Emissions of heavy metals (Mercury, Lead, Cadmium and other metals)

Heavy metals differ significantly in their properties and ability to long-range atmospheric transport. In particular, mercury is well known as a global scale pollutant with distinct ability to intercontinental transport. On the other hand, such particle bound heavy metals as lead and cadmium are mostly transported regionally. Therefore, mercury has the first priority for modeling on a global scale and requires global emissions data. Nevertheless, in many cases modeling other heavy metals also requires emissions data covering territories wider than the region of the primary interest. For example, lead and cadmium pollution levels in the EMEP countries and, in particular, in the EECCA countries can be affected by emission sources located in neighboring non-EMEP countries of Northern Africa, Meddle East, Eastern and Southern Asia.

Although, there are detailed heavy metal anthropogenic emissions inventories developed for a number of regions (e.g. Europe, North America), global emission datasets are quite rare. Table contains characteristics of available global emission datasets and inventories for some heavy metals. Global spatially resolved data on mercury emission are available for a number of years (1990, 1995, 2000, 2005). The only gridded emission dataset for lead relates to 1990. The lead emissions inventory has been updated for mid-nineties but these data have no spatial distribution. Only aggregated emission estimates without spatial distribution of emission sources are available for cadmium and relate to mid-nineties.


Chemical Period of time Global emissions, t/y Spatial resolution Dataset location Reference


1990 2 144 1º—x1º CGEIC  CGEIC website
1995 2 317 0.5º—x0.5º



 Pacyna et al., 2003
2000 2 190 0.5º—x0.5º  Pacyna et al., 2006
2005 1930 0.5º—x0.5º  AMAP/UNEP, 2008
2010 1960 0.5º—x0.5º  AMAP/UNEP, 2013


1989 167 889 - 206 435  1º—x1º CGEIC  Pacyna et al., 1995
1995 119 259 n/a n/a  Pacyna and Pacyna, 2001
 Cadmium 1995 2 983 n/a n/a  Pacyna and Pacyna, 2001


Mercury (Hg)

First global dataset of mercury anthropogenic emissions relates to 1990 and is available at website of the Canadian Global Emissions Interpretation Centre (CGEIC). It covers the globe with spatial resolution 1ºx—1º and distinguishes emissions from point sources (power plants, smelters and other large sources with stack height and a sufficiently buoyant plume that the emissions are transported into the atmosphere above 100 m) and area sources (e.g. residential and commercial boilers, mobile sources etc.)

Later two newer global mercury emissions datasets were released for 1995, 2000, 2005, 2010. They are available at the website of the Arctic Monitoring and Assessment Programme (AMAP) and contain considerably more detailed information on mercury emissions. The datasets have spatial resolution 0.5º—x0.5º and include the following parameters:

  • Gridded anthropogenic emissions separately from point and area sources;
  • Three emission height levels (below 50 m, 50-150 m, above 150 m);
  • Speciation of mercury emissions into elemental gaseous mercury (GEM), reactive gaseous mercury (RGM) and particulate mercury (TPM).



Lead (Pb)

The only global spatially resolved emissions inventory for lead relates to the year 1989 and is available at the CGEIC website. This dataset includes gridded information on lead emissions from point and area sources with spatial resolution 1º—1º. Besides, it provides low and high estimates of the emission rates based on emission factors that were assumed to be the highest and the lowest values within the emission factor ranges for a given source category. The figure shows global distribution of anthropogenic emissions of lead in 1990 (kg/km2/y). According to this inventory the most significant emission sources were located in Europe, Southeastern Asia, Mexico and South Africa. Global emission estimates varied from 168 to 206 kt/y.

The global emissions inventory for lead has been updated by Pacyna and Pacyna (2001) for mid-nineties. Estimates of lead emissions from anthropogenic sources have been revised based on new information from national experts in combination with statistical data on the consumption of raw material, industrial goods production. However, the updated inventory did not include spatial distribution of emissions. According to the inventory combustion of gasoline continued to be the major source of atmospheric emissions of lead, contributing about 74% of global emission in 1995.



Cadmium and other heavy metals

There is no spatially resolved global emission datasets available so far for cadmium and other heavy metals and metalloids (arsenic, chromium, copper, nickel, selenium and zinc). The most recent inventory for these metals includes total emission estimates for different continents and various source categories and relates to 1995 (Pacyna and Pacyna, 2001). The inventory contains estimates of heavy metal emissions from heat and power production, non-ferrous metal production, iron and steel industry, cement production and waste disposal. Emission of heavy metals from major anthropogenic source categories is presented in the Table. The estimates indicated that stationary fossil fuel combustion was the major source of chromium (69%), nickel (90%), and selenium (89%). Whereas non-ferrous metal production was the largest source of atmospheric emissions of cadmium (73%), arsenic (69%), copper (70%), and zinc (72%).


Source category Cd As Cr Cu Ni Se Zn
Stationary fossil fuel combustion 691 809 10 145 7 081 86 110 4 101 9 417
Non-ferrous metal production 2 171 3 457 18 071 8 878 466 40 872
Iron and steel production 64 353 2 825 142 36 7 2 118
Cement production 17 268 1 335 - 134 3 2 670
Waste disposal 40 1 124 425 621 129 24 1 933
Total 2 983 5 011 14 730 25 915 95 287 4601 57 010


Asia was the largest contributor to global heavy metal emissions made up 40-60% of total value. Contributions of other continents vary for different metals. Europe contributed 23% of global emissions of chromium and 21% of nickel; North America made up 24% of selenium and 22% of chromium emissions; South America was responsible for 21% of global copper emissions. The lowest contributions were made by Australia.


Emissions of Persistent Organic Pollutants (PCBs, HCHs, HCB, PCDD/Fs, PAHs)

A number of inventories of the global production, usage and emission for selected POPs were developed recently. The information on available inventories including the estimates of POP usage, production and emissions is presented in the table. Some of the inventories contain the data on total emissions for particular countries, their temporal variations, and gridded emission data. In particular, PCB emission inventory presents a full set of information on spatial variations of emissions on global scale and their temporal variations with annual resolution in period 1930-2100. Gridded emissions data are also available for a-HCH and y-HCH for 1980, 1990, and 2000. For other POPs the estimates of total emissions of particular countries are available. Nevertheless this information can also be used for modeling purposes. Below a description of available POP emission inventories is presented.

Chemical Use Period of time Global use/ production, kt Global emissions, kt Reference
(as kg TEQ)
By-products 1990   50 Brzuzy and Hites, 1996
1995   10.5
8.3 - 36
UNEP, 1999
HCB Pesticide
Mid 1990   0.023
0.012 - 0.092
Bailey, 2001


Various 1930-2000
up to 2100
1320 7.71
0.44 - 91.7
Breivik et al., 2002a
Breivik et al., 2002b
Breivik et al., 2007
PAHs By-products 2004   520 Zhang and Tao, 2008




For the evaluation of global scale transport and boundary conditions, required for regional EMEP modelling, expert estimates of global PAH emissions, produced by the research group of Peking University [Shen et al., 2013], were applied. Global PAH emission inventories with 0.1°x0.1° spatial resolution were developed using a bottom-up approach for the period from 1960 to 2014. Distribution of global emission fluxes of the sum of 4 PAHs and B(a)P is presented in Figure.

Spatial distribution of global annual emissions of 4PAH (a) and B(a)P (b), g/km2/y, with spatial resolution 1°x1°, applied in the model simulations for 2016

A global historical emission inventory for selected PCB congeners (22 PCB congeners) was prepared by Breivik et al. (2002a). The available data of the historical production of PCBs and the chemical composition of various technical mixtures have been compiled from the literature. Information of imports, exports and consumption, as well as restrictions on production and import has further been analyzed for individual countries. The estimates account for a reported historical global usage of approximately 1300 kt of PCBs. The results suggest that almost 97% of the global historical use of PCBs have occurred in the Northern Hemisphere (Breivik et al., 2002a).

In the second part of this study, an attempt has been made to estimate the historical anthropogenic emissions as a direct result of the widespread usage of PCBs for a period of 70 years. The historical emission of 22 PCB congeners is estimated to be between 440 and 91722 t, with 7709 as the average value (Breivik et al., 2002b). In accordance with this three emission scenario were developed, in particular, low emission, average emission, and high emission scenario. In spite of the major uncertainties, the study is considered as a first important step towards the establishment of a global PCB emission inventory with congener resolution. The importance of temperature as a key parameter in controlling and affecting both the absolute value of PCB emissions as well as the PCB emission profile is emphasized.

The third part of this study presents an update of preceding PCB emission database. This work takes into account updated information on PCB production, as well as new data on the chemical composition of various technical mixtures for which less information had been available earlier. The methodology to estimate temporal trends of PCB usage is improved. The authors have also included projected emissions up to year 2100, which can facilitate predictions of future environmental exposure. The national emission data for each of the 114 countries considered is spatially resolved on a 1º—1º grid for each congener and year, using population density as a surrogate (Breivik et al., 2007), as indicated for PCB-153 for 2000 (kg/cell/y).


Spatial distribution of global annual emissions of PCB-153, g/km2/y (c) with spatial resolution 1°x1°, used in model simulations for 2016



HCB had been actively used in industry and for agricultural purposes. HCB was first introduced in 1933 as a fungicide on the seeds of onions, sorghum and crops such as wheat, barley, oats and rye. It is believed that agricultural use of HCB dominated its emissions during the 1950s and 1960s. The peak of HCB production was the late 1970s and early 1980s worldwide. Production of HCB declined as a result of restrictions on its use. The banning of HCB for agricultural use in the 1970s removed the largest single primary source of HCB in the environment (Barber et al., 2005). Although HCB production has ceased in most countries, it is still being generated inadvertently as a by-product and/or impurity in several chemical processes, such as the manufacture of chlorinated solvents, chlorinated aromatics and pesticides (Jacoff et al., 1986). HCB can be also released to the environment by incomplete combustion and from old dumpsites.

Global emission estimate for HCB was presented in (Bailey, 2001). This study presents the information from a variety of sources of HCB emissions and gives a quantitative estimate of the global HCB emissions in the mid 1990s. The best estimates of global HCB emissions from different categories of sources are as follows: pesticides application - 6500 kg/y; manufacturing - 9500 kg/y; combustion – 7000 kg/y, including 500 kg from biomass burning. This adds up to total current HCB emissions of approximately 23,000 kg/y with an estimated range 12,000-92,000 kg/y (Bailey, 2001). The author suggests that a substantial portion of HCB measured in the atmosphere is thought to come from volatilization of y-HCB on the soil from past contamination along with unidentified sources. Globally, the magnitude of these secondary emissions is currently not known, but this is a potentially very important aspect of the current global HCB cycle (Barber et al., 2005).

To describe contemporary levels of HCB pollution experimental scenario of recent and historical global HCB emissions for the period from 1945 to 2013 was developed [Shatalov et al., 2010]. Following available information the application of HCB, largely in agricultural activities as a fungicide, was started from 1945 and reached its maximum in 1980-s. Starting from that period the agricultural use of HCB was banned in many countries world-wide resulting in considerable decrease of primary emission of HCB and the increase of relative importance of HCB re-volatilization. According to constructed scenario emissions of HCB decreased from its maximum in 1978 to 2013 by more than 400 times.

The largest HCB emission fluxes took place in Southern and Eastern Asia. Releases of HCB to the atmosphere in the EMEP region were relatively lower.


Spatial distribution of global annual emissions of HCB, ng TEq/m2/y (c) with spatial resolution 1°x1°, used in model simulations for 2016





PCDD/Fs enter the environment as by-products of industrial processes. The most significant type of sources of PCDD/F emission to the atmosphere is low-temperature, incomplete incineration of chlorine-containing materials such as plastics. Other major sources include thermal processes, such as motor vehicle fuel combustion in countries where leaded fuel containing chlorine scavengers is still used, and metallurgical industries. PCDD/Fs are also trace contaminants in chlorophenoxy herbicides, PCB formulations, and chlorophenol wood preservatives (AMAP, 2004). Rough estimates of the global emissions of PCDD/Fs are available from Brzuzy and Hites (1996). The authors calculated annual global emission to approximately 50 kg TEQ for 1990. They based on assumptions that the global PCDD/F emissions from different categories of sources are as follows: waste incineration - 20 kg TEQ/y; cement kilns - 17 kg TEQ/y; manufacturing - 7 kg TEQ/y, and combustion - 6 kg TEQ/y.

In addition, an overview of available emission inventories for PCDD/Fs on global level can be found in (UNEP, 1999). This report includes estimates of PCDD/F emissions for Western Europe, Northern America (Canada and the United States). For Asian region there is only one inventory for Japan, which covers a few source categories, and the estimates for emissions from waste incinerators from South Korea. From the southern hemisphere, only Australia has estimated annual emissions. Japan and the United States were considered the largest emission sources of PCDD/Fs (4 kg TEQ/y and 2.7 kg TEQ/y, respectively). The average estimate of annual “global” PCDD/F emission is amounted to approximately 10.5 kg TEQ for 1995. The lower estimate is about 8.3 kg TEQ/y and the upper estimate is approximately 36 kg TEQ/y. The waste incineration is still the major emitter of dioxins and furans to the atmosphere with contribution to the total emission reaching almost 70 %. The majority of PCDD/PCDF emissions from this sector are due to the municipal (and to a lesser extent to the hazardous) waste incinerators located in Japan. This source sector alone is responsible for almost 34 % of the total PCDD/PCDF inventory to air from identified sources so far (UNEP, 1999).

To evaluate global transport and fate of PCDD/Fs experimental emission scenario was constructed on the basis of the information on dioxins and furans releases compiled under UNEP SC. National inventories of annual PCDD/F emissions were available for 68 countries representing the level of emissions during the recent decade (Fiedler, 2007; Fiedler et al., 2012; UNEP, 2013). Compiled inventories covered main modes of entry of emissions to the environment (the atmosphere, land, water, residues, and products), among which major releases were indicated to the atmosphere (45%). Analysis of this information, performed by Cao et al. (2013) permitted to reveal correlation between the intensity of PCDD/F emissions in the countries and their economic status. Obtained regression relationship between the national PCDD/F emissions and the data on GDP and total population of the countries was applied to estimate emissions for the other countries, for which the information on emissions was not available. Thus based on this information amount of PCDD/Fs released to the environment annually was estimated to about 80 kg TEQ. Spatial distribution of PCDD/F emissions was generated using the population density as a surrogate.



Spatial distribution of global annual emissions of PCDD/F, ng TEq/m2/y (c) with spatial resolution 1°x1°, used in model simulations for 2016



AMAP, 2004 AMAP Assessment 2002: Persistent Organic Pollutants in the Arctic. Arctic monitoring and Assessment Programme (AMAP), Oslo, Norway. xvi+310 pp.

AMAP/UNEP [2013] Technical Background Report for the Global Mercury Assessment 2013. Arctic Monitoring and Assessment Programme, Oslo, Norway / UNEP Chemicals Branch, Geneva, Switzerland. vi + 263 pp. (http://www.amap.no/documents/download/1265).

AMAP/UNEP [2008] Technical background report to the global atmospheric mercury assessment. Arctic Monitoring and Assessment Programme / UNEP Chemicals Branch. 159 pp. (http://www.chem.unep.ch/mercury/).

Bailey R.E., 2001 Global hexachlorobenzene emissions. Chemosphere 43, pp.82-167.

Barber J.L., Sweetman A.J., Wijk D. and K.C.Jones 2005 Hexachlorobenzene in the global environment: Emissions, levels, distribution, trends and processes. Sci, Total Environ., 349, pp.1-44.

Breivik K., Sweetman A., Pacyna J.M. and K.C.Jones, 2002a Towards a global historical emission inventory for selected PCB congeners - a mass balance approach. 1. Global production and consumption. Sci.Total Environ., 290, pp.181-198.

Breivik K., Sweetman A., Pacyna J.M. and K.C.Jones, 2002b Towards a global historical emission inventory for selected PCB congeners - a mass balance approach. 1. Emissions. Sci.Total Environ., 290, pp.199-224.

Breivik K., Sweetman A., Pacyna J.M. and K.C.Jones, 2007 Towards a global historical emission inventory for selected PCB congeners - a mass balance approach. 3. An update. Sci.Total Environ., 377, pp.296-307.

Brzuzy L.P. and R.A. Hites, 1996 Global Mass Balance for Polychlorinated Dibenzo-p-dioxins and Dibenzofurans. Environ. Sci. Technol. 30, pp.1791-1804.

Cao Z., H. Fiedler, B. Wang, T. Zhang, G. Yu, J. Huang, S. Deng [2013] Economic status as a determinant of national PCDD/PCDF releases and implications for PCDD/PCDF reduction, Chemosphere 91, 328-335.

Fiedler H. [2007] National PCDD/PCDF release inventories under the Stockholm Convention on Persistent Organic Pollutants, Chemosphere 67, S96–S108.

Fiedler H., Cao Z., Huang J., Wang B., Deng S., and Yu G. [2012] PCDD/PCDF inventories 1990 vs. 2012. Organohalogen Compounds, 74, 1521-1524.

IHPA, 2006 The legacy of lindane HCH isomer production. Annexes. A global overview of residue management, formulation and disposal by John Vijgen. http://www.ihpa.info/library_access.php.

Jacoff F.S., Scarberry R. and D.Rosa 1986 Source assessment of hexachlorobenzene from the organic chemical manufacturing industry. In: Morris C.R., Cabral

Pacyna J. M., Scholtz M. T. and Y.-F. Li (1995) Global Budgets of Trace Metal Sources, Environmental Reviews, 3, 145-159.

Pacyna, J.M. and Pacyna, E.G. (2001): An assessment of global and regional emissions of trace metals to the atmosphere from anthropogenic sources worldwide. Environmental Reviews, 9: 269-298.

Pacyna J, Pacyna E, Steenhuisen F, Wilson S. (2003) Mapping 1995 global anthropogenic emissions of mercury. Atmos. Environ. 37(S1), S109-S117.

Pacyna E. G., Pacyna J. M., Steenhuisen F. and Wilson S. (2006) Global anthropogenic mercury emission inventory for 2000. Atmos. Environ. 40(22), 4048-4063

Sang S., Petrovic S. and V.Cuddeford, 1999 Lindane A Review of Toxicity and Environmental Fate, prepared by World Wildlife Fund Canada.

UNEP, 1999 Dioxin and Furan Inventories. National and Regional Emissions of PCDD/PCDF. Prepared by UNEP Chemicals. Geneva, Switzerland.

UNEP [2013] Toolkit for Identification and Quantification of Releases of Dioxins, Furans and Other Unintentional POPs.

Zhang Y. and S.Tao, 2008 Global atmospheric emission inventory of Polycyclic Aromatic Hydrocarbons (PAHs) for 2004. Atmos. Environ.