Tailings Site Research Articles

English

The Keating Tailings site is located in Broadwater County, Montana on land administered by the Bureau of Land Management. These low pH (4 standard units) wastes resulting from historic gold and copper mining operations contain phytotoxic levels of several metals and are generally devoid of vegetation. With an estimated volume of 110,100 m 3 , these tailings represent unacceptable risk to the environment and human health. The objective of conducting a phytostabilization study at the Keating Tailings Site was to provide BLM managers and decision makers with site specific information and data relating to the implementation, and effectiveness of phytostabilization so that it may be applied to other similar acid metalliferous mine tailings sites administered by the Bureau. To achieve this management objective, replicated experimental plots were implemented using soil amendments, lime and organic matter, designed to ameliorate the plant inhibiting chemical characteristics of the tailings. The plots were seeded with a mix of indigenous native plant species. Vegetation performance of plants grown in the amended or phytostabilized tailings was compared to results for plants seeded into tailings that were not amended, and performance of plants seeded in an adjacent off-site, but non-impacted area. Response variables evaluated in the first growing season, 2004, included emergence and establishment, density, and canopy cover. Concentrations of metals in vegetation were evaluated in terms of plant sufficiency/excess, and in terms of maximum allowable dietary levels for cattle. Changes in soil rootzone pH, conductivity, and soluble metal concentrations before and after treatment were also determined.

Difficulties in radon measurements at workplaces

Different legislation systems can be found in the world concerning radon levels at workplaces. Following the European Union suggestion, a reference level for radon concentration in the air at workplaces was established in several European countries. In Hungary, the relevant legislation has come into effect on 1 January 2003. The determination of average radon concentration might present a problem, especially in places where the monthly average concentrations vary to a great extent. For example, the monthly averages measured in a hospital cave used for treating respiratory diseases showed a 24-fold difference depending on the chosen month. In such cases, attention should be paid when choosing the months and using the results of measurements for dose assessment. Another uncertainty emerges when estimating the annual dose, based on the data coming from long-term measurements, usually using integrated methods such as track detectors. There is a considerable difference between the averages measured during the working hours and over the total time (including nights and weekends), mostly in the cases of rooms with frequent air change like schools, kindergartens and ventilated workplaces. This can lead to a significant overestimation in dose calculation. Special attention needs to be paid to workplaces such as mines, tunnels and open air uranium tailings sites. This paper discusses the possible inaccuracies caused by the improper selection of time periods and methods in the measurements of the average radon concentration at workplaces.

Geostatistical and local cluster analysis of high resolution hyperspectral imagery for detection of anomalies

This paper describes a new methodology to detect small anomalies in high resolution hyperspectral imagery, which involves successively: (1) a multivariate statistical analysis (principal component analysis, PCA) of all spectral bands; (2) a geostatistical filtering of noise and regional background in the first principal components using factorial kriging; and finally (3) the computation of a local indicator of spatial autocorrelation to detect local clusters of high or low reflectance values and anomalies. The approach is illustrated using 1 m resolution data collected in and near northeastern Yellowstone National Park. Ground validation data for tarps and for disturbed soils on mine tailings demonstrate the ability of the filtering procedure to reduce the proportion of false alarms (i.e., pixels wrongly classified as target), and its robustness under low signal to noise ratios. In almost all scenarios, the proposed approach outperforms traditional anomaly detectors (i.e., RX detector which computes the Mahalanobis distance between the vector of spectral values and the vector of global means), and fewer false alarms are obtained when using a novel statistic S 2 (average absolute deviation of p-values from 0.5 through all spectral bands) to summarize information across bands. Image degradation through addition of noise or reduction of spectral resolution tends to blur the detection of anomalies, increasing false alarms, in particular for the identification of the least pure pixels. Results from a mine tailings site demonstrate the approach performs reasonably well for highly complex landscape with multiple targets of various sizes and shapes. By leveraging both spectral and spatial information, the technique requires little or no input from the user, and hence can be readily automated.

Molecular Mechanism of Prothrombin G20210A Variant: Critical New Role of Exon Splicing Enhancer in Poly (A) Tailing.

Although prothrombin G20210A variant was first found in 1996 as a genetic risk factor of thrombosis, the underlying molecular mechanism for elevation of circulatory prothrombin levels due to the variant has remained elusive. The position 20210 is at the poly (A) tailing site, and several research groups reported different, contradictive possibilities including enhancement in poly (A) tailing, change in RNA stability or no difference in prothrombin mRNA level between G (wild type) and A (variant) at position 20210. Such confusions appear to be due to no reliable assay model availability. To overcome these problems, we developed a minigene assay system composed of only autologous components of human prothrombin gene, including its promoter, coding regions, introns and 3′UTR with poly (A) signal.A minigene containing the first intron and the last intron did not express at any appreciable levels, failing proper splicing of the last intron. Inclusion of upstream other introns was found essential for its appropriate splicing. Although it is generally accepted that the last intron enhances the efficiency of poly (A) tailing, we found that it is not the case for the prothrombin gene. Instead, an exon-splicing enhancer (ESE) of another upstream intron was found to be responsible for efficient poly (A) tailing. ESE, therefore, plays a critical role not only in alternative splicing but also in constitutive RNA processing, including poly (A) tailing. Etiology of prothrombin G20210A variant studied with culture cells and transgenic mice, supported an enhancement of poly (A) tailing, and not a change in mRNA stability.

Indicators of Microbial Sulfate Reduction in Acidic Sulfide-Rich Mine Tailings

Sulfate-reducing bacteria (SRB) are thought to be actively involved in the cycling of sulfur in acidic mine tailings. However, most studies have used circumstantial evidence to assess microbial sulfate activity in such environments. In order to fully ascertain the role of sulfate-reducing bacteria (SRB) in sulfur cycling in acidic mine tailings, we measured sulfate reduction rates, sulfur isotopic composition of reduced sulfide fractions, porewaters and solid-phase geochemistry and SRB populations in four different Cu-Zn tailings located in Timmins, Ontario, Canada. The tailings were sampled in the summer and in the spring, shortly after snowmelt. The results first indicate that all four sites showed very high sulfate reduction rates in the summer (∼100–1000 nmol cm− 3d−1), which corresponded to the presence of sulfide in the porewaters and to high SRB populations. In some of the sites, zones of microbial sulfate reduction also corresponded to a decline of organic carbon and to an apparent pyrite (with slightly negative δ34S values) enrichment around the same depth. Microbial sulfate reduction was also important in permanently acidic (pH 2–3) mine tailings sites, suggesting that SRB can be active under very acidic conditions. Secondly, the results showed that microbial sulfate reduction was greatly reduced in the spring, suggesting that temperature might be a key factor in the activity of SRB. However, a closer look at the results indicated that temperature was not the sole factor and that acidic conditions and limited substrate availability in the spring appeared to be important as well in limiting microbial sulfate par reduction in sulfidic mine tailings. Finally, the results indicate that sulfur undergoes rapid cycling throughout the year and that microbial sulfate reduction and metal sulfide precipitation do not appear to be a permanent sink for metals.

Comparison of in situ uranium KD values with a laboratory determined surface complexation model

Reactive solute transport simulations in groundwater require a large number of parameters to describe hydrologic and chemical reaction processes. Appropriate methods for determining chemical reaction parameters required for reactive solute transport simulations are still under investigation. This work compares U(VI) distribution coefficients (i.e. K D values) measured under field conditions with K D values calculated from a surface complexation model developed in the laboratory. Field studies were conducted in an alluvial aquifer at a former U mill tailings site near the town of Naturita, CO, USA, by suspending approximately 10 g samples of Naturita aquifer background sediments (NABS) in 17-5.1-cm diameter wells for periods of 3 to 15 months. Adsorbed U(VI) on these samples was determined by extraction with a pH 9.45 NaHCO 3/Na 2CO 3 solution. In wells where the chemical conditions in groundwater were nearly constant, adsorbed U concentrations for samples taken after 3 months of exposure to groundwater were indistinguishable from samples taken after 15 months. Measured in situ K D values calculated from the measurements of adsorbed and dissolved U(VI) ranged from 0.50 to 10.6 mL/g and the K D values decreased with increasing groundwater alkalinity, consistent with increased formation of soluble U(VI)-carbonate complexes at higher alkalinities. The in situ K D values were compared with K D values predicted from a surface complexation model (SCM) developed under laboratory conditions in a separate study. A good agreement between the predicted and measured in situ K D values was observed. The demonstration that the laboratory derived SCM can predict U(VI) adsorption in the field provides a critical independent test of a submodel used in a reactive transport model.

Comparison of in situ uranium KD values with a laboratory determined surface complexation model

Reactive solute transport simulations in groundwater require a large number of parameters to describe hydrologic and chemical reaction processes. Appropriate methods for determining chemical reaction parameters required for reactive solute transport simulations are still under investigation. This work compares U(VI) distribution coefficients (i.e. KD values) measured under field conditions with KD values calculated from a surface complexation model developed in the laboratory. Field studies were conducted in an alluvial aquifer at a former U mill tailings site near the town of Naturita, CO, USA, by suspending approximately 10 g samples of Naturita aquifer background sediments (NABS) in 17-5.1-cm diameter wells for periods of 3 to 15 months. Adsorbed U(VI) on these samples was determined by extraction with a pH 9.45 NaHCO3/Na2CO3 solution. In wells where the chemical conditions in groundwater were nearly constant, adsorbed U concentrations for samples taken after 3 months of exposure to groundwater were indistinguishable from samples taken after 15 months. Measured in situ KD values calculated from the measurements of adsorbed and dissolved U(VI) ranged from 0.50 to 10.6 mL/g and the KD values decreased with increasing groundwater alkalinity, consistent with increased formation of soluble U(VI)-carbonate complexes at higher alkalinities. The in situ KD values were compared with KD values predicted from a surface complexation model (SCM) developed under laboratory conditions in a separate study. A good agreement between the predicted and measured in situ KD values was observed. The demonstration that the laboratory derived SCM can predict U(VI) adsorption in the field provides a critical independent test of a submodel used in a reactive transport model.

Outdoor Radon Concentration at a Uranium Tailings Site and Its Occupational Radiation Dose

Outdoor Radon Concentration at a Uranium Tailings Site and Its Occupational Radiation Dose

Methods for estimating adsorbed uranium(VI) and distribution coefficients of contaminated sediments.

Assessing the quantity of U(VI) that participates in sorption/desorption processes in a contaminated aquifer is an important task when investigating U migration behavior. U-contaminated aquifer sediments were obtained from 16 different locations at a former U mill tailings site at Naturita, CO (U.S.A.) and were extracted with an artificial groundwater, a high pH sodium bicarbonate solution, hydroxylamine hydrochloride solution, and concentrated nitric acid. With an isotopic exchange method, both a KD value for the specific experimental conditions as well as the total exchangeable mass of U(VI) was determined. Except for one sample, KD values determined by isotopic exchange with U-contaminated sediments that were in equilibrium with atmospheric CO2 agreed within a factor of 2 with KD values predicted from a nonelectrostatic surface complexation model (NEM) developed from U(VI) adsorption experiments with uncontaminated sediments. The labile fraction of U(VI) and U extracted by the bicarbonate solution were highly correlated (r2 = 0.997), with a slope of 0.96 +/- 0.01. The proximity of the slope to one suggests that both methods likely access the same reservoir of U(VI) associated with the sediments. The results indicate that the bicarbonate extraction method is useful for estimating the mass of labile U(VI) in sediments that do not contain U(IV). In-situ KD values calculated from the measured labile U(VI) and the dissolved U(VI) in the Naturita alluvial aquifer agreed within a factor of 3 with in-situ KD values predicted with the NEM and groundwater chemistry at each well.

Copper and health status of cattle grazing high-molybdenum forage from a reclaimed mine tailing site

High concentrations (21–44 mg kg-1 dry matter) of Mo have been identified in forage in several reclaimed mining areas in British Columbia. Since Mo concentrations greater than 5 mg kg-1 in forage dry matter may result in molybdenosis because of a secondary Cu deficiency in ruminants, a study was undertaken to determine if cattle can safely graze the reclaimed land. For 12-wk grazing periods in 1994, 1995 and 1996, 32 cow/calf pairs grazed high-Mo forage at a reclaimed mine tailings site located at the Highland Valley Copper mine near Logan lake, BC. Half of the animals in the trial received a Cu supplement (All-Trace copper bolus) and the other half served as a control group. There were no significant differences (P < 0.05) in weight gain, liver Mo, serum Cu and Mo, and milk Cu and Mo between the two treatment groups of cows. Liver Cu was higher for the Cu bolus group at certain time periods in 1994 and 1995, indicating that the bolus was effective at supplying Cu. At all times, the liver Cu levels for the animals remained above the recommended critical level of 25 mg kg-1 dry matter. Animals appeared healthy and no signs of Cu deficiency were observed. Key words: Molybdenum, molybdenosis, copper, mine tailings, reclamation, animal health

Development of a functionalized polymer-coated silica for the removal of uranium from groundwater.

A new active material for the treatment of uranium-contaminated groundwater using permeable reactive barriers has been developed. This material, called PANSIL, is an example of a tailored ligand system that selectively removes a contaminant from solution. The active medium in PANSIL is a polyacryloamidoxime resin derived from polyacrylonitrile, which is deposited from solution onto the surface of quartz sand to form a thin film coating. PANSIL is highly effective at sequestering UO2(2+) from solution when the pH is between about 5 and 8 and can preferentially sequester UO2(2+) from solutions that are typical of the groundwater from a mine tailings site, due to the stability of the polyacryloamidoxime uranyl complex formed. Uranium sequestration capacity will depend on the surface area of the sand that is resin coated, but in the batch of PANSIL tested (<2% resin by weight), it exceeds 4000 mg of UO2 per kg of PANSIL at pH 4.5 when the dissolved UO2(2+) concentration is greaterthan 300 mg/L. PANSIL largely retains the permeability and strength of the sand employed and therefore has suitable engineering properties for permeable reactive barrier applications.

A probabilistic model and software tool for evaluating the long-term performance of landfill covers

A probabilistic model and software tool has been developed to assist in the selection, design, and monitoring of long-term covers for contaminated subsurface sites. The software platform contains multiple modules that can be used to simulate relevant features, events, and processes that include water flux through the cover, source-term release, vadose-zone transport, saturated-zone transport, gas transport, and exposure pathways. The component models are integrated into a probabilistic total-system performance-assessment model within a drag-and-drop software platform. Uncertainty and sensitivity analyses can be conducted that yield the following primary benefits: (1) quantification of uncertainty in the simulated performance metrics; (2) identification of parameters most important to performance; and (3) comparison of alternative designs to optimize cost and performance. A case study has been performed using the Monticello Mill Tailings Site in Utah to illustrate the important features and benefits of the modeling approach and software.

Tailings erosion and resuspension in two mine tailings ponds due to wind waves

One of the approaches used for environmental management of mine tailings is the implementation of shallow water covers, typically up to 2 m. As oxygen is one of the key factors in the oxidation of reactive iron-bearing sulfide minerals present in the tailings, water covers provide an effective control measure, owing to the low diffusivity and solubility of oxygen in water. However, wind-induced waves can resuspend flooded tailings, thus increasing the possibility of oxidation and acid generation. Summer field measurements of wind velocity and wave heights at two mine tailings sites in New Brunswick and Ontario are presented and discussed. Frequency diagrams of winds show that in summer wind speeds as high as 15 m/s are anticipated at these sites. Bottom shear stresses are calculated using measured wave heights and compared to critical shear stress of the tailings. Calculated bottom shear stresses exceeded the critical shear stress of the tailings at wind speeds higher than 9 m/s. Based on sediment trap data at the two sites, it is shown that resuspension is not a function of depth and wind-generated stresses alone, but also depends on other factors, such as cohesion and self-consolidation of tailings and site topography. Areas of erosion and accumulation were delineated at the two sites using particle size distribution, metal concentrations and an erosion–transportation–accumulation (ETA) diagram. The findings confirm resuspension work carried out by previous researchers at these two sites.

Characterization of microbial activities and U reduction in a shallow aquifer contaminated by uranium mill tailings.

A characterization of the Shiprock, NM, uranium mill tailing site focused on the geochemical and microbiological factors governing in-situ uranium-redox reactions. Groundwater and aqueous extracts of sediment samples contained a wide concentration range of sulfate, nitrate, and U(VI) with median values of 21.2 mM, 16.1 micro M, and 2.7 micro M, respectively. Iron(III) was not detected in groundwater, but a median value of 0.3 mM in sediment extracts was measured. Bacterial diversity down gradient from the disposal pile reflected the predominant geochemistry with relatively high numbers of sulfate- and nitrate-reducing microorganisms, and smaller numbers of acetogenic, methanogenic, nitrate-dependent Fe(II)-oxidizing, Fe(III)-reducing, and sulfide-oxidizing bacteria. In aquifer slurry incubations, nitrate reduction was always preferred and had a negative impact on sulfate-, Fe(III)-, and U-reduction rates. We also found that sulfate-reduction rates decreased sharply in the presence of clay, while Fe(III)-reduction increased with no clear impact on U reduction. In the absence of clay, iron and sulfate reduction correlated with concentrations of Fe(III) and sulfate, respectively. Rates of U(VI) loss did not correlate with the concentration of any electron acceptor. With the exception of Fe(III), electron donor amendment was largely unsuccessful in stimulating electron acceptor loss over a 2-week incubation period, suggesting that endogenous forms of organic matter were sufficient to support microbial activity. Our findings suggest that efforts to accelerate biological U reduction should initially focus on stimulating nitrate removal.

A case against Kd-based transport models: natural attenuation at a mill tailings site

This study compares numerical modeling results of contaminant transport using a multi-component coupled reactive mass transport model and a distribution coefficient ( K d)-based transport model. The study site is a contaminated groundwater aquifer underneath a uranium mill tailings pond in the western USA. Advective–dispersive–reactive transport is simulated for a 5-year period of intrusion of tailings fluid into the shallow aquifer, followed by flushing with uncontaminated upgradient groundwater for 1600 years. The coupled model shows that groundwater–sediment interactions result in multiple concentration waves, strong interactions among solutes, and chemical heterogeneity in both space and time. As a result, calculated K d values vary spatially and temporarily. None of these characteristics can be simulated with a K d-based model. These results illustrate the shortcomings of the K d approach, the usage of which is prevalent in the regulatory environment.

Probability of failure of waste disposals sites in Žirovski vrh uranium mine

The only Uranium mine in Slovenia Žirovski vrh was closed in 1990 due to economic reasons. After the closure extensive decommissioning works in the mine and in the surrounding began. In the very beginning after the closure great landslide has been occurred in the mill tailings site and recalculation of stability of existent and alternate sites were performed. In this calculations I used statistical scatter of input variables and calculated probability of failure of sites.

Spatial analysis of radiometric fractions from high-resolution multispectral imagery for modelling individual tree crown and forest canopy structure and health

Research was conducted in a forest adjacent to an abandoned acid mine tailings site to assess forest structural health using high spatial and spectral resolution digital camera imagery. Conventional approaches to this problem involve the use image spectral information, basic spectral transformations, or occasionally spatial transformations of image brightness. This research introduces fractional textures and semivariance analysis of image fractions. They were integrated with conventional image measures in stepwise multiple regression modelling of forest structure (canopy and crown closure, stem density, tree height, crown size) and health (a visual stress index). The goal was to conduct a relative comparison of the potential of the various image variable types in modelling of forest structure and health. Analysis was conducted for both canopy (crowns and shadows) and individual tree crown sample data sets extracted from 10 nm bandwidth spectral bands at three resolutions (0.25, 0.5, 1.0 m). Spatial transformations (texture, semivariogram range) of image brightness (DN) and image fractions (IF) were consistently the most significant and first entered variables in the best models of the forest parameters. At the canopy-scale, despite a limited number of available plots (6), stable models were produced that demonstrated the potential for spatially transformed variables. Semivariogram range explained 88% of the total variation of 9 of the 18 models and represented 56% of the variables used in all models while texture variables explained 51% of model variance in 8 of the 18 models and represented 40% of the variables used. At the tree crown scale ( n=31), 88% of the total variation of six of eight models was explained by texture variables and 6% by semivariogram variables. DN and IF variables that were not spatially transformed contributed little to the models at both scales. They represented 4% and 6%, respectively, of the variables used in all models. Spatial information in image fractions and image brightness has proven to be more significant than spectral information in these analyses. Of the spatial resolutions evaluated, 0.5 m consistently produced similar or better models than those using the 0.25 or 1.0 m resolutions. These results demonstrate the potential for integration of spatial transforms of image fractions and raw brightness in high-resolution modelling of forest structure and health.

Tin mining and processing in Nigeria: cause for concern?

The range and means of measured dose rates and activity concentrations of 40K, natural U and Th in environmental samples of a Nigerian tin mine and its tailing sites are presented. The measured concentration levels of eU and eTh in tailing samples and the measured dose rates were found to be elevated with values up to approximately 100 times above background levels of control soils. For the purpose of protection of workers and the public it is therefore advisable that these sites should be properly monitored, controlled and potential remediation strategies be considered in order to avoid any radiological impact to the population.

A mineralogical and geochemical study of element mobility in sulfide mine tailings of Fe oxide Cu–Au deposits from the Punta del Cobre belt, northern Chile

Two flotation tailings sites (Ojancos and P. Cerda) from the Fe oxide Cu–Au Punta del Cobre belt, south of Copiapó, Atacama desert, northern Chile, are geochemically (largely using sequential extractions) and mineralogically compared. Main ore minerals are pyrite, magnetite and/or hematite and chalcopyrite. Gangue is dominantly calcite with minor quartz. The host silicate assemblage is largely controlled by hydrothermal alteration and consists of variable amounts of the following minerals: K-feldspar±Ca-amphibole±biotite±sericite±chlorite±tourmaline±epidote±quartz. In this study, both the Ojancos and the P. Cerda tailings were deposited in valley dam impoundments and when they were filled, new tailings were deposited upstream. As a result, high quantities of seepage migrated downstream into the older tailings impoundment. At Ojancos, the recent upstream tailings have excess of acid potential (7.1 wt.% calcite and 3.5 wt.% pyrite), whereas the older downstream tailings are characterized by alternations of several meter-thick intervals with high neutralizing potential (about 40 wt.% calcite and 2 wt.% pyrite) and intervals with high acid potential (about 3 wt.% calcite and 4 wt.% pyrite). Acid mine drainage (AMD) with the precipitation of schwertmannite (pH 3.15) and chalcoalumite (pH 4.9) flows out at the interface between the uphill and downstream tailings. Strong downstream element transport is taking place and contributes to the formation of the cementation zone (mainly gypsum, ferrihydrite and goethite, and locally jarosite) in the older downstream impoundment. The cementation zone (pH=4) shows strong enrichment of heavy metals (e.g., up to 6800 ppm Cu, 680 ppm Zn, 1100 ppm As), mainly adsorbed and as secondary sulfides (e.g., covellite). In contrast, at the P. Cerda, tailings impoundment carbonates are homogeneously distributed and the overall neutralization potential exceeds the acid potential (average of about 10 wt.% calcite and up to 2.5 wt.% pyrite). The up to 5-m thick oxidation zones (paste pH=6.9–8.3) at P. Cerda are characterized by interlayering of coarser dark gray unoxidized layers with fine-grained, Fe(III) hydroxide-rich, ochre to red-brown colored horizons. The hyperarid climate dries out first the coarse, sulfide-rich horizons of the tailings and limits so the oxidation, which is restricted to the fine-grained, due to their higher moisture retention capacity. However, results indicate that during operation an important element transfer from the younger upstream tailings to the older downstream tailings impoundment took place, possibly by sorptive transport at ferric polymers or colloids in the form of neutral mine drainage (NMD). This would explain the metal enrichments in the cementation zone, which are mainly associated to the exchangeable fraction and not as secondary sulfides. This results, in both cases (in Ojancos mainly as AMD and in P. Cerda mainly as NMD), in Fe(III) input as ferric cation, as ferric polymer, or CO 3 complexes to the downstream impoundment. This constitutes a very effective acid potential transfer to the older downstream material because oxidation via input of external Fe(III) produces 16 mol of protons per mol FeS 2, i.e., eight times more than via oxidation with oxygen. In addition, the created acidity favors dissolution of the abundant Fe oxides magnetite and hematite of this ore deposit type providing so additional Fe(III) for sulfide oxidation.

Air–surface exchange of mercury in natural and anthropogenically impacted landscapes in Atlantic Canada

Understanding the relative importance of various sources of mercury within ecosystems and their subsequent impact on biota requires a thorough understanding of the biogeochemical cycling of mercury. An important component of the biogeochemical mercury cycle involves the exchange of gas-phase mercury between the atmosphere and various landscape surfaces. This study examines air–surface exchange (flux) of gaseous elemental mercury in selected undisturbed and anthropogenically impacted aquatic and terrestrial landscape settings. The objective of this study was to quantify air–surface mercury exchange rates in these contrasting landscape settings, to put constraints on these rates and provide insight into physical processes controlling mercury flux. Mean daily mercury flux was typically low over natural forest soils varying from −0.4 (net deposition) to 2.2 ng m −2 h −1 (net evasion). Minimum and maximum flux rates measured at these sites over the diurnal cycle ranged from a minimum of −1.3 to a maximum of 5.7 ng m −2 h −1 . Low flux rates from forest soils was partially due to poor light penetration through the forest canopy. Average daily mercury flux over undisturbed glacial till soil at an open field site with full sun exposure was also low (0.9 ng m −2 h −1 ) compared to the same soil mixed over a depth of 2 m (8.0 ng m −2 h −1 ). Mean daily mercury flux rates measured over pristine freshwater lake surfaces were also low (0.7–6.5 ng m −2 h −1 ), although the overall range in flux varied more widely (−0.3 to 44 ng m −2 h −1 ). Total mercury concentrations in marine water at a moderately polluted coastal harbour site (mean c. 0.8 ng l −1 ) and mean daily mercury flux rates over salt water (0.7 ng m −2 h −1 ) were similar to those measured over freshwater lakes. The highest daily average and diurnal range in mercury flux rates were measured at two abandoned gold mine tailings sites (130 and 237 ng m −2 h −1 ). The abandoned mine sites comprised tailings high in mercury content due to the past use of the mercury amalgamation process in gold extraction from the ore. Flux rates from tailings were two orders of magnitude higher than those observed over undisturbed native soils of similar parent material. Higher flux rates at the mine tailings sites were accompanied by ambient air concentrations 5–10 times background levels at 20 cm above the tailings. Mercury flux from vegetation has not been widely considered. A preliminary study of mercury flux from a white pine tree ( Pinus strobus ) indicated that actively growing trees might play a role in the atmospheric mercury cycle.