Global and Regional Mercury Cycles: Sources, Fluxes and Mass Balances

In 1977 – 1980, 1990, and 1994 we measured the latitudinal distribution of atmospheric mercury over the Atlantic Ocean. The results of these measurements show a pronounced concentration gradient between the northern and southern hemispheres and a rather small variability of mercury concentrations within the hemispheres, especially within the southern hemisphere. The measurements made during the cruises also suggest globally increasing atmospheric mercury concentrations by 17.5% in the northern and 14.0% in the southern hemispheres between 1977 – 1980 and 1990 to concentrations of 2.25 ng Hg m3 in the northern and 1.50 ng Hg nr3 in the southern hemispheres in 1990 and a global decrease between 1990 and 1994. The decrease between 1990 and 1994 of 20.4% in the northern and 21.2% in the southern hemispheres is consistent with the decrease of 23.3% observed in the same period at the Wank summit (1780 m a.s.l) in southern Germany. The results of all these measurements are reviewed. From these results and from measurements reported by others, we derive several constraints on the budget of atmospheric mercury, its atmospheric burden, residence time, sources, and sinks.

The global mobilization and biogeochemical cycling of Hg at the Earth’s surface has been studied using a variety of mass balance formulations [e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. The prominent role of atmospheric and oceanic processes, and anthropogenic Hg emissions in modern Hg cycle are evident. Most research however, has focused on terrestrial systems, with the biogeochemistry of Hg in fresh water systems receiving much attention. Paradoxically, the oceans have been largely ignored, yet the primary exposure of humans to methylmercury (MMHg) is through the consumption of marine fish and fish products. MethylHg compounds are considerably more toxic than elemental Hg and its inorganic salts, and prenatal life is more susceptible to MMHg-induced brain damage than adults [8]. The risk to public health is evident in the fish consumption advisories that have been issued in Canada, Scadanavia, by more than 30 states, the US FDA, the World Health Organization (WHO) and numerous other governments. Moreover, the coupling between atmospherically-borne Hg contamination and high MMHg concentrations in fish has been recognized. For example, a principal theme for the recently formed SCOPE Committee on Hg in the environment is a reevaluation of the influence of atmospheric Hg cycling and sea-air exchange on MMHg levels in oceanic fish [11]. In the U.S.A., the 1990 Clean Air Act Amendments require an assessment of health risk to humans and wildlife caused by Hg emissions. The potential adverse impact of atmospheric Hg deposition to coastal waters is contained in a planned Environmental Protection Agency (EPA) report to Congress [12].

Using Lake Ontario as an example, this paper presents estimates of atmospheric inputs of mercury based on an atmospheric deposition and air-water exchange paradigm, invoking recent environmental measurement data obtained in the vicinity of this lake. These calculations consider not only direct deposition (wet & dry) to the lake surface, but also take into account atmospheric loading to the surrounding watershed (drainage basin), as well as mass transfer of elemental Hg at the air-water interface.

One of the most important current tasks in research on mercury behaviour in the environment is the increase of our knowledge concerning its cycles and balances in regional scales, including those for different types of regions. Such studies depends very much upon the completeness and quality information regarding local emissions and fluxes of mercury (Hg). However, the magnitudes of the anthropogenic emissions of Hg from different point and diffuse sources, and the fluxes due to natural emissions from the Earth surface are still very uncertain in major cases. Recently the main natural sources and anthropogenic emissions, as well as Hg budgets, were estimated for Siberia, the largest part of the Russia (about 10 million km 2). The huge size of this region makes the problem not only locally significant, but globally too.

Field experiments have been carried out at different sites in north-western and central Europe to measure mercury concentrations in air and precipitation. Additionally, ozone and aerosol black carbon (soot) concentrations have been measured at these sites. The data sets contain results from a background site on the west coast of Ireland, a site which was impacted by regional scale atmospheric transport of mercury and a factory site located in the area of the strongest emissions of Central Europe in former GDR. Deposition estimates based on experimental data are compared with model results for these locations.Furtheron, local transport and deposition of mercury in the vicinity of partly inactivated chlor-alkali plants in former GDR was estimated by combination of field measurements and numerical modelling. With this approach the strengths of individual mercury sources located on the factory site were estimated and a mass balance for the entire site was calculated.

Although quite extensive information exists on environmental and health effects of mercury and its behavior in the environment, much less information is available on the emission fluxes of the element. Preliminary studies conclude that on the global scale the Hg emission to the air is comparable with direct inputs of the element to the aquatic environment and are almost a half of the direct releases to the terrestrial environment. Thus, the atmosphere is an important pathway for mercury cycling in the environment. Globally, combustion of fossil fuels to produce electricity and heat is the major source of atmospheric emissions of Hg. Both national and regional emission inventories indicate that the combustion of fuels, particularly coal, emits more than half of the atmospheric Hg in Europe. Major portion of Hg emissions from combustion of fuels is in a gaseous phase. In the combustion zone Hg present in coal or other fossil fuels evaporates in elemental form and then most likely a portion of it is oxidized while in the flue gases. The oxidized forms of Hg can be retained in modern flue gas cleaning systems. Refuse incineration seems to be the second largest source of Hg emissions to the atmosphere. Emission generation process for Hg during the incineration of wastes is similar to that during combustion of fossil fuels. However, more Hg in the oxidized form is expected from incinerators due to the higher content of chlorine in the wastes compared to fossil fuels.

Looking at emission quantities of chemicals is useful for an initial comparison of sources, but such a comparison does not relate linearly to health risk to human and non-human populations, but is a key input for risk assessment. Accurate emissions data are just now becoming available in the US for the trace element, mercury (Hg). We can use these emissions estimates to constrain risk assessments, to bound emissions for countries which lack data, and to further constrain the global cycle. Two parts of the equation remain highly uncertain: the amount of emitted Hg that participates in the global cycle and the quantification of natural sources; the latter is our greatest research need.

This article reviews current knowledge of atmospheric mercury processes and describes activities in Europe and North America to simulate these processes by means of tropospheric chemistry/transport models for regional-scale applications. Advantages and limitations of relatively simple Lagrangian models are discussed within the context of issues currently facing the environmental scientific and policymaking communities. The current state and future direction of comprehensive Eulerian models in simulating the tropospheric chemistry and transport of mercury species is outlined. A number of central improvements in these models are discussed, with consideration of the key progress necessary to include feedbacks and interactions between formation and distribution of clouds and mercury atmospheric chemistry.

An overview of the state of the art of airborne mercury transport simulation and of the modern knowledge in this field was made at the expert meeting in March, 1994 (Expert Panel on Mercury Atmospheric Processes, March 16–18, 1994, Tampa, Florida, USA) and at the EMEP Workshop (Pacyna et al., 1993).

A mass balance budget for mercury at Ocean margins is proposed on the basis of recent data on exchange fluxes with terrestrial, atmospheric and marine reservoirs. The largest single mercury flux consists of particulate species from rivers (~5 Mmol a-1), which is, primarily, unreactive and quickly buried in nearshore sediments. The most important source of mobile mercury for ocean margins is the atmosphere directly via deposition and indirectly via upwellings (total ≈ 5 Mmol a-1). The direct atmospheric input to coastal zones rivers (~2 Mmol a-1), exceeds considerably the dissolved inputs from rivers (~0.13 Mmol a-1). The evasion of elemental mercury from coastal surface waters is balanced by the mercury deposition from the atmosphere, but geographical differences exist suggesting a net mercury transfert to the higher latitudes. The overall residence time for mercury on the continental shelves is about 4 months. This is less than the mean residence time of water on the shelves (~1.3 year) and confirms the known reactivity of mercury in aquatic environments. However, two fractions of mercury are actively recycled (through the atmosphere and through the organic carbon recycling). The highly productive zones associated with frontal stuctures near the shelf edges appear to be very active in redistributing mercury species between ocean and coastal areas. One consequence is that the main source of methylmercury for coastal waters is upwelled oceanic waters.

A review of the available information on the sources and sinks for oceanic mercury (Hg) illustrates the importance of the ocean in the global Hg cycle. The principal source of oceanic Hg is atmospheric deposition with riverine sources contributing about 10% of the total inputs. The primary loss term is gas evasion at the sea surface. Burial of Hg in ocean sediments is a minor sink (10% or less of the total flux). Mass balance estimates suggest that hydrothermal sources do not contribute significantly to the oceanic Hg pool. Overall, about 11 Mmol/yr is currently being added to (and lost from) the ocean reservoir. Much of this Hg is of anthropogenic origin. Mercury deposited to the ocean is effectively reduced, principally by biologically-mediated processes, to elemental Hg and this leads to a rapid recycling to the atmosphere of much of the deposited Hg. Reactive Hg is converted in deeper ocean waters to methylated Hg species. Methylated Hg species and elemental Hg are important components of the Hg pool and the interconversion between the different Hg forms determines, to a large extent, the fate and transport of Hg in deep ocean waters, and the extent of Hg accumulation in marine food chains.

The particularities of mercury pollution of the Katun river basin (West Sibiria) are summarised. On the base of natural data the amount of different forms of mercury transported by the river was calculated. It is concluded that suspended sediments play the main role in the mercury transport and at present time the Katun river hydrochemical conditions hinder the process of mercury accumulation in the biota.

From 1992 till 1994 sampling campaigns were carried out in the Scheldt estuary the year round, in order to assess the behaviour of the various Hg species. On the basis of these data annual and seasonal budgets were calculated for particulate Hg, total dissolved Hg and dissolved methylmercury. These budgets allow evaluation of the consistency of each individual flux or process such as the inputs of the various Hg species into the North Sea, the sedimentation flux of particulate Hg in the area of high turbidity, the evasional fluxes, the formation rates of Hg°, etc. in relation to the other ones.

Research on mercury distribution in Lake Baikal was carried out in September 1990, June 1992 and March 1993. The data of these expeditions were used to estimate fluxes of total Hg and MeHg through the various environmental compartments (water, atmosphere, biota and sediments) in the lake.Atmospheric deposition (wet and dry) is the major source of mercury to the lake. Annually this flux accounts for 3.5% of the total Hg load in the lake. Riverine inputs are an important MeHg source; the annual flux accounts for 1.6% of the MeHg load in lake water. The concentrations of MeHg are much higher in river waters and shallow bays compared to the open waters of Baikal. Atmospheric inputs of MeHg are of the same order of magnitude as the flux to the biotic compartment. MeHg concentrations in Baikal fish are significantly lower than in North American and Scandinavian lakes. The sediments are the largest sink for Hg. Estimated gaseous emissions account for 15% of the atmospheric depositional flux.

The Elbe river, running through the Czech Republic and Germany is polluted heavily by a large variety of chemicals, but mercury is the dominant pollutant. Chloralkaline plants in the Czech Republic and in Eastern Germany, as well as an acetaldehyde production, discharged often more than 25 t Hg/year into the river and its tributaries. The measured concentrations for total mercury often exceed 100 mg Hg/kg in sediments or suspended matter, compared to a background of about 0.2 mg/kg.The organic species of mercury encountered are methylmercury compounds in both the sediments and the floodplain soils and dimethylmercury in the soil air. These concentrations vary during the year the highest concentrations are found in summer, as shown by the analysis of sediment cores.The speciation of mercury changes along the length profile of the Elbe river: as the mercury is transported towards the North Sea, its association with high molecular weight humic substances increases, which is of importance for the availability of this metal.

Mass balance studies in Wisconsin, Canada, and Sweden indicate that atmospheric deposition is the principal source of Hg to many lakes, whether delivered directly or indirectly via terrestrial runoff. The reported depositional flux varies 10-fold among these northern regions (3 to 30g Hg/km2/y). Watershed export rates vary from about 1 to 6 g Hg/km2/y, indicating that a substantial fraction of the atmospheric load is retained by terrestrial catchments. Hg residence times vary from 150 to 300 days in small Wisconsin lakes. Losses occur through sedimentation, outflow, and gaseous evasion. Environmental factors such as pH and dissolved humic matter (DHM) effect the budgets, concentrations, and speciation of Hg in lakewaters. Sedimentation is favored by low pH and low DHM, gaseous evasion is favored by high pH, and high MeHg concentrations are favored by high DHM and low pH. Multiple regression models containing DHM and pH explained 85% to 90% of the variability in waterborne Hg (0.2 to 4.8 ng/L) and methyl mercury (MeHg: 0.04 to 2.2 ng/L) among WI lakes ranging in pH from 4.5 to 8 and in organic carbon from 1 to 22 mg/L. It is unclear whether DHM affects MeHg production in lakes or simply reflects co-transport with organic C from riparian wetland. Northern wetlands export 0.1 to 1 g MeHg/km2/y along with 10 to 100 kg C/ha/y (versus <0.05 g MeHg/km2/y from uplands and 0.03 to 0.3 g MeHg/km2/y deposited atmospherically). In Little Rock Lake, WI, pelagic fish are the largest reservoir for MeHg while organic sediments are the largest Hg(II) reservoir. Biotic MeHg concentrations increase 2-fold to 4-fold with increasing trophic level, but MeHg turnover rates are highest at low trophic levels. We estimate turnover rates of 5 g MeHg/km2/y by pelagic microorganisms versus 0. 5 g MeHg/km2/y by pelagic fish. These high turnover rates may indicate that in situ rates of MeHg formation and destruction are correspondingly high.

Atmospheric sources are recognized to be significant in the cycling of Hg in the biosphere, yet there are few reliable data on air/surface exchange rates of Hg in forests. We have developed a tower-based micrometeorological method for measuring gas-phase Hg fluxes over environmental surfaces, and have used this approach to measure Hg° fluxes over soils, vegetation, and water surfaces. These fluxes have been combined with modeling results based on measurements of atmospheric Hg concentrations and speciation to quantify the overall flux of Hg between the atmosphere and the ground. These results are compared with a study of the biogeochemical cycle of Hg in the temperate deciduous forest at Walker Branch Watershed in the southeastern United States. Our preliminary results suggest that the largest Hg fluxes in forests involve gas exchange at the air/vegetation interface. Given the magnitude of these fluxes and their level of uncertainty, this forest could act as a net source or sink for atmospheric Hg, indicating the importance of better understanding the role of Hg exchange at the vegetation surface.

In Scandinavia 1979, measurements of Hg in air was started at the Swedish west-coast. This site, the Rörvik station, has since then been part of several Swedish and Nordic atmospheric Hg networks. The results from the measurements in these networks clearly demonstrate the existence of a decreasing south-north gradient over Scandinavia. A compilation of available data suggest a decrease in Hg concentrations in air starting in 1990. This coincides with drastically decreased Hg emissions following the closing down of large Hg emitters in the former GDR. Scandinavian measurements of total Hg in precipitation are available from 1987. The coupling between atmospheric Hg in Scandinavia and large continental emission sources, is supported by a clear decreasing south-north gradient and by the fact that the total Hg precipitation data also start to decline around 1990. MeHg in air has recently been measured, but the data set is still too limited to allow trend analysis. The MeHg precipitation data set is much larger, but a rather unclear south-north gardient is present and no coupling to the dramatic changes 1990 in Eastern Europe can be detected. By reviewing the seasonality and long-term changes in our atmospheric Hg data set, together with our Hg/MeHg budget calculation for forested ecosystems and changes in output rates, we try to derive any coupling between the load to a forested ecosystem and the output to surface waters. A lowered export rate of Hg and MeHg over time is indicated in catchment studies in south-western Sweden. However, it is not yet possible to statistically confirm the coupling between input and ouput fluxes of Hg/MeHg from a forested catchment.

The data about the role of natural organic substances in dissolution, transport and concentrating of mercury in environment are generalized. The interaction of mercury (II) with fulvic acids (FA) and humic acids (HA) have been studied. It has been shown that in the reaction of mercury ions with FA stable soluble high-molecular mercury (II) fulvate complexes are formed. These complexes are predominating form of mercury in surface fresh waters. The interaction of mercury (II) with FA leads to the abrupt increase of mercury mobility in waters and sails. On the contrary HA behave as complexing sorbents promoting mercury concentrating in soils and bottom sediments. Another direction of mercury (II) interaction with humus acids is the formation of organomercury compounds. Principal possibility of abiological methylation of mercury by means of interaction of mercury (II) with FA has been proved. The output of methylmercury increases with concentration of FA and pH.

Gold-mining in the Chita Oblast, Eastern Zabaikalye (Transbaikal), causes mercury contamination of the environment. Contrast and localization of the contamination depend on duration and intensity of amalgamation usage in concentration of gold-bearing sands and ores and also on natural landscape and climatic conditions. Mercury concentrations in the environmental compartments of the gold-mining areas and background zones vary between <0.001 and 0.78 mgkg-1 for rocks and minerals, <1 and 183 ngm-3 for atmospheric air, 0.013 and 3.59 mgkg-1 for soils, <5 and 5 000 ngl-1 for dissolved mercury, <5 and 27 800 ngl-1 for particulate mercury of natural and industrial waters, 0.008 and 54.2 mgkg-1 for bottom sediments.

The present work is a first attempt to give an overall estimate of anthropogenic mercury pollution in Siberia. The enterprises of three main groups of industries emitting environmentally harmful amounts of mercury and mercury-containing compounds are considered (non-ferrous industry; chemical,electronic and electrical industries; oil,gas,coking industries). The figures of mercury emission cited in the present work do not claim to be fully complete and absolutely accurate. For this reason, many estimates presented here will certainly need a refinement. In spite of this, the authors belive, that what is presented is a rather comprehensive picture.

The combined technique of thermal and atomic absorption analyses is developed using the synthetic minerals with definite mercury-containing species (forms) as reference samples. It is shown that natural pollutants which are mainly resultant from ore-forming processes and geothermal fluids are characterized by a high level and irregular distribution of Hg concentrations and could easily be distinguished by the relatively abundant sulfide form and the presence of the high-temperature isomorphous one. By contrast, under the anthropogenic pollution only low-temperature, sorbed forms are found, whereas sulfide and isomorphous ones are quite rare or fully absent. This allowed rough estimates of anthropogenic and natural contributions to Hg pollution to be obtained.

Natural sources which accumulate mercury are discussed. Among natural suppliers of mercury into the atmosphere, degassation of the Earth’s crust (including mercury mineralization zone) and evaporation from oceans play a Key role. This paper concers natural sources of mercury in Siberia, to which Hg deposits, various Hg-containing deposits and rocks with high Hg contents due to the development of dispersion haloes are assigned. The Hg deposits are located in three ore regions over the territory of Siberia: Altai-Sayan, Transbaikalia and the Upper Yana-Kolyma regions. The estimated natural emmission of mercury due to the surface degassation is about 40 t/y.

Presented data are based on results of study, which were started in Institute of Geology of Ore Deposits, Petrography, Mineralogy and Geochemistry of the RAS and continue more than 40 years. Regional and local regularities was shown related to mercury concentrations in natural deposits of different type: ore (except cinnabar) and non-ore (including hydrocarbons, coal and oil shales) deposits, products of volcanic and hydrothermal activity. Reasons of high Hg concentration appearance were discussed for different geological systems in connection with processes of mantle degasing through deep fault zones. New type of ore-mercury belts was proposed, based on finding of mercury-bearing ore and oil-gas deposits. Evaluations of mercury concentration scales and environmental conclusions were produced.

The paper reports data on mercury abundance in soils of of Southern West Siberia. The natural geochemical background of mercury in soils for the territories without natural anomalies estimated over upper horizons, is 45±3 ppb (background range is from 14 to 73 ppb). The presence of technogenic constituent of geochemical background is least within Altai Territory and is considerably higher in the Novosibirsk and Kemerovo regions. The levels of mercury contents in soils from nearly background to abnormally high values are in direct dependence on their concentrations in soil-forming rocks. In the domain of background values it is shaded by various natural processes, atmospheric falls included.

Hg occurrences in Hungary are related to the Neogene Inner Carpathian Volcanic Belt but are of no economic importance. There are also polluted areas due to mining activity. The chemical industry is the main user of mercury. Kazincbarcika (northern Hungary) is an area where extreme and uncommon Hg pollution has occurred During hydrochloric acid production hundreds of tons of mercury has leaked out, contaminating the upper 4–5 m of sediments. Another seriously high mercury content has been detected at Balatonfűzfő village (near the lake Balaton), in the grounds of the Nitrokémia Works. Concentrations of Hg of limited extent can be detected in the Budapest agglomeration area too.

The mercury occurs as cinnabar and in different ore minerals. Its concentration depends on the geological conditions at the time when mercury migrated upwards: the most favourable during the rifting periods (Triassic rifting in the Dinarides), less favourable during postcollisional phases (the Vardar zone and the Serbo-Macedonian mass), and unfavourable over a subducted plate (the Upper Cretaceous Timok volcanic zone — TVZ).In the TVZ a vertical zonality in the mercury distribution is evident: the deepest generated ores are the poorest and the mineralization originated in the highest levels the richest. However, the metallurgical treatment even of the in mercury poorest ores, because of high production, induce ecological damage.

Starting in Latin America in the 1980s a revival of the use of amalgamation in gold panning has spread rapidly in tropical areas and roughly 10 million people was estimated to be engaged in in this activity. This is the major reason for this project started by SCOPE. An overview is given of the aims of the project and the meetings held in Stockholm and Rio de Janeiro as well as the future plans of meetings and fact finding missions in Africa and Asia. Important aspects of the project is a comparison of the behaviour, distribution and transformation of mercury in the tropics as compared to temperate regions and the global load of mercury in the environment.

Knowledge about the behaviour of mercury in aquatic and terrestrial ecosystems largely result from studies carried out in Europe, North America and Japan e.g. in temperate areas. The SCOPE mercury project aims at complementing this information with results from studies in other areas including the tropics. One of the reasons for this interest is the rapidly growing use of mercury in connection with small scale gold mining in developing countries all over the globe. This paper presents some of the preliminary findings from studies in the Amazone region.