Environ. Sci. Technol. 2011, 45, 392–399
Scaling Non-Point-Source Mercury Emissions from Two Active Industrial Gold Mines: Influential Variables and Annual Emission Estimates C . S . E C K L E Y , †,‡ M . G U S T I N , * ,† M. B. MILLER,† AND F. MARSIK§ Department of Natural Resources & Environmental Science, University of Nevada, Reno, Nevada 89557, United States, and Department of Atmospheric, Oceanic and Space Sciences, University of Michigan, 2455 Hayward Avenue, Ann Arbor, Michigan 48109-2143, United States
Received May 28, 2010. Revised manuscript received September 21, 2010. Accepted November 17, 2010.
Open-pit gold mines encompass thousands of hectares of disturbed materials that are often naturally enriched in mercury (Hg). The objective of this study was to estimate annual nonpoint-source Hg emissions from two active gold mines in Nevada. This was achieved by measuring diel and seasonally representative Hg fluxes from mesocosms of materials collected from each mine. These measurements provided a framework for scaling emissions over space and time at each mine by identifying the important variables correlated with Hg flux. The validity of these correlations was tested by comparisons with measurements conducted in situ at the mines. Of the average diel fluxes obtained in situ (92 daily flux measurements), 81% were within the 95% prediction limits of the regressions developed from the laboratory-derived data. Some surfaces at the mines could not be simulated in the laboratory setting (e.g., material actively leached by cyanide solution and tailings saturated with cyanide solution), and as such in situ data were applied for scaling. Based on the surface areas of the materials and environmental conditions at the mines during the year of study, non-point-source Hg releases were estimated to be 19 and 109 kg · year-1. These account for 56% and 14%, respectively, of the overall emissions from each mine (point + nonpoint sources). Material being heapleached and active tailings impoundments were the major contributors to the releases (>60% combined) suggesting that as mining operations cease, releases will decline.
Introduction Mercury (Hg) released to the air can be transported and deposited to aquatic systems, converted to methylmercury, and accumulated in fish (1). Because anthropogenic releases of Hg are an important component of the atmospheric Hg pool (2), it is important to accurately identify the magnitude of anthropogenic Hg emissions. Currently, the focus of * Corresponding author. † University of Nevada. ‡ Present address: Canadian Government Laboratory, Environment Canada, 201-401 Burrard St., Vancouver, British Columbia V6C 3S5. § University of Michigan. 392
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 45, NO. 2, 2011
regulatory agencies has been on reducing Hg releases from large point-source emitters such as coal-fired power plants and chlor-alkali facilities (3). There is much less known about non-point-source emissions associated with anthropogenic activities (4). Industrial gold mines have been shown to have significant point-source releases of Hg to the atmosphere associated with ore processing facilities (5, 6). The state of Nevada is one of the largest gold-producing regions in the world with ∼20 operating mines releasing approximately 2000 kg of Hg · year-1 as stack emissions (e.g., autoclaves, roasters, etc) (6). The Hg emitted is naturally associated with the ore materials. Since large amounts of subterranean material are extracted and distributed over the mine surface, there is the potential for significant non-point-source emissions because Hg can volatize from substrate under ambient conditions (7). Thus, it is important that these emissions be quantified to understand the total (point + nonpoint source) emissions from active gold-mining operations. Previous work has shown that Hg fluxes from geogenic materials are influenced by material Hg concentrations and environmental conditions (7-9). Meteorological variables that have been found to be important include solar radiation, temperature, precipitation (and the resulting effect on soil moisture), and atmospheric turbulence (7-9). Disturbance of substrate has also been found to facilitate Hg release (9, 10). To quantify non-point-source emissions from active industrial gold mines, flux must be understood within the context of these parameters. The objective of this study was to determine the annual non-point-source Hg emissions from two active Nevada (USA) gold mines. Although the two mines had different degrees of natural Hg enrichment, their overall geologic settings and ore-processing methods represent those ongoing within the state. Emission estimates were developed by applying derived relationships between flux and specific parameters within a Geographic Information System (GIS) framework. This allowed us to take into account the spatial and temporal variability of flux occurring at each mine over a year.
Materials and Methods Site and Materials Description. The two mines selected for this project are located in central Nevada: Twin Creeks mine (41°15′ N, 117°9′ W, 1560 m), operated by Newmont Mining Corporation, and Cortez-Pipeline mine (40°4′ N, 116°42′ W, 1570 m), operated by Barrick Gold Corporation. Both have a diversity of ore types (oxide ore as well as refractory ore that has high carbon and/or sulfur content), material processing methods (heap leach, autoclave, carbon-in-leach), and age of surface materials (active, inactive, and reclaimed) (see Tables S1-S3 in Supporting Information for additional geologic information). The major surfaces at each mine include the open pit(s), waste rock dumps, leach pads, ore stockpiles, tailings impoundments, and reclaimed materials. Blasting of rock within active open-pit mines occurs almost daily. The fragmented rock is then transported to specific locations depending on gold content. Waste rock dumps consist of material that does not contain economical gold concentrations. Materials in heap leach piles consist of low-grade ore (i.e., a low concentration of economical gold). These are irrigated with a dilute cyanide solution to extract the gold. At any given time, there are leach pads being constructed (preleach materials), those where cyanide is being applied (active leach materials), and areas that are no longer actively 10.1021/es101820q
2011 American Chemical Society
Published on Web 12/10/2010
leached (postleach). High-grade stockpiles consist of ore with high-enough gold concentrations that it is economical for processing in the mill, where it is crushed into a fine powder. After crushing, high-grade oxide ore is leached in vats (carbon-in-leach), whereas refractory ore is processed in either a roaster or autoclave to liberate the gold prior to vat leaching. After processing of the high-grade materials, they are transported in cyanide-containing solution to the tailings impoundments. Over the life of the mine, surfaces are reclaimed or restored to resemble the natural landscape. Reclamation can include contouring surfaces or capping a material such as tailings with soil and subsequent seeding or adding of vegetation. Minor mine surfaces not included in our study include the surfaces of roads used to travel between mining areas, areas for parking mining vehicles/ equipment, and leach solution storage ponds, which together made up 0.08 for intercepts and >0.4 for slopes). This indicates that the relationship between substrate Hg concentration, environmental conditions, and flux are similar between the two mines. 394
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 45, NO. 2, 2011
The organization of the Hg substrate-flux regressions by solar level is not meant to imply that solar radiation alone was affecting emissions from dry materials but that changes in solar radiation were the best representative of the complex meteorological interactions affecting the Hg fluxes from dry materials on diel and seasonal time steps. Similarly, the material Hg concentration is most likely not the only substrate characteristic that affects the amount of Hg released. For example, the Cortez-Pipeline carbonaceous ore had a very high Hg concentration (e.g., 22-86 µg · g-1), but relatively low flux (
