Radiation Dosimetry on Revegetated Uranium Mill Tailings in Western South Dakota

Measurement of gamma radiation using thermoluminescent dosimeters on three uranium mill tailings areas and a control area showed exposure rates below ground depended on the amount and type of soil covering. Covering tailings with 30 cm of shale and 60 cm of topsoil reduced gamma radiation exposure belowground to rates similar to the control area. Soil covering of 30 cm was adequate to reduce gamma radiation exposure rates aboveground to background levels. Gamma exposure rates to small mammals were correlated to measurements belowground near the surface. Gamma radiation exposure to small mammals inhabiting uranium mill tailings exceeded standards recommended for the general public. The waste materials from uranium milling processes (tailings) contain low amounts of radioactive materials in the U-238 decay series. To assess the health risks from these radioactive materials, it is necessary to determine the amounts of nuclear waste materials in soils and vegetation on these sites, and the radiation exposure to fauna, including humans. Studies have been conducted to evaluate the levels of radiation exposure to small mammals (Halford and Markham 1978, Gano 1979) and waterfowl (Halford et al. 1982) inhabiting spent nuclear fuel waste disposal sites. However, these sites contain nuclear reactor by-products, higher in radioactivity than uranium mill sites. Information pertaining to the radiation received by small mammals inhabiting uranium mill tailings sites is lacking. This study compares the levels of gamma radiation exposure to small mammals and their environment on uranium mill tailings to background levels on a control site. Study Area and Methods The uranium mill tailings immediately west of Edgemont, South Dakota were selected for this study. The mill has been inactive since 1972 and is scheduled for decommissioning. Four sites were selected for study. Sites 1 and 2 were uranium mill tailings covered with 30 cm of topsoil, with established vegetation dominated by crested wheatgrass (Agropyron cristatum) and yellow sweetclover (Melilotus officinalis) seeded in the fall of 1973 and spring of 1974, respectively. Silver King Mines, Inc., manages the site for the Tennessee Valley Authority. Site 2 differed from site 1 in that background gamma scan counts measured 1 m aboveground indicated a higher level of gamma radiation on the site. Site 3, also uranium mill tailings, was covered with 30 cm of shale and 60 cm of topsoil with the dominant species of vegetation being crested wheatgrass, western wheatgrass (Agropyron smithii), and yellow sweetclover all seeded in 1978. The control site was a pasture located west of the mill, dominated by crested wheatgrass, red threeawn (Aristada longiseta), and sand dropseed (Sporobolus cryptandrus). On each site, a grid of 77 (7 x 11) Sherman collapsible small mammal live-traps (23 x 9 x 8 cm) spaced at 7 m intervals, was established. Trapping was conducted monthly from June through September of 1981 and April through September of 1982. The goal was to catch as m a n y s m a l l m a m m a l s a s p o s s i b l e , a t t a c h dosimeters and recapture these animals at a later date. Because trapping success was poor on the control site during the first trapping period, an additional 30 traps were set adjacent to the control site. All traps were baited with rolled oats mixed with peanut butter and were opened five consecutive nights at monthly intervals. Animals captured on all sites were toe-clipped for identification and age, sex, weight, and species were recorded. During 1981 all animals captured were anesthetized and surgically implanted (subcutaneously) with a waterproof packet containing two, 0.32 x 0.32 cm thermoluminescent dosimeters (TLD's) (Halford and Markham 1978, Gano 1979). A different technique was used during Northwest Science, Vol. 60, No. 3, 1986 145 1982; TLD’s were attached to both ears of the animals with two Monel metal tags. Each TLD chip was affixed to the ear tag by a piece of 0.3%cm heat-shrink tubing. The animals were then released at the capture location. Animals with TLD’s recaptured after 30 or more days had the old TLD’s replaced with new ones and the animal was released again. No new TLD’s were put on small mammals in September of 1981. Gamma radiation on the sites was measured by sets of TLD’s attached to wooden rods at 80, 60,40, and 20 cm belowground and at 0, 10,20, 40,60, and 80 cm aboveground. One set of TLD’s was placed at the center of each small mammal trap grid and one set of TLD’s was placed at the center of each quarter of the small mammal grid. These TLD’s were in the field 30 days. Exposed TLD’s were placed in lead containers while in the field and during storage. The TLD’s, which had been calibrated with a Ra-226 source were initially selected to be within t 10 percent error due to chip variability and error associated with the reading. All TLD’s exposed in the field and some TLD’s stored in a lead container in the laboratory were sent for analysis. Analyses were conducted by Eberline Laboratories.’ Exposure levels of TLD’s kept in the laboratory were subtracted from those exposed in the field to account for irradiation during shipping and handling. TLD exposure data are given in milliRoentgens per day (mR/day). Radiation exposure data from these TLD’s are only representative of gamma and some highenergy beta forms of radiation. Differential shielding by the techniques used for attaching TLD’s was not of consequence since most gamma radiation easily passes through up to 1.0 inches of iron (U.S. Department of Health, Education and Welfare 1970). Radiation exposure data on all treatments were analyzed separately for aboveground and belowground in a split-plot design (Steel and Torrie 1960). Sites were considered whole plot effects and depths were treated as subplot effects. Since the interaction of site by depth for the belowground analysis was significant (P < O.Ol), separate one-way analyses of variance were performed at each depth. Differences among exposure rates belowground for sites at each depth were evaluated using Dunnett’s T3 method for non-homogeneous variances (Dunnett 1980). Significant differences from the aboveground procedure were evaluated using Tukey’s multiple comparisons method for homogeneous variances (Steel and Torrie 1960). Radiation exposures to the small mammals were analyzed by two-way analysis of variance for species and sites. Tukey’s method was used to determine where differences occurred. Probability level for all tests was