Multielement analysis and mercury speciation in atmospheric samples from the Toronto area

Inductively coupled plasma atomic emission spectrometry. (ICP-AES), graphite furnace atomic absorption (GFAAS) and gas chromatography with an atomic absorption detector (GC-AAS) were used for the analysis of atmospheric samples collected in the Toronto area of Southern Ontario. Airborne particulate matter and particulate and soluble fractions of snow were analyzed for S, Al, Fe, Ca, Mg, P, Pb, Zn, Cu, Cr, Cd, Ni, As, and Hg. Also, determinations of organic Hg compounds in filtered snow were performed. The element concentrations in airborne particulate matter from Toronto were similar in amount to those found in other major cities and were the highest in winter. Regional differences in the chemical composition of snow were shown to be related to industrial density. The majority of elements existed in the highest concentrations in the soluble fraction of snow, except for Al, Fe, and Pb, which were largely present in particulate matter. Differences in the distribution of elements in particulate matter in fresh and old snow were observed and depended on particle size and element. The results demonstrated the usefulness of snow as an indicator of atmospheric pollution.     

Modelization of the mercury fluxes at the air-sea interface

Aqueous and atmospheric Hg° concentrations for remote marine areas such as the equatorial Pacific Ocean and for coastal seas such as the North Sea and the Scheldt Estuary are discussed. Biological processes seem to be at the origin of the supersaturated Hg° concentrations in the water. On the other hand, transfer velocities across the air-sea interface were calculated with a classical shear turbulence model and with a wave breaking model. With these data, Hg° fluxes from the sea to the atmosphere were calculated: in the Pacific Ocean they range from 0.43 to 6.5 μg g Hg.m^?2. yr^?1 at a wind speed of 2.8 m.s^?1 and from 10.3 to 156 μg Hg.m^?2 yr^?1 at a wind speed of 54 m.s^?1, but they are still higher when wave breaking is considered (from 11 to 168 μg Hg.m^?2.yr^?1). These transfer fluxes are an order of magnitude higher in the Scheldt Estuary.     

Lidar measurements of atmospheric mercury

Mercury is the only atmospheric pollutant that is present in the atmosphere in atomic form. The optical resonance line at 254 nm can be used for absorption measurements using different optical remote sensing techniques. Range-resolved Hg mapping can be performed using the differential absorption lidar (DIAL) technique. We have used the lidar technique both for mapping of industrial plumes and for background concentration measurements. Our studies also include Hg of geophysical origin.     

Cycling of methyl mercury and mercury — Responses in the forest roof catchment to three years of decreased atmospheric deposition

Studies of the biogeochemistry of total mercury (Hg) and methyl mercury (MeHg) in the Lake Gårdsjön watershed have shown that the atmosphere is the most important source of Hg and MeHg in the ecosystem. Soils are accumulating most of the deposited Hg and MeHg, but transport of Hg and MeHg from the forested catchments into the lake ecosystems is enough to explain elevated concentrations of MeHg in fish in more than 10 000 Swedish lakes. An experimental roof was constructed to study effects of decreased atmospheric input on an entire forested catchment. The experiment started in April 1991, and decreases in the output of both MeHg and Hg occurred during 1991, 1992 and 1993. Runoff fluxes from the control catchment during the pre-treatment period were related to the experimental catchment using regression analyses. Since April 1991, after three year experiment, predicted compared to measured fluxes showed that Hg output decreased by 32% and MeHg by 28%. The decrease in Hg was most obvious during high water flows in winter/spring while MeHg decreased during all seasons of the year. The decreased input of Hg and MeHg to the Forest Roof Catchment is the most probable explanation to the rapid decrease in output of Hg and MeHg by runoff from the catchment basin.     

Occurrence and turnover of atmospheric mercury over the nordic countries

A seasonal variation of both particle and gaseous Hg concentrations in the atmosphere is present in south-western Sweden. An average gaseous Hg level of 3.7 ng m^?3 is found in winter, compared to 2.8 ng m^?3 in summer. A weak decreasing south-north gradient for gaseous Hg in air over the Nordic countries is also present, with yearly average values from 3.2 to 2.8 ng m^?3. A gradient for particulate Hg is less clear. An air parcel trajectory sector classification of gaseous Hg levels in air, and to some extent the particulate associated Hg, clearly demonstrates the increased concentrations in the southern sectors, especially in south-western Sweden where the gaseous Hg increase is about I ng m^?3. These observations are consistent with an influence from the European continent. The average concentrations of Hg in precipitation at the various stations show a pronounced decreasing south-north gradient. A major portion of the total Hg present in precipitation is associated with particles. For the southern stations, a strong correlation between Hg and sulfate, or pH, is present suggesting a connection between Hg in precipitation and anthropogenic activities.     

Industrial emissions of mercury in Finland between 1967 and 1987

Mercury pollution has been studied in Finland since 1960. In the 1960 's, industrial use of Hg was about 50 t yr⁻¹, 80% of which was consumed by chlor-alkali plants. In the 1980's, about 19, to 25 t yr⁻¹ Hg was used and compared to past consumption, its use has decreased by 40 to 50%. In 1987, the import of Hg through raw materials and other sources in Finland was, however about 170 t yr⁻¹ which is rather high. There are no detailed statistics about Hg emissions in the 1960's. In 1967, air and water emissions of Hg were 2.8 and 8.3 t yr⁻¹ respectively. In 1987, estimated air emission of Hg was 3.48 t yr⁻¹ whereas water emission was 0.15 t yr⁻¹ Aquatic emission has been lowered by 50% which is due to the fact that pulp and paper industries stopped using Hg as slimicide. On the other hand, the air emission of Hg has not decreased. This is due to the high Hg content in raw materials used by metallurgical and power plants.     

An intensive multi-site pilot study investigating atmospheric mercury in Broward County,Florida

An intensive multi-site pilot study of atmospheric Hg was conducted in Broward County, Florida in August and September of 1993. Broward County, which contains the city of Fort Lauderdale, is located in southeastern Florida. The county borders the Florida Everglades on the west and the Atlantic Ocean on the east. A network of four sampling sites was set up for 20 days throughout Broward County to measure Hg in both the vapor phase and the particle phase as well as Hg in precipitation. The mean concentrations of total vapor phase Hg measured at two inland sites were found to be significantly higher (3.3 and 2.8 ng/m³) than that measured at a site located on the Atlantic shore (1.8 ng/m³). The mean concentrations of particle phase Hg collected at the two inland sites (51 and 49 pg/m³) were found to be 50% greater than that measured at the coastal site (34 pg/m³). In addition, event precipitation samples were collected at four sampling sites over the 20 day study period and were analyzed for both reactive and total Hg. The mean concentration of total Hg in the precipitation samples was found to be 44 ng/L, with a range of 14 to 130 ng/L. It was determined that further meteorological analysis and a more complete characterization of the aerosol and precipitation composition are needed to identify the probable source(s) contributing to the increased deposition of Hg.     

Pre-industrial atmospheric deposition of mercury: Uncertain rates from lake sediment and peat cores

Lacustrine sediment cores from depositional areas have frequently been used to estimate pre-industrial rates of atmospheric Hg deposition. However, this approach tends to result in overestimates, partly because of Hg inputs from the catchment, partly because of a horizontal redistribution of sediments within lakes. Peat core studies may suffer from a vertical migration of Hg due to water table fluctuations. A natural Hg deposition rate around 2 μg m^?2 y^?1 is suggested to be more realistic than values of 3 to 12 μg m^?2 y^?1 reported from recent studies. The anthropogenic impact on the present Hg deposition may have been underestimated accordingly.     

International field intercomparison of atmospheric mercury measurement methods

To determine the extent of comparability of sampling and analytical procedures for atmospheric mercury (Hg) being used by different scientific groups around the world and hence the compatibility of measurement results, the Atmospheric Environment Service (AES) co-ordinated a field intercomparison study in Windsor, Ontario, over a period of 5 days- during Sept./Oct.,1993. This study brought together 2 groups (University of Michigan Air Quality Laboratory; Chemistry Institute of GKSS) which performed conventional (manual) sample collection procedures for total gaseous mercury (TGM) and for particulate-phase mercury (PPM), followed by cold-vapor atomic fluorescence spectrophotometric (CVAFS) analysis in the respective laboratories. Two other groups (Ontario Hydro, and the Ontario Ministry of Environment & Energy) each operated a novel mercury vapor analyzer produced by Tekran Inc. of Toronto. As is the case for the manual methods, this analyzer also uses gold amalgamation and CVAFS. During the intercomparison, meteorological parameters (air temperature, barometric pressure, wind speed/direction and relative humidity) were obtained at the study site.     

Linkages between atmospheric mercury deposition and the methylmercury content of marine fish

Enhanced Hg deposition to productive marine systems may result in concurrent increases in monomethyl Hg (MMHg) concentrations of marine fish. Consequently, it is important to understand what effects an increasing Hg supply may have on the marine food chain. A simple ocean model is employed to estimate the fraction of total Hg inputs which is required to sustain “average” marine fish MMHg concentrations annually. Calculations show that upwelling zones require 20% of total annual Hg inputs, coastal zones 5%, and open-ocean regions only 0.02%. The value for coastal areas is similar to that calculated for the acidified basin of Little Rock Lake, Wisconsin, a small fresh water seepage lake. These calculations point to Hg source strength and rates of particle scavenging as being key factors in controlling the rate of transport to sites of methylation (and subsequent entry into the marine food chain). If biological variables (scavenging rates, primary productivity) remain constant while anthropogenically-derived Hg deposition increases, it is likely that concentrations in marine biota (including fish) will rise in accord.     

Ambient levels and dry deposition fluxes of mercury to Lakes Huron,Erie and St. Clair

Ambient concentrations and dry deposition fluxes of Hg in the gas and particle phase to Lakes St. Clair, Erie and Huron were estimated with a hybrid receptor-deposition model (HRD). The ambient gas and particulate phase Hg concentrations were predicted to vary by a factor of 12 to 18 during the transport of air masses traversing the lakes. The ensemble average deposition fluxes of fine particle Hg ranged from 7 pg/m²-h to 15.3 pg/m²-h over Lake St. Clair, 0.5 to 4.2 pg/m²-h over Lake Huron and 5.1 to 20.6 pg/m²-h over Lake Erie. The deposition flux of coarse particle Hg was in the range of 50 to 84 pg/m²-h over Lake St. Clair, 4.7 to 24.2 pg/m²-h over Lake Huron and 5.1 to 20.6 pg/m²-h over Lake Erie. Gaseous Hg volatilized at a rate of 0.21 to 0.52 ng/m²-h from Lake Huron and 0.13 to 0.36 from Lake Erie. Gas phase Hg was deposited at a rate of 5.9 ng/m²-h and/or volatilized at a rate of 0.5 ng/m²-h from Lake St. Clair depending upon the location of the sampling site used in the HRD model. The effect of meteorological conditions, particle size distributions and type and location of the sampling sites played an important role in the transfer of atmospheric Hg to and/or from the lakes.     

Biomonitoring of atmospheric mercury in the vicinity of Kilauea,Hawaii

The release of mercury from volcanic eruptions on the Island of Hawaii has been well documented, but the dispersion characteristics have not previously been studied in detail. Samples of lichens were collected from around the island, and subsequent analysis showed levels ranging from < 8 to 59 µg Hg g^-1. The distribution of the accumulation of mercury clearly shows that there is a major input from the active eruption sites, but that this is significantly modified by variations in wind patterns. Superimposed on this regional distribution are two localised sites where the mercury accumulation is exceptionally high; this is attributed to the release of steam from the circulation of shallow ground waters.     

An assessment of adult risks of paresthesia due to mercury from coal combustion

This paper presents a probabilistic assessment of the risks of transient adult paresthesia (tingling of the extremities) resulting from ingestion of methylmercury (MeHg) in fish and shellfish. Two scenarios are evaluated: the “baseline,” in which the MeHg dose results from the combined effects of eating canned tuna fish, various marine seafoods, and freshwater sportfish, and an “impact” scenario in which the Hg content of the freshwater sportfish is increased due to local deposition from a hypothetical 1000 MW_e coal-fired power plant. Measurements from the literature are used to establish the parameters of the baseline, including atmospheric rates of Hg deposition and the distributions of MeHg in fish. The Hg intake for the impact scenario is then based on linear scaling of the additional annual Hg deposition as estimated from a Gaussian plume dispersion model. Human health responses are based on a logistic fit to the frequencies of paresthesia observed during a grain poisoning incident in Iraq, 1971–2. Based on a background prevalence rate of 2.2% for adult paresthesia, the assessment predicts a 5% chance that the increase in paresthesia prevalence due to either baseline or incremental MeHg doses might approach about 1% of the background prevalence rate.     

Interpreting the variations in atmospheric methane fluxes observed above a restored wetland

The eddy flux of methane (CH₄) was measured over 14 months above a restored wetland in western Denmark. The average annual daily CH₄ flux was 30.2 mg m⁻² d⁻¹, but the daily emission rates varied considerably over time. Several factors were identified that explained some of this variation. (1) Grazing cattle moving through the source area of the eddy flux mast increased the measured emission rates by one order of magnitude during short time periods. (2) Friction velocity exerted a strong control on the CH₄ flux whenever there were water pools on the surface. (3) An exponential response of the daily CH₄ flux to soil temperature at 20 cm depth was found for most of the study period, but not for parts of the summer season that coincided with a low water level in the river flowing through the wetland. (4) Additional variations in the CH₄ emission rates were related to the spatial heterogeneity of the source area. This area covered not only different plant communities but also a gravel road and a river surface, and it had a microtopography that visibly induced a large spatial variability in the wetness of the top soil. It is shown that the control mechanisms for the methane emission from restored wetlands are more complex than those reported for natural wetlands, since they include both management activities and slow adaptive processes related to changes in vegetation and hydrology. On the basis of eddy fluxes of carbon dioxide measured at the same site it is finally demonstrated that the variability in the CH₄ fluxes strongly affects the greenhouse gas sink strength of the restored wetland.     

Long-term changes in concentration and deposition of atmospheric mercury over Scandinavia

Samples for measurements of total gaseous mercury (Hg) in air have been collected since 1980 in south-western part of Scandinavia. A collection program for precipitation samples used to determine changes in depositional fluxes of total Hg has been in operation since 1987. A comparison of today's total gaseous Hg levels in air and the total Hg concentrations in precipitation with the ones found earlier, shows a clear decrease with time. At the Swedish west-coast, yearly average air concentrations and median levels of 3.3 and 3.1 (1980–1984), 3.2 and 2.8 (1985–1989), and 2.7 and 2.6 ng Hg/m³ (1990–1992), respectively, were found. Increased average and median winter concentrations were always found, with levels at 3.7 and 3.4, 3.7 and 3.3, and 3.0 and 2.7 ng Hg/m³ for the respective time period. Higher winter values were expected due to increased anthropogenic emissions and changes in the mixing height of the atmosphere. The corresponding total wet deposition rates decreased from 27 (1987–1989) to 10 μg Hg/m² yr. (1990–1992). A finding of special interest was the decreased number of episodic events of high total gaseous Hg levels in air, from 1990 and further on. In addition, the frequency distribution of the concentrations of Hg in air seems to be different for these years compared to the other two time periods. A frequency distribution of air concentrations of Hg more resembling a normal distribution was found for the years 1990 to 1992. The decrease of the atmospheric burden of total gaseous Hg and deposition of total Hg are most probably connected to lower emissions in source areas on the European continent. It seems logical to state that the problem of high Hg depositional fluxes to Scandinavia, is best solved by abatement strategies on the regional scale.     

Is mercury increasing in the atmosphere? The need for an atmospheric mercury network (AMNET)

Mercury uses in human endeavors will lead to a general, though variable, volatilization of Hg. Current estimates for anthropogenic interferences range from about 50 to 75% of the total annual Hg emissions to the atmosphere. Recent modeling suggests that the present atmospheric Hg burden has increased by a factor of 3 during the last 100 years with a current rate of increase of about 0.6% yr^?1 (ca. 0.01ng m^?3yr^?1). This impact, which is significant, can be examined and assessed empirically. To date, however, atmospheric Hg programs have not employed an experimental design sufficient to account for short time scale atmospheric Hg variations of natural and anthropogenic origin, and to resolve the long term temporal pattern. I am proposing an international research program, AMNET, or Atmospheric Hg Network, to address the important question, “Is Hg increasing in the atmosphere?” AMNET would examine temporal and spatial variations in atmospheric Hg and assess the influence of natural and anthropogenic sources on the global atmospheric Hg cycle. This program requires international support and cooperation. The experimental design of AMNET would follow the successful Atmospheric Lifetime Experiment Program (ALE), which examined the contemporary temporal changes in the atmospheric concentrations of the freons, methyl chloroform, carbon tetrachloride, and nitrous oxide. Following the ALE design, AMNET sampling stations would be maintained in both hemispheres and at sites free from strong local pollution sources of Hg (e.g., remote islands). Measurements would be made for a period of three to five years. The precision and accuracy of the Hg⁰ determinations must be ≥ 1%. The accurate resolution of the variability and secular trends in the atmospheric Hg burden can provide: (1) a direct quantitative assessment of the scale to which anthropogenic processes are affecting the natural biogeochemical cycling of Hg; (2) an essential refinement and constraint currently lacking in mass balance models; (3) an enhanced knowledge of the behavior of Hg in the atmosphere, and (4) an accurate data base required for global circulation atmospheric chemical Hg models.     

An integrative method to quantify contaminant fluxes in the groundwater of urban areas

Background, Aims, and Scope  Groundwater in urban areas is often contaminated and emission sources can be located close to groundwater wells. The delineation of contaminant plumes is difficult because of the various potential emission sources. Thus, detection, quantification and remediation of contaminated sites in a city need more integrative approaches. Methods  A method has been developed which allows quantification of mass fluxes of contaminants in groundwater between control planes. Budget zones along the flow path are defined to calculate a contaminant balance and to quantitatively reveal input areas. Concentrations and water budgets are used to calculate mass balances for each contaminant. The city of Darmstadt (Germany) was chosen to evaluate the method. Results  The groundwater monitoring wells (GMWs) upstream of the city showed anthropogenically superposed background values for all naturally occurring inorganic species. The contaminant concentrations increased in the city (probably influenced by road traffic, gas stations, leaking sewers, etc.). Downstream from the city, concentrations usually decreased. Organic compounds typical for urban environments, such as polycyclic aromatic hydrocarbons (PAH), locally exceeded drinking water regulations. In GMWs with high concentrations of organic contaminants in the city or downstream from industrial areas, a significant increase in Fe²⁺ and Mn²⁺ could be observed, in some cases coinciding with a decrease in NO₃, SO₄ and an increase in NH₄. Discussion  For typical urban contaminants, a positive budget was calculated in several zones, which shows that emissions from urban sources are reaching the groundwater. Negative budgets can be mainly explained with diving plumes and degradation. The input calculated from the individual budget zones is usually higher than the input estimated from urban emissions. Differences between the calculated and the estimated input can be explained with additional sources or (bio)degradation processes. Conclusions  It was confirmed that high concentrations of contaminants do not necessarily correlate with high fluxes. Integrative approaches can reveal areas of high contaminant mass input. The results obtained with the new method are plausible compared to the land use and the estimated urban input. The concentration pattern of Fe²⁺, Mn²⁺, SO₄ and NO₃ is partly due to natural processes, triggered by the degradation of organic matter and organic contaminants. Recommendations  Since this method includes mass balances and flux calculations, avoiding an overestimation of single point contaminant concentration, it is recommended to use this approach to quantify groundwater contamination in cities. Further research is focusing on the role of urban soils as natural reservoirs for the input of contaminants.     

Benthic fluxes of mercury during redox changes in pristine coastal marine sediments from the Gulf of Trieste (northern Adriatic Sea)

Purpose

The Gulf of Trieste (northern Adriatic Sea) is an example of a coastal environment contaminated with mercury (Hg). Contamination is a consequence of nearly 500?years of activity at the Idrija Mine (western Slovenia), which is the second largest Hg mine in the world. Oxygen depletion can be common in the Gulf of Trieste due to late summer stratification of the water column and accumulation of labile organic matter. Since changing redox conditions can have an impact on Hg transformations, we studied the effect of oxygen depletion, in parallel with sulphide, iron (Fe), manganese (Mn), fluorescent dissolved organic matter (FDOM) and nitrogen (N) and phosphorus (P) availability, on total Hg and methylmercury (MeHg) fluxes from sediments.

Materials and methods

Pore water concentrations and benthic fluxes of total dissolved Hg and MeHg were studied in situ and in microcosm laboratory experiments using flux chambers encompassing three different stages: oxic, anoxic and reoxidation.

Results and discussion

Our experiments showed that in the oxic stage there were small effluxes of MeHg to the water column, which increased in the anoxic stage and dropped rapidly in a subsequent reoxic stage, showing influx. Our results support the hypothesis that MeHg desorption from reduced metal hydroxides under anoxic conditions, and co-precipitation with Fe-oxides and MeHg demethylation in the reoxidation stage, may play a major role in determining MeHg benthic fluxes. For Hg and MeHg, it appears that there is little relationship between their pore water distribution and flux and that of FDOM, i.e. humics.

Conclusions

The results indicate that there was no significant difference in Hg and MeHg pore water levels and their benthic fluxes between the contaminated northern and central parts of the Gulf of Trieste and the pristine southern part. This suggests that shallow and stratified coastal marine environments, in general, represent areas with a risk of high benthic release of toxic MeHg.