluorapatite, Ca5(PO4)3F, in the aquifer sediments (Spruill and others, 1998).
Analysis of an independent data set collected by NAWQA during 1993-1995 indicates that nitrate concentration in ground water generally increases with increasing risk group as defined on a national map. The risk of nitrate contamination of ground water is highest in areas with high nitrogen input, well-drained soils, and less extensive woodland relative to cropland. Median nitrate concentration was 4.3 mg/L in wells representing the high-risk group, and the USEPA'S MCL of 10 mg/L was exceeded 24% of the time. In contrast, median nitrate concentration was 0.34 mg/L in wells representing the low-risk group, and the MCL was exceeded only 5% of the time.
Nitrate-reduction appears to be an important attenuation process for ground waters of selected areas in the southeastern United States. Principal components analysis of ground-water quality data from the region yielded a "nitrate-reduction" component which indicates that dissolved oxygen and nitrate are inversely related to ammonium, iron, and manganese. The nitrate-reduction component explained 23% of the total variation in the data set.
Ator, S.W., Blomquist, J.D., Brakebill, J.W., Denis, J.M., Ferrari, M.J., Miller, C.V., and Zappia, H., 1998, Water quality in the Potomac River Basin, Maryland, Pennsylvania, Virginia, West Virginia, and the District of Columbia, 1992-1996: U.S. Geological Survey Circular 1166, 38 p.
Centers for Disease Control and Prevention, 1996, Spontaneous abortions possibly related to ingestion of nitrate-contaminated well water--LaGrange County, Indiana, 1991-1994: Morbidity and Mortality Weekly Report, v. 45, n. 26, p. 569-572.
Helsel, D.R., and Hirsch, R.M., 1992, Statistical methods in water resources: Elsevier, New York, 522 p.
Kellogg, R.L., Maizel, M.S., and Goss, D.W., 1992, Agricultural chemical use and ground water quality: where are the potential problem areas?: U.S. Department of Agriculture.
Korom, S.F., 1992, Natural denitrification in the saturated zone: a review: Water Resources Research, v. 28, n. 6, p. 1657-1668.
Lowrance, R., 1992, Groundwater nitrate and denitrification in a Coastal Plain riparian forest: Journal of Environmental Quality, v. 21, p. 401-405.
Mueller, D.K., Hamilton, P.A., Helsel, D.R., Hitt, K.J., and Ruddy, B.C., 1995, Nutrients in ground water and surface water of the United States--an analysis of data through 1992: U.S. Geological Survey Water-Resources Investigations Report 95-4031, 74 p.
Mueller, D.K., and Helsel, D.R., 1996, Nutrients in the Nations waters--too much of a good thing?: U.S. Geological Survey Circular 1136, 24 p.
Nolan, B.T., Ruddy, B.C., Hitt, K.J., and Helsel, D.R., 1997, Risk of nitrate in groundwaters of the United Statesa national perspective: Environmental Science and Technology, v. 31, n. 8, p. 2229-2236.
Nolan, B.T., Ruddy, B.C., Hitt, K.J., and Helsel, D.R., 1998, A national look at nitrate contamination of ground water: Water Conditioning and Purification Magazine, v. 39, n. 12, p. 76-79.
Puckett, L.J., 1994, Nonpoint and point sources of nitrogen in major watersheds of the United States: U.S. Geological Survey Water-Resources Investigations Report 94-4001.
Puckett, L.J., and Bricker, O.P., 1992, Factors controlling the major ion chemistry of streams in the Blue Ridge and Valley and Ridge physiographic provinces of Virginia and Maryland: Hydrological Processes, v. 6, p. 79-98.
Solley, W.B., Pierce, R.R., and Perlman, H.A., 1993, Estimated use of water in the United States in 1990: U.S. Geological Survey Circular 1081, 76 p.
Spalding, R.F., and Exner, M.E., 1993, Occurrence of nitrate in groundwatera review: Journal of Environmental Quality, v. 22, no. 3, p. 392-402.
Speiran, G.K., 1996, Geohydrology and geochemistry near coastal ground-water- discharge areas of the Eastern Shore, Virginia: U.S. Geological Survey Water-Supply Paper 2479, 73 p.
Spruill, T.B., Harned, D.A., Ruhl, P.M., Eimers, J.L., McMahon, G., Smith, K.E., Galeone, D.R., and Woodside, M.D., 1998, Water quality in the Albemarle-Pamlico drainage basin, North Carolina and Virginia, 1992-1995: U.S. Geological Survey Circular 1157, 36 p.
Tiedje, J.M., Sexstone, A.J., Myrold, D.D., and Robinson, J.A., 1982, Denitrification: ecological niches, competition and survival: Antonie van Leeuwenhoek, v. 48, p. 569-583.
U.S. Environmental Protection Agency, 1995, Drinking water regulations and health advisories: Office of Water, Washington, D.C.
Ward, M.H., Mark, S.D., Cantor, K.P., Weisenburger, D.D., Correa-Villaseņor, A., and Zahm, S.H., 1996, Drinking water nitrate and the risk of non-Hodgkin's lymphoma: Epidemiology, v. 7, n. 5, p. 465-471.