Steven A. Frenzel 1996
U.S. Geological Survey Open-File Report 96-223
Surface and ground water in Nebraska may contain contaminants resulting from human activities. For purposes of this publication, a contaminant is any element or compound whose presence may affect the water's suitability for certain uses. For example, herbicide concentrations may exceeed the U.S. Environmental Protection Agency's (USEPA) Health Advisory Levels (HAL) for drinking water or trace-element concentrations may exceed guidelines for the protection of aquatic life. In general, the contaminats discussed in this report enter the aquatic system through nonpoint-source runoff from agricultural lands that dominate the Nebraska landscape. However,because this assessment was conducted as part of a larger, national program, a screening for contaminants with non-agricultural origins was included.
The measurement of water quality involves a variety of steps, each contributing unique information while also aggregating to an overall assessment. One aspect of water-quality assesment is to describe the occurrence and distribution of contaminants. Some contaminants may be hundreds or thousands of times more concentrated in the tissues of aquatic organisms or in fine sediments than they are in the water. As a result, fish tissue and streambed sediments are well suited for the detection of certain contaminants. For example, pesticides used in the United States prior to the early 1970's, such as DDT, may have degraded into more stable but still toxic compounds that are highly concentrated in fish tissues. Conversely, other contaminants are not concentrated in sediments or tissues but are readily detected in water samples. Organonitrogen herbicides (such as atrazine), atrazine), the most commonly used herbicides in Nebraska, are examples of water-soluble contaminants.
Several sampling strategies were used to address specific questions. Some sites were sampled repeatedly through time and during all hydrologic conditions, whereas others were sampled only once to determine presence of contaminants. Because a strong relation between concentration and streamflow often exists for contaminants originating from nonpoint sources, streams typically were sampled near gaging stations that monitor streamflow.
The areas described are the drainages from the confluence of the North and South Platte Rivers near North Platte, Nebraska to the mouth of the Platte River near Omaha, Nebraska. This study area, the Central Nebraska Basins (see map), is one of 60 study areas identified by the U.S. Geological Survey as part of the National Water-Quality Assessment (NAWQA) Program. Four environmental settings-- Sandhills, Loess Hills, Glaciated Area, and Platte Valley-- characterize the study area. Sampling sites were distributed to represent areas that were indicative of these environmental settings or to represent the integration of multiple environmental settings. The width of sampled streams varied from about 10 to about 2,000 feet and represented drainages from 132 to 30,000 square miles.
In general, the Sandhills lie in the western part of the study unit and have highly permeable soils yielding little surface runoff. Land cover of the Sandhills is grassland. The Loess Hills are located in the central part of the study area. Soils are less permeable here than in the Sandhills due to a higher content of silt and clay. Land cover is mostly irrigated rowcrops with some rangeland in the western part of the Loess Hills. The most eastern environmental setting is called the Glaciated Area; it has clay-rich soils and land cover is predominantly rowcrops. Runoff is greatest from this area, not only because of the less permeable soils but also because of greater precipitation than in the other areas. The Platte Valley is located along the southern part of the study area. Platte Valley soils typically are sandy and land cover is a mixture of pasture and rowcrops.
More than 150 surface-water samples were analyzed for these compounds during 1992-94 (see map). The symbol color on the map shows whether samples were collected during low-flow conditions (magenta, typically late summer), during spring runoff after herbicides had been applied (red), or on a monthly schedule throughout the study (blue). Bar length on the map indicates the percentage of samples in which any of the herbicides exceeded an HAL or MCL.
Twenty-eight surface-water sites were sampled for organonitrogen herbicides; 16 of these were sampled only during low-flow conditions. Only one low-flow organonitrogen herbicide sample exceeded a HAL or MCL. That sample was collected from Silver Creek, a small stream in the Platte Valley, and the atrazine concentration was 4.3 ,ug/L. Twelve sites were sampled during spring runoff. No samples from the Tri-County Supply Canal near North Platte or the Dismal River exceeded the HAL or MCL for organonitrogen herbicides. Those sites drain areas with little rowcrop production. Three sites on the Loup River system and the Platte River near Grand Island represent mixtures of rangeland and cropland. Cropland areas in these basins generally are located in river valleys near the sampling location. Spring-runoff samples from those sites exceeded the HAL or MCL for organonitrogen herbicides between 17 and 86 percent of the time. Those same levels of organonitrogen herbicides were exceeded in 87 to 100 percent of the spring-runoff samples from sites draining rowcrop dominated areas. Nine sites representing a variety of land uses were sampled on a monthly basis and included a complete range of streamflows. The percentage of organonitrogen herbicide samples that exceeded the HAL or MCL ranged from 0 percent in rangeland areas to 53 percent in rowcrop areas. Metolachlor was the only organonitrogen herbicide that did not exceed the HAL or MCL in any sample.
Ground-water samples for the determination of organonitrogen herbicide concentrations were collected from 66 monitoring and supply wells in the Platte Valley during 1994-95. Wells were located from near North Platte to near Louisville, but mostly were clustered between Kearney and Grand Island. Shallow wells drawing water from the alluvial aquifer in the Platte Valley were targeted for sampling because shallow ground water is the most vulnerable to contamination by agricultural chemicals. None of the 142 ground-water samples contained herbicides at concentrations exceeding a HAL or MCL for drinking water. Alachlor, cyanazine, and metolachlor were not present above the detection level of 0.05 1lg/L. The largest atrazine concentration was 1.1 µg/L and 86 percent of the samples contained atrazine at concentrations less than 0.05, µg/L.
Four ecologically significant organochlorine residues were present in fish tissues at sites sampled during 1992. The frequency of detection of these compounds was smallest in the Sandhills and largest in the Platte Valley (see map). Trans-nonachlor, p, p'-DDE, and dieldrin typically are largest in agricultural areas, whereas PCBs, which also had some agricultural use, are more commonly associated with industrial areas. These compounds are considered hydrophobic in that they are poorly soluble in water yet have an affinity for lipids. Nationally, organochlorine residues have a trend of decreasing concentrations in whole fish sampled from 1976 to 1984 for the National Contaminant Biomonitoring Program of the U.S. Fish and Wildlife Service.
Fish from four sites were sampled from 1992-94 and revealed considerable annual variation in concentrations of these contaminants (see graphs; note that vertical axes are logarithmic so small vertical shifts in plotted symbols reflect large changes in concentration). The most upstream site sampled on the Platte River was at Brady, Nebraska which drains mostly rangeland. The amount of cropland in each basin increases from the Platte River at Brady, to Grand Island, and to Louisville. Land use in the Elkhorn River drainage is predominantly cropland.
DDT was a broadly applied pesticide until its use was banned in the United States in 1972. The most environmentally persistent metabolite of DDT is p, p'-DDE. The presence of p, p'-DDE in fish tissues in the study area shows some relation to streamflow (see graphs). Flooding can increase contaminant transport from watersheds to streams as well as resuspend previously deposited sediments within the stream. Increased contaminant levels in streams would be reflected by increased concentrations of those contaminants in the tissues of resident fish. In 1993, streamflow and p,p'-DDE concentrations in fish tissues from the Platte River (near Grand Island and near Louisville) and from the Elkhorn River near Waterloo were substantially larger than in 1992 (see charts). During the same period, p, p'-DDE was at or less than detection levels at the Platte River at Brady site where streamflows generally were less than normal (normal streamflow shown as the shaded area on the charts). Concentration of p, p'-DDE was largest (1,600 µg/kg) in fish collected from the Platte River near Louisville following extremely high streamflows in 1993. For much of the sampling period at the Elkhorn River, streamflow was well above normal levels and p, p'-DDE concentrations in fish showed an increasing trend. In contrast, p, p'-DDE was not detected in ground-water samples collected from the Platte Valley, and was detected in only 1 of 28 streambed-sediment samples and 3 of 159 surface-water samples collected across the study area. These data demonstrate the ability of organisms to concentrate hydrophobic contaminants to levels many times greater than their surrounding environment, as well as the effectiveness of tissue sampling in detecting contaminant occurrence.
Total PCB concentration in fish tissue was greater from streams draining larger areas and integrating different land uses. The largest concentrations were observed from samples collected at Platte River sites and the farthest downstream site on the Elkhorn River. Trans-nonachlor is the most environmentally persistent component of chlordane. Although chlordane had past agricultural uses, its presence in the environment often is associated with urban areas because of recent residential uses. The largest transnonachlor was in a sample collected from the Wood River downstream from the city of Grand Island.
Dieldrin, a metabolite of the insecticide aldrin as well as an insecticide itself, was banned in the early 1970s. Dieldrin was more abundant in fish-tissue samples from streams that drained areas with large amounts of rowcrops. Three streams selected to represent areas dominated by rowcrop production were included in the network of streams sampled during the study. Two of those streams did not have a sufficient fish population for collection of a tissue sample. The largest concentrations of dieldrin in fish tissues were at Maple Creek, the third site indicative of intense rowcrop production, and the nearby Elkhorn River site. As with p, p'-DDE, concentrations of PCBs, trans-nonachlor, and dieldrin were rarely detected in sampling media other than fish tissues. At the four sites sampled from 1992-94, there was not a discernable relation between trans-nonachlor, PCBs, or dieldrin concentrations in fish tissue and streamflow.
Comparisons of selected trace-element concentrations determined during 1992-95 in surface water, ground water, streambed sediments finer than 0.063 millimeters, and fish livers are shown using boxplots (see graphs). Boxplots show the statistical distribution of data values such that a box, called the interquartile range (IQR), contains the middle 50 percent of the values. Values considered to be outliers fall in the range of 1.5 to 3.0 times the IQR beyond the boundaries of the box. Extreme values are more than 3.0 times the IQR from the boundaries of the box. The number of data points for each boxplot is denoted by "n".
Differences in concentrations between the media reflect the affinity of trace elements for sediments and the biomagnification of trace elements by fish. Livers were sampled because trace elements tend to be most concentrated in that organ. Comparisons of data collected in the Central Nebraska Basins from 1992-95 with the National Contaminant Biomonitoring Program data and USEPA's standards for the protection of fish-eating wildlife are not possible because those values refer to concentrations in whole fish.
Cadmium and lead in surface and ground water were at or less than the detection level of 1.0 µg/L and less than the drinking-water MCLs of 5.0 and 15 µg/L, respectively. Cadmium was most concentrated in fish livers and lead was most concentrated in streambed sediments. Lead concentration was nearly 100 times greater in the sediments (median concentration 18,000 µg/kg) than in fish livers (median concentration 200 µg/kg). Arsenic and selenium concentrations ranged from the detection level of 1.0 µg/L to over 10 µg/L in surface- and ground-water samples; the drinking-water MCL for these elements is 50 µg/L . Arsenic was most concentrated in the sediments whereas selenium was most concentrated in fish livers. Mercury was not analyzed in water samples and was detected in only 2 of 13 streambed-sediment samples. The maximum concentration of mercury in sediments was 30 µg/kg. Mercury concentration in fish livers exceeded the detection level of 100 µg/kg in 14 of 19 samples. The largest mercury concentration observed was 350 µg/kg. Nationally, concentrations of arsenic, cadmium, lead, and selenium in whole fish decreased from 1976 to 1984, whereas mercury concentrations remained essentially unchanged.
Volatile organic compounds (VOCs) generally are associated with industrial activities although some also are used as inert ingredients in pesticide formulations. Because many VOCs are carcinogenic, their occurrence in water supplies is of concern. There were no detections for any of more than 60 different VOCs that were analyzed in each of four surface-water and eight ground-water samples collected from sites in the Platte Valley during 1995. These results suggest that VOC contamination of surface and ground water in the Platte Valley from industrial and agricultural sources was not evident.
Five surface-water and 11 ground water samples were collected from the Platte Valley in 1995 to test for presence of alpha and beta radioactivity. Alpha radioactivity generally indicates the presence of naturally occurring radium or radon and has a MCL of 15 pCi/L (picocuries per liter) in drinking water. Alpha radioactivity ranged from 9.9 to 18 pCi/L in surface water and from 1.5 to 79 pCi/L in ground water. The largest concentrations in surface water were from a diversion canal near North Platte, although similar concentrations were observed from the Platte River near Grand Island. Beta radioactivity, where present above 50 pCi/L, may indicate the presence of fission products such as potassium-40 or strontium-90. Beta radioactivity ranged from 11 to 35 pCi/L in surface water and from 5.2 to 55 pCi/L in ground water. The North Platte River Basin, which is upstream from the Central Nebraska Basins, contains areas of uranium-rich rock that is a likely source of alpha and beta radioactivity.
Callam, Michael, Lund, John, Bubb, Dave, and Schumacher, Dave, 1994, Findings of the 1993 ambient fish tissue and follow-up programs in Nebraska: State of Nebraska Department of Environmental Quality, Surface Water Section, Water Quality Division, unpaginated.
Emmons, P.J., 1996, Water quality in a wet meadow, Platte River Valley, central Nebraska: U.S. Geological Survey Fact Sheet FS-097-96.
Exner, M.E. and Spalding, R.F., 1990, Occurrence of pesticides and nitrate in Nebraska ground water: Lincoln, University of Nebraska, Water Center, 39p.
Huntzinger, T.L. and Ellis, M.J., 1993, Central Nebraska River Basins, Nebraska: Water Resources Bulletin, vol. 29, no. 4, p. 533-574.
Schmitt, C.J. and Brumbaugh, W.G., 1990, National contaminant biomonitoring program-- Concentrations of arsenic, cadmium, copper, lead, mercury, selenium, and zinc in U.S. freshwater fish, 1976-84: Archives of Environmental Contamination and Toxicology, vol. 19, p. 731-747.
Schmitt, C.J., Zajicek, J.L., and Peterman, P.H., 1990, National contaminant biomonitoring program-- Residues of organochlorine chemicals in U.S. freshwater fish, 1976-1984: Archives of Environmental Contamination and Toxicology, vol. 19, p. 748-781.
Stamer, J.K., and Wieczorek, M.E., 1995, Pesticides in streams in central Nebraska: U.S. Geological Survey Fact Sheet FS-232-95.
Zelt, R.B. and Jordan, P.R., 1993, Water-quality assessment of the central Nebraska basins-- Summary of data for recent conditions through 1990: U.S. Geological Survey Open-File Report 93422, 179 p.