A case study of cost effective detailed waste characterisation techniques for closure of abandoned uranium mines

摘要

Mining activities that occurred during the 1950s and early 1960s at the North Cave Hills in Harding County, South Dakota, are a significant part of South Dakota's uranium mining history. Under the General Mining Laws, per the Atomic Energy Act of 1946 and Public Law 357, unrestricted strip mining took place within the North Cave Hills during this time. More than 1 square kilometre of mine-affected land has been approved by the US Forest Service and the US Environmental Protection Agency as a non-time critical removal action under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) in 2007. Cleanup levels for surface soil are established for total arsenic and radium-226 based on a site-specific risk assessment. While numerous scientific studies have been conducted at the site, none of these has characterised the site at the level of detail achieved using the sampling methods presented here. This case study presents the methodology and results of two different sampling methods used to independently characterise the distribution of heavy metals and radionuclide soil concentrations at an abandoned uranium mine. Statistical evaluations used in sizing sampling grids, evaluation of the feasibility of utilising double sampling methods versus simple random sampling, geostatistical mapping to establish cleanup boundaries, and an economic evaluation of the final methods selected are presented in this paper. A site-wide X-ray fluorescence (XRF) field survey involving collection of more than 800 measurements was conducted on a defined square sampling grid pattern. A strong correlation was established (r = 0.95) between in situ XRF arsenic measurements and laboratory-reported total arsenic concentrations in surface soils. This correlation was the basis for development of a site-specific regression model used in estimating total arsenic concentrations in surface soils. Additionally, a continuous mobile gamma radiation survey was conducted on a defined square sampling grid pattern. A strong correlation (r = 0.96) was established between gamma exposure rate and radium-226 concentrations in surface soils. This correlation was the basis for development of a site-specific regression model used in estimating the radium-226 concentrations in surface soils. More than 60,000 gamma exposure rate measurements were collected and used in the estimation of radium-226 concentration within 5,988 block-averaged 100-square-meter (m~2) grids. Advanced geospatial analysis techniques were applied to the arsenic and radium-226 data sets, and cleanup boundaries were estimated. The surficial extent of soils with arsenic and radium-226 concentrations above the cleanup levels were estimated at 14 percent (87,817 m~2) for arsenic and 2.7 percent (16,682 m~2) for radium-226 of the total study area. An economic analysis compared the incurred cost of the case study with conventional soil sampling methods that would achieve the same data quality objectives. A project savings factor of 4.4 was calculated for the XRF field survey and 84 for the gamma survey. The data and information collected during this study were used to develop site-wide removal action designs that will be implemented as part of the mine closure process. It is concluded that there is significant economic benefit in using the sampling approaches presented in this report to characterise mine waste.