The Garrison Diversion Unit is being constructed to transfer water from the Missouri River (Lake Sakakawea) to areas in east-central and southeastern North Dakota for expanded irrigation of agricultural lands. During initial investigations of irrigation return flows in 1969-76, the potential effects of toxic elements were considered, and the U.S. Bureau of Reclamation concluded these elements would have no adverse effects on streams receiving return flows. After the development of problems associated with selenium in irrigation return flows in the western San Joaquin Valley, Calif., in 1985, the U.S. Bureau of Reclamation initiated additional studies, including an investigation conducted in cooperation with the U.S. Geological Survey, to assist in collecting and evaluating trace-element data. Also, in 1986, with the passage of the Garrison Diversion Unit Reformulation Act, Congress mandated that soil surveys be conducted to determine if there are "*** soil characteristics which might result in toxic or hazardous irrigation return flows."

In order to address this issue, an investigation was conducted during 1995-87 by the U.S. Geological Survey in cooperation with the U.S. Bureau of Reclamation to determine the occurrence and distribution of arsenic, selenium, and other trace elements in the soils of six potential irrigation areas along the Garrison Diversion Unit route and in the James River basin. A total of 165 soil samples were collected and analyzed for total concentrations of as many as 42 elements, including arsenic and selenium. In addition, 81 of the samples were analyzed for water-extractable concentrations of 14 elements, including arsenic and selenium, to aid in determining the extent to which they might be mobilized by the irrigation water. In a detailed phase of the investigation, 376 water samples were collected in one of the six potential irrigation areas, the west Oakes irrigation area. Most of these samples were analyzed for arsenic, selenium, and as many as 28 other elements.

Results of the investigation indicate that soils in the potential irrigation areas contain small concentrations of arsenic, selenium, and other trace elements. The geometric mean concentrations of total arsenic and selenium were 4.15 and 0.13 milligrams per kilogram, respectively, which are considerably smaller than those measured in the western San Joaquin Valley, Calif., and soils from other areas in the western United States. Water-extractable concentrations of arsenic and selenium, determined on 1:5 soil to water extractions, generally were less than 10 percent of the total concentrations. The geometric mean water-extractable concentrations for both elements were 0.02 milligram per kilogram or less.

The median and maximum concentrations of all constituents and properties indicative of irrigation drainage were tens to hundreds of times smaller in the Oakes test area drains than in western San Joaquin Valley drains. The maximum arsenic concentration in ground-water samples was 44 micrograms per liter, and the median concentration was 4 micrograms per liter. The maximum concentration in drain samples was 11 micrograms per liter, and the median concentration was 3 micrograms per liter.

Only 22 percent of the water samples collected from wells in the Oakes test area contained detectable concentrations (1 microgram per liter or more) of selenium. However, selenium was detected in 63 percent of the samples collected from sites on drains. The greater incidence of detection of selenium in the drain samples is interpreted as an effect of the more oxidizing environment of the drains, which are about 8 feet below land surface near the top of the water table. The median selenium concentration in the drain samples, however, was only 1 microgram per liter, and the maximum concentration in 63 drain samples was 4 micrograms per liter. For comparison, the median selenium concentrations reported for drains in the western San Joaquin Valley, Calif., ranged from 84 to 320 micrograms per liter. Mater from two observation wells had the largest selenium concentrations (8 and 9 micrograms per liter) measured during the investigation. These were the only two samples that exceeded any of the water-quality regulations, standards, or criteria for selenium.

Mercury and boron were the only other trace elements that exceeded standards and criteria. The median concentration of mercury was less than 0.1 microgram per liter, and the maximum concentration was 0.8 microgram per liter. The chronic freshwater-aquatic-life criterion for mercury (0.012 microgram per liter) is about 10 times less than the laboratory detection limit and is derived from bioconcentration factors based on methylmercury. Two boron samples exceeded the irrigation criteria of 750 micrograms per liter. Comparisons with criteria and standards indicate that the concentrations of trace elements determined in samples from wells and drains in the Oakes test area during this investigation should not adversely affect human and aquatic life or irrigated crops.

The data collected indicate that the soils and ground water in the Garrison Diversion Unit contain small concentrations of trace elements, including arsenic and selenium. Based on a detailed study of soils and ground water in the west Oakes irrigation area, however, there is no evidence that expanded irrigation will mobilize these elements in concentrations large enough to adversely affect aquatic life in the James River ecosystem, based on current regulations, standards, and criteria. Data are not currently available to make definitive statements about selenium concentrations in ground water in Garrison Diversion Unit irrigation areas other than the west Oakes Irrigation area. Data available on total and water-extractable selenium concentrations in soils t however, indicate that concentrations in ground water would be similar to those determined in the west Oakes irrigation area. Plans have been developed to sample ground water in the additional areas.