- Email: [email protected]
Human influence on the global cycling of trace metals
Palaeogeography, Palaeoclimatology, Palaeoecology (Global and Planetary Change Section), 82 (1990): 113-120
113
Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands
Human influence on the global cycling of trace metals 1 JEROME O. NRIAGU National Water Research Institute, Box 5050, Burlington, Ont. L7R 4A6 (Canada) (Received April 24, 1989; accepted April 28, 1989)
Introduction
An objective of this workshop is the exploratory description and prediction of the trajectories of ecosystems or biogeochemical processes in time. The available information strongly suggests that the trajectories of the global and regional cycles of the trace metals have become markedly influenced by mankind. At present, the human influence may be manifested by increased accumulation and concentration in different levels of an ecosystem, by changes in rates of transfer of trace metals between the environmental reservoirs, and by hypeinopenia in the sensitive organisms and populations. Acute ecosystem stress has been demonstrated at the local scale but the chronic effects associated with regional and global pollution with toxic metals cannot yet be diagnosed. Nevertheless, recent studies show t h a t there is no threshold concentration for adverse effects from universal exposure to toxic metals, and the biochemical mechanisms mediating the low-level toxicity in biological and natural systems remain unclear. If the current rates of emission continue, the levels of trace metals in many environmental compartments are bound to become increasingly
1 Discussion paper prepared for the IUGS Workshop on Past Global Changes Interlaken, Switzerland, April 24-28, 1989. 0921-8181/90/$03.50 © 1990 - Elsevier Science Publishers B.V.
stressful to many organisms over an extended period of time. Historical
Contaminating the environment with toxic metals is an old human pursuit. Before the Industrial Revolution, mining and smelting operations represented the major source of trace metals in the environment, and the available records leave no doubt as to the severity of pollution in some of the classical mining regions. Xenophon (Book 3, Verse 6), for example, noted that the district around the famous ancient silver mines of Laurion (in Greece) was too contaminated [unhealthy] to warrant a visit by Ariston's son, Glaucon. The ancient Greek poet Lucretius (De r e r u m n a t u r a , Book VI) was particularly concerned about the health effects of the noxious emissions from precious metal mines: And when there is mining for veins of gold and silver Which men will dig far deep down in the earth What stenches arise, as at Scaptensula! How deadly are the exhalations of gold mines! You can see the ill effects in the miners' complexions. All these exhalations come from the earth And are breathed forth into the open light of day.
Vitruvius (Book 8, verse 3) spoke about extensive water pollution around mines, while Pliny (Book 33, verse 31) noted that emissions from mines and smelters were dangerous to all animals and especially to dogs. Del Mar (1880) main-
114
tained that the concern for environmental quality was as important as the social and economic factors in the interdiction of mining operations near ancient cities, a perfect example being the Roman edict forbidding any mining activities in Italy. Will Durant shared the same sentiments and in his highly acclaimed work, The Story o/ Civilizations, aptly observed that "Laurion pays the price of wealth it produces, as mining always pays the price for metal industry; plants and men wither and die from furnace fumes, and the vicinity of the works becomes a scene of desolation". In ancient times, the dispersion of trace metals from polluting sources was local, or at most regional, in scope. The mine workings were done on a small scale but the uncontrolled smelting in open fires often resulted in severe local contamination (as noted above). Evidence of pollution in ancient times comes from the trace metal profiles of peats in the mining district of Giordano Valley near Bristol which show elevated concentrations suggesting that the Roman smelting operations produced measurable flux of pollutants to the local environment (Martin et al., 1979; Livett, 1988). Elevated lead concentrations in the Featherbed Moss also attest to the environmental impacts of Roman exploitation of the Derbyshire lead deposits (Lee and Tallis, 1973). During the 16 Century, the development of large furnaces equipped with tall stacks (see Agricola, 1555) forever changed the sphere of influence of the smelters. Increased trace metal deposition dating back to the Middle Ages which has been recorded in peat profiles at Giordano Valley, Featherbed Moss, Glenshieldaig (nortwest Scotland) and in Lake Windermere sediment suggests that emissions from smelters in Britain were already affecting the regional environment (Livett, 1988). Furthermore, the profiles of trace metals in peat at Dravel Mose, Denmark (Aaby et al., 1979), the ice cores from Jotunheimen Mountains, southern Norway (Jaworowski et al., 1975), and lake sediments from southern parts of Norway and Finland (Davis et al., 1983; Verta et al., 1989) all point to the fact that by the middle of the 17th century,
J.O. NRIAGU
the pollutant trace metals released from industries in Britain and central Europe were reaching the most remote regions of Scandinavia. In their study of ice cores from Greenland, Murozumi et al. (1969) also found that accelerated deposition of lead from anthropogenic sources began in the middle the 18th Century. The health effects of high levels of toxic metals in the air were of some concern even during the Middle Ages. In his 1661 tract entitled Fumifugium, John Evelyn condemned air pollution from coal combustion: "New Castle cole, as an expert Physician affirms, causeth consumptions, phthisicks, and indisposition of the lungs, not only by the suffocating abundance of smoke, but also by its virulency: for all subterrany fuel hath a kind of virulent or arsenical vapor rising from it" (Cited in Lenihan, 1988, p. 121). In John Evelyn's days, arsenic was synonymous with poison, and his perception of the dangers of toxic metal pollution from coal combustion was remarkable. Analyses of recent layers of peat, ice and aquatic sediments have provided historical records, dating back to the early 1800s, of ubiquitous trace metal pollution of increasing severity especially in the northern hemisphere (see MARC, 1985; Levitt, 1988). Since the middle of the last century, the production and discharge of trace metals in the environment have been increasing almost at a logarithmic scale (Nriagu, 1979). It has been estimated that between 1850 and 1900, the worldwide industrial emissions of Cd, Cu, Pb, Ni and Zn to the atmosphere averaged about 380, 1800, 22,000, 240 and 17000 tonnes per year respectively (Nriagu, 1979). Between 1900 and 1980, the industrial emission rates for Cd, Cu, Pb, Ni and Zn increased by 8-, 6-, 9-, 51-, and 8-fold respectively (Fig. 1). During the first 80 years of this century, the mine outputs of Cu, Pb, Ni and Zn have totalled about 250, 160, 17 and 185 million tonnes respectively (Nriagu, 1979, 1988). Considering that, ultimately, most of the mine-produced metals will also be dispersed in the biosphere, there can be no doubt that mankind is playing a key role in the geochemical re-distribution of the trace elements. The paper by
HUMAN
INFLUENCE
ON THE GLOBAL CYCLING
OF TRACE
115
METALS
Emissions of Cu, Pb, Zn; 1900 to 1980
5000
® Cu
I
4000
m £0
uJ o v tO 03
3000
Pb
I
Zn
2000
E UJ
1000
0
1900
1920 1910
1940 1930
1960 1970
1950
Decade
Consumption of Cu, Pb, Zn; 1900 to 1980
90[- ® Cu
80-
1
70-
Pb
O'J
m
60U.I 0 tO
E "3 ¢0
Zn
504030-
0
20-
0
1930
1910
1920
1950 1940
1970 1960
1980
Decade Fig. 1. A-C. Historical trends in global emission and production of trace elements (from Nriagu, 1979).
I
116
J.O. NRIAGU
Consumption and Emissions of
Cadmium (1900-1980) 160 -
[-----q
©
Consum.
140 ~D W O
120 -
Emissi.
v cO
100 -
co .i
E
UI c-
o cL
8060-
E cO
0
2O 0
1900
1920 1910
1940 1930
1960 1950
1970
Decade Fig. 1. (continued).
Pacyna and Winchester (this issue) demonstrates the pervasive nature of trace metal pollution even in the Arctic region. W o r l d w i d e e m i s s i o n s of trace m e t a l s into t h e air, w a t e r a n d soils
The following discussions are based on global inventories of emissions of trace metals from natural and anthropogenic sources that have recently been published (Nriagu and Pacyna, 1988; Nriagu, 1989). The data strongly suggest t h a t mankind has destabilized the steady-state (pretechnological) geochemical cycles of the trace metals and that the various components of the biosphere are gaining large quantities of toxic metals at a rate that is determined by the anthropogenic inputs.
Perturbation of the atmosphere cycle From the published inventories (Table I), the ratios of anthropogenic to natural fluxes of trace
metals to the atmosphere have been estimated to be 18 for Pb, 4.8 for Cd, 2.3 for Zn, 1.7 for V, 1.6 for As, 1.4 for Hg and Ni, 1.0 for Sb, 0.85 for Cu, 0.71 for Cr, 0.66 for Mo, 0.63 for Se and 0.08 for Mn. With the exception of Mn, the fluxes from industrial sources either exceed or are comparable to the emissions from natural sources implying that mankind has become the key eleTABLE I N a t u r a l versus anthropogenic emissions of trace metals ( × 10 6 k g / y r ) Element As Be Cd Cr Hg Ni Pb Se V
Anthropogenic ~ 19 ? 7.6 30 3.6 56 332 6.3 86
Natural b 12 ? 1.3 44 2.5 30 12 9.3 28
TotM 31 8.9 74 6.1 86 344 16 114
a F r o m Nriagu and Pacyna (1988). b From Nriagu (1989).
117
HUMAN INFLUENCE ON THE GLOBAL CYCLING OF TRACE METALS
ment in the global and regional atmospheric cycling of the trace elements. Most of the industries are located in the northern hemisphere and the perturbation of the trace metal cycles in this region should be more profound. Field measurements have de m ons t r at ed causal relationships between emissions and ambient concentrations of trace metals in air in many parts of the world, and have thereby confirmed the over-riding importance of anthropogenic inputs on the regional and global atmospheric fluxes of the trace elements (see Nriagu and Davidson, 1986). In a recent extensive survey of the global distribution of Pb in the marine atmosphere, Volkening et al. (1988) found the highest concentrations near the highly industrialized areas such as the west coast of Europe and the coastal areas of eastern South America. The lowest concentrations were recorded over the Ekstrom ice shelf in Antarctica. A prior study of ice cores from Antarctica had found a recent 4-fold increase in Pb concentration (Boutron and Patterson, 1987) which would suggest that no place on earth is now free of lead pollution. It should be noted, however, that there has been a significant decrease in the Pb fluxes which can be attributed to the reduced use of leaded gasoline. As to be expected, industrial emissions have drastically changed the concentrations of trace elements in the rainfall. Assuming t hat 70% of all the emissions to the atmosphere (Table I) is deposited on the terrestrial environment, and t h a t the volume of rainfall on land is 1.1 x 1017 1
T A B L E II Concentrations (ng/1) of trace metals in atmospheric precipitation Element
Predicted a
Urban b
Rural b
Remote b
As Cd Cr Pb Hg Ni V
200 55 570 2200 36 545 730
< 1000 < 700 < 3000 < 25000 65 < 12000 < 6000
285 110 380 4500 20 ? 1200
19 8 50 220 2 ? 160
a See text for the calculation. b Compiled from the' literature.
TABLE III Global emissions of trace metals into the atmosphere, water and soils (in i000 metric tonnes per year) a Element
Air
Water
Soil
As Cd Cr Cu Hg Mn Mo Ni Pb Sb Se V Zn
19 7.6 30 35 3.6 38 3.3 56 332 3.5 3.8 86 132
41 9.4 142 112 4.6 262 11 113 138 18 41 12 226
82 22 896 954 8.3 1670 88 325 796 26 41 132 1372
a From Nriagu and Pacyna (1988).
(Berner and Berner, 1987), the average concentrations of the trace elements in bulk precipitation have been calculated and shown in Table II. It would seem reasonable t hat the calculated concentrations are comparable to those of rural areas and lie between the values observed in urban and remote locations. A notable feature in Table II is t hat the levels of trace metals in some urban areas now exceed the levels considered safe in drinking water. This should be a matter of some concern in developing countries where rain water is consumed by a large number of neople in the urban areas.
Pollution of lakes and rivers Estimates of the worldwide discharges of trace metal pollution into the aquatic environments are shown in Table III. It should be noted t hat industrial discharges of waste effluents are often localized so t h a t the atmosphere is the dominant contributor to the trace metal economy of many lakes in rural and remote areas. As a first approximation, I have assumed t hat atmospheric inputs were responsible for 20-30% of the trace metals in lakes in prehistoric times; the dominant fraction was derived from rock weathering. By assuming also t h a t 5-10% (a high fraction in fact) of the global emissions of trace metals from natural sources got into the lakes and river, the baseline concentrations of trace elements have
118
J.O. NRIAGU
T A B L E IV Calculated average (Total) concentrations of trace elements (ng/1) in lakes and rivers Element
Background concentration
Median
Industrial contribution
Median
As Cd Cr Cu Hg Mo Ni Pb Sb Se V Zn
11 - 31 2.4- 7.2 37 -110 29 - 86 1.2- 3.5 2.8- 8.5 29 - 87 10 - 35 4.7- 14 11 - 32 3.1- 9.2 58 -174
(21) (4.8) (74) (58) (2.4) (5.6) (58) (22) (9.4) (22) (6.2) (116)
15 - 46 1.7- 5.0 56 -170 36 -108 3.2- 9.6 3.8- 12 38 -115 15 - 46 3.1- 9.2 12 - 36 36-108 58 -173
(31) (3.4) (113) (72) (6.4) (7.9) (76) (31) (6.2) (24) (72) (115)
been estimated (Table IV). For comparison, the average pollution component of the trace metal burden in lakes and rivers have also been calculated (Table IV) assuming that 10-30% of the total anthropogenic discharges into the aquatic ecosystems goes into lakes and rivers with global volume of 1.3 x 10 i7 1 (Berner and Berner, 1987) and that the average retention time for trace metals is 4 months. The total concentrations of the trace metals in lakes and rivers are estimated as the sum of the baseline and pollution burdens. For most of the trace elements, the calculated baseline concentrations are similar to the levels found in the open ocean (see Wong et al., 1983). The very low baseline concentrations imply that small changes in pollutant metal inputs can significantly perturb the distribution of trace metals in fresh water ecosystems. The calculations (Table IV) suggest that the anthropogenic contributions now constitute about 50% or more the total trace metal burdens in lakes in rivers. The determination of the pollution component in lakes and rivers has been hampered by the fact that the available data bases are severely compromised by contamination artifacts. I am encouraged by the fact that the predicted concentrations are in general agreement with some of the most recent concentrations that have been obtained using the ultra-clean laboratory technique.
The concentrations of trace metals in rainfall in the urban and rural areas (Table II) are generally higher than the levels in lakes (Table IV). Apparently, lakes have developed effective self-cleaning processes that involve such factors as pH, water renewal time, complexation capacity, p a r t i c u l a t e concentration, biological processes (such metal methylation and assimilation), bioturbation of sediments, etc. Our studies have clearly demonstrated that the cycling of trace metals in lakes is strongly dominated by particulate dynamics in the water column (Wong and Nriagu, 1984; Nriagu and Wong, 1986). I believe t h a t the trace metal content of the seston (suspended particulates) may be a better indicator of the level of metal pollution than the concentration of the trace metals in the water. Several studies have shown that aquatic ecosystems respond rather quickly to changes in anthropogenic inputs of trace metals. For example, the surface ocean waters off the northeastern United States have shown a 30-40% decline in lead concentration due to the phase-out of leaded gasoline. The isotopic composition of Pb in the surface waters of the Great Lakes is very similar to the isotopic signature of the ambient aerosol (Flegal et al., 1989); the close correlation stems from the fact that the life-time of lead in the epilimnion is only a few days and the Pb concentration is controlled by recent atmospheric inputs. Also, there has been a drastic shift in the isotopic composition of Pb in Lake Erie sediments which reflects the declining role of atmospherically derived automotive lead (Mueller, 1988). The installation of a tall stack at the C u / N i smelters in Sudbury (Ontario) has reduced the rates of accumulation of trace metals in the local lake sediments (Nriagu and Rao, 1987), while the reduced consumption of leaded gasoline has engendered the recent decline in Pb accumulation in aquatic sediments which has been documented in many parts of the world (Levitt, 1988). Influence on m a r i n e trace m e t a l cycle
The current massive industrial emissions of trace metals have greatly increased the atmo-
HLrMAN I N F L U E N C E ON T H E GIX)BAL CYCLING OF T R A C E METALS
spheric fluxes of trace metals into the ocean (see above for the regional distribution of Pb in m a r i n e atmosphere). F u r t h e r m o r e , model calculations (above) suggest t h a t t h e river-borne delivery of trace metals to the oceans has p r o b a b l y doubled in recent times. To w h a t extent h a v e the n a t u r a l geochemical cycles been p e r t u r b e d b y the anthropogenic inputs? In addition to the mass balance considerations, there are several lines of evidence which show t h a t the anthropogenic inputs h a v e completely overwhelmed the n a t u r a l m a r i n e cycle of lead: (a) there is a 15-fold increase in Pb content of recent coral growth layers c o m p a r e d to those of a century ago; (b) the m a r k e d change in the f o r m of Pb profiles in ocean w a t e r column reflects the strong a t m o s p h e r i c source intensity; (c) a 26-fold difference in the Pb concentrations in mixed surface layers has been noted between the N o r t h Atlantic (polluted) and the S o u t h Pacific (relatively unpolluted); and (d) a measurable shift in the isotopic signature of Pb in surface waters towards the isotopic ratio of industrial lead has also been reported (see P a t t e r son and Settle, 1986, for detailed discussion of oceanic lead pollution). A recent analysis of the Atlantic Ocean sediments show a s h a r p increase in the Pb content of the surficial layer which has been a t t r i b u t e d to increased influx of anthropogenic lead (Veron et al., 1987). T h e h u m a n influence on the oceanic cycling of the other trace elements has not been fully assessed a l t h o u g h industrial activities are also transferring large quantities of these trace elem e n t s into the marine environment.
Contamination o/soils Soils are the u l t i m a t e sink for p o l l u t a n t metals in the terrestrial e n v i r o n m e n t (Table III). Assuming t h a t the discharges from industrial sources are dispersed uniformly over the land area (150 × 1012 m2), the annual rates of m e t a l addition to soils are estimated to v a r y from 0.1 g h a - 1 y r - 1 for Cd, to a b o u t 5 g h a - 1 y r - 1 for Pb, Cu and Cr, to over 6 g h a - 1 y r - i for Zn and Mn. T h e large background reservoir of trace metals in soils often obscures such high industrial load-
119
ings and increased concentrations are rarely observed except in u r b a n soils and near m a j o r point sources of pollution. Several studies h a v e d o c u m e n t e d an increase in the Cd content of soils following continued, long-term applications of p h o s p h a t e fertilizers to agricultural soils (see Adriano, 1986). Since the 1850s, an increase of 27-55% in the Cd burden of the soil plough layer (0-23 cm) has been reported a t the R o t h a m s t e d E x p e r i m e n t a l Station, a semi-rural location in s o u t h e a s t England (Jones and Johnston, 1989). Such a change in solid Cd burden of 1.9-5.4 g ha 1 y r - 1 has been a t t r i b u t e d solely to increased atmospheric deposition of this element (Jones and Johnston, 1989). T h e r e is some evidence t h a t large areas of J a p a n and central E u r o p e have become severely c o n t a m i n a t e d with c a d m i u m (Asami, 1984; Kloke et al., 1984). Increased concentration of Cd in soils is of h u m a n health concern because (a) there a p p e a r s to be a positive relationship between the Cd levels in soils and plants, and (b) crops represent the m a j o r source of dietary Cd for the general population.
References Aaby, B., Jacobsen, J. and Jacobsen, O.S. Dan. Geol. Unders. Arbog, 1978: 5-68. Adriano, D.C., 1986. Trace Elements in the Terrestrial Environment. Springer, New York, N.Y. Agricola, G., 1555. De re Metallica (Translated 1950, by H.C. Hoover and L.H. Hoover, Dover Publications, New York). Asami, T., 1984. In: J.O. Nriagu (Editor), Changing Metal Cycles and Human Health. Springer, Berlin, pp. 95-11. Berner, E.K. and Berner, R.A., 1987. The Global Water Cycle. Prentice-Hall, Englewood Cliffs, N.J. Boutron, C.F. and Patterson, C.C., 1987, J. Geophys. Res., 92: 8454-8464. Davis, R.B., Norton, S.A., Hess, C.T. and Brakke, D.F., 1983. Hydrobiologia, 103: 113-123. Del Mar, A., 1880. History of Precious Metals. Bell, London. Flegal, A.R., Nriagu, J.O., Coale, K.H. and Niemeyer, S., 1989. Nature, 339: 455-458. Jones, K.C. and Johnston, A.E., 1989. Environ. Pollut. 57: 199-216. Kloke, A., Sauerbeck, D.R. and Vetter, H., 1984. In: J.O. Nriagu (Editor), Changing Metal Cycles and Human Health. Springer, Berlin, pp. 113-141. Jaworowski, Z., Bilkiewicz, J., Dobosz, E. and Wodkiewicz, L., 1975. Environ. Pollut., 9: 305-315. Lee, J.A. and Tallis, J.H., 1973. Nature, 245: 216-218. Lenihan, J., 1988. The Crumbs of Creation. Hilger, Bristol.
120 Livett, E., 1988. Advanc. Ecol. Res., 18: 65-176. MARC, 1985. Historical Monitoring. Monit. Res. Assess. Cent., Chelsea College, Univ. London. Martin, M.H., Coughtrey, P.J. and Ward, P., 1979. Proc. Bristol Nat. Soc., 37: 91-97. Mueller, E.L., The Variation in Lead Isotope Ratios in Lake Erie Sediments. Thesis. Dept. Physics, Univ. Toronto. Murozumi, M., Chow, T.J. and Patterson, C.C., 1969. Geochim. Cosmochim. Acta, 33: 1247-1294. Nriagu, J.O., 1979. Nature, 279: 409-411. Nriagu, J.O., 1988. Environ. Pollut., 50: 139-161. Nriagu, J.O., 1989. Nature, 338: 47-49. Nriagu, J.O. and Davidson, C.I. (Editors), 1986. Toxic Metals in the Atmosphere. Wiley, New York, N.Y. Nriagu, J.O. and Wong, H.K.T. 1986. Water, Air and Soil Pollut., 31: 999-1006.
J.O. NRIAGU Nriagu, J.O. and Pacyna, J.M., 1988. Nature, 333: 134-139. Nriagu, J.O. and Rao, S.S., 1987. Environ. Pollut., 44: 211218. Patterson, C.C. and Settle, D.M., 1987. SEAREX Newsl., 10 (1). Veron, A., Lambert, C.E., lsley, A., Linet, P. and Grousset, F., 1987. Nature, 326: 278-281. Verta, M., Tolonen, K. and Simola, H., 1989. Sci. Total Environ., 87: 1-18. Volkening, J., Baumann, H. and Heumann, K., 1988. Atmos. Environ., 22: 1169-1174. Wong, H.K.T. and Nriagu, J.O., 1984. Trace Substances in Environ. Health, 18: 416-426. Wong, C.S., Boyle, E.A., Bruland, K.W., Burton, D. and Goldberg, E.D. (Editors), 1983. Trace Metals in Sea Water. Plenum, New York, N.Y.
113
Elsevier Science Publishers B.V., Amsterdam - Printed in The Netherlands
Human influence on the global cycling of trace metals 1 JEROME O. NRIAGU National Water Research Institute, Box 5050, Burlington, Ont. L7R 4A6 (Canada) (Received April 24, 1989; accepted April 28, 1989)
Introduction
An objective of this workshop is the exploratory description and prediction of the trajectories of ecosystems or biogeochemical processes in time. The available information strongly suggests that the trajectories of the global and regional cycles of the trace metals have become markedly influenced by mankind. At present, the human influence may be manifested by increased accumulation and concentration in different levels of an ecosystem, by changes in rates of transfer of trace metals between the environmental reservoirs, and by hypeinopenia in the sensitive organisms and populations. Acute ecosystem stress has been demonstrated at the local scale but the chronic effects associated with regional and global pollution with toxic metals cannot yet be diagnosed. Nevertheless, recent studies show t h a t there is no threshold concentration for adverse effects from universal exposure to toxic metals, and the biochemical mechanisms mediating the low-level toxicity in biological and natural systems remain unclear. If the current rates of emission continue, the levels of trace metals in many environmental compartments are bound to become increasingly
1 Discussion paper prepared for the IUGS Workshop on Past Global Changes Interlaken, Switzerland, April 24-28, 1989. 0921-8181/90/$03.50 © 1990 - Elsevier Science Publishers B.V.
stressful to many organisms over an extended period of time. Historical
Contaminating the environment with toxic metals is an old human pursuit. Before the Industrial Revolution, mining and smelting operations represented the major source of trace metals in the environment, and the available records leave no doubt as to the severity of pollution in some of the classical mining regions. Xenophon (Book 3, Verse 6), for example, noted that the district around the famous ancient silver mines of Laurion (in Greece) was too contaminated [unhealthy] to warrant a visit by Ariston's son, Glaucon. The ancient Greek poet Lucretius (De r e r u m n a t u r a , Book VI) was particularly concerned about the health effects of the noxious emissions from precious metal mines: And when there is mining for veins of gold and silver Which men will dig far deep down in the earth What stenches arise, as at Scaptensula! How deadly are the exhalations of gold mines! You can see the ill effects in the miners' complexions. All these exhalations come from the earth And are breathed forth into the open light of day.
Vitruvius (Book 8, verse 3) spoke about extensive water pollution around mines, while Pliny (Book 33, verse 31) noted that emissions from mines and smelters were dangerous to all animals and especially to dogs. Del Mar (1880) main-
114
tained that the concern for environmental quality was as important as the social and economic factors in the interdiction of mining operations near ancient cities, a perfect example being the Roman edict forbidding any mining activities in Italy. Will Durant shared the same sentiments and in his highly acclaimed work, The Story o/ Civilizations, aptly observed that "Laurion pays the price of wealth it produces, as mining always pays the price for metal industry; plants and men wither and die from furnace fumes, and the vicinity of the works becomes a scene of desolation". In ancient times, the dispersion of trace metals from polluting sources was local, or at most regional, in scope. The mine workings were done on a small scale but the uncontrolled smelting in open fires often resulted in severe local contamination (as noted above). Evidence of pollution in ancient times comes from the trace metal profiles of peats in the mining district of Giordano Valley near Bristol which show elevated concentrations suggesting that the Roman smelting operations produced measurable flux of pollutants to the local environment (Martin et al., 1979; Livett, 1988). Elevated lead concentrations in the Featherbed Moss also attest to the environmental impacts of Roman exploitation of the Derbyshire lead deposits (Lee and Tallis, 1973). During the 16 Century, the development of large furnaces equipped with tall stacks (see Agricola, 1555) forever changed the sphere of influence of the smelters. Increased trace metal deposition dating back to the Middle Ages which has been recorded in peat profiles at Giordano Valley, Featherbed Moss, Glenshieldaig (nortwest Scotland) and in Lake Windermere sediment suggests that emissions from smelters in Britain were already affecting the regional environment (Livett, 1988). Furthermore, the profiles of trace metals in peat at Dravel Mose, Denmark (Aaby et al., 1979), the ice cores from Jotunheimen Mountains, southern Norway (Jaworowski et al., 1975), and lake sediments from southern parts of Norway and Finland (Davis et al., 1983; Verta et al., 1989) all point to the fact that by the middle of the 17th century,
J.O. NRIAGU
the pollutant trace metals released from industries in Britain and central Europe were reaching the most remote regions of Scandinavia. In their study of ice cores from Greenland, Murozumi et al. (1969) also found that accelerated deposition of lead from anthropogenic sources began in the middle the 18th Century. The health effects of high levels of toxic metals in the air were of some concern even during the Middle Ages. In his 1661 tract entitled Fumifugium, John Evelyn condemned air pollution from coal combustion: "New Castle cole, as an expert Physician affirms, causeth consumptions, phthisicks, and indisposition of the lungs, not only by the suffocating abundance of smoke, but also by its virulency: for all subterrany fuel hath a kind of virulent or arsenical vapor rising from it" (Cited in Lenihan, 1988, p. 121). In John Evelyn's days, arsenic was synonymous with poison, and his perception of the dangers of toxic metal pollution from coal combustion was remarkable. Analyses of recent layers of peat, ice and aquatic sediments have provided historical records, dating back to the early 1800s, of ubiquitous trace metal pollution of increasing severity especially in the northern hemisphere (see MARC, 1985; Levitt, 1988). Since the middle of the last century, the production and discharge of trace metals in the environment have been increasing almost at a logarithmic scale (Nriagu, 1979). It has been estimated that between 1850 and 1900, the worldwide industrial emissions of Cd, Cu, Pb, Ni and Zn to the atmosphere averaged about 380, 1800, 22,000, 240 and 17000 tonnes per year respectively (Nriagu, 1979). Between 1900 and 1980, the industrial emission rates for Cd, Cu, Pb, Ni and Zn increased by 8-, 6-, 9-, 51-, and 8-fold respectively (Fig. 1). During the first 80 years of this century, the mine outputs of Cu, Pb, Ni and Zn have totalled about 250, 160, 17 and 185 million tonnes respectively (Nriagu, 1979, 1988). Considering that, ultimately, most of the mine-produced metals will also be dispersed in the biosphere, there can be no doubt that mankind is playing a key role in the geochemical re-distribution of the trace elements. The paper by
HUMAN
INFLUENCE
ON THE GLOBAL CYCLING
OF TRACE
115
METALS
Emissions of Cu, Pb, Zn; 1900 to 1980
5000
® Cu
I
4000
m £0
uJ o v tO 03
3000
Pb
I
Zn
2000
E UJ
1000
0
1900
1920 1910
1940 1930
1960 1970
1950
Decade
Consumption of Cu, Pb, Zn; 1900 to 1980
90[- ® Cu
80-
1
70-
Pb
O'J
m
60U.I 0 tO
E "3 ¢0
Zn
504030-
0
20-
0
1930
1910
1920
1950 1940
1970 1960
1980
Decade Fig. 1. A-C. Historical trends in global emission and production of trace elements (from Nriagu, 1979).
I
116
J.O. NRIAGU
Consumption and Emissions of
Cadmium (1900-1980) 160 -
[-----q
©
Consum.
140 ~D W O
120 -
Emissi.
v cO
100 -
co .i
E
UI c-
o cL
8060-
E cO
0
2O 0
1900
1920 1910
1940 1930
1960 1950
1970
Decade Fig. 1. (continued).
Pacyna and Winchester (this issue) demonstrates the pervasive nature of trace metal pollution even in the Arctic region. W o r l d w i d e e m i s s i o n s of trace m e t a l s into t h e air, w a t e r a n d soils
The following discussions are based on global inventories of emissions of trace metals from natural and anthropogenic sources that have recently been published (Nriagu and Pacyna, 1988; Nriagu, 1989). The data strongly suggest t h a t mankind has destabilized the steady-state (pretechnological) geochemical cycles of the trace metals and that the various components of the biosphere are gaining large quantities of toxic metals at a rate that is determined by the anthropogenic inputs.
Perturbation of the atmosphere cycle From the published inventories (Table I), the ratios of anthropogenic to natural fluxes of trace
metals to the atmosphere have been estimated to be 18 for Pb, 4.8 for Cd, 2.3 for Zn, 1.7 for V, 1.6 for As, 1.4 for Hg and Ni, 1.0 for Sb, 0.85 for Cu, 0.71 for Cr, 0.66 for Mo, 0.63 for Se and 0.08 for Mn. With the exception of Mn, the fluxes from industrial sources either exceed or are comparable to the emissions from natural sources implying that mankind has become the key eleTABLE I N a t u r a l versus anthropogenic emissions of trace metals ( × 10 6 k g / y r ) Element As Be Cd Cr Hg Ni Pb Se V
Anthropogenic ~ 19 ? 7.6 30 3.6 56 332 6.3 86
Natural b 12 ? 1.3 44 2.5 30 12 9.3 28
TotM 31 8.9 74 6.1 86 344 16 114
a F r o m Nriagu and Pacyna (1988). b From Nriagu (1989).
117
HUMAN INFLUENCE ON THE GLOBAL CYCLING OF TRACE METALS
ment in the global and regional atmospheric cycling of the trace elements. Most of the industries are located in the northern hemisphere and the perturbation of the trace metal cycles in this region should be more profound. Field measurements have de m ons t r at ed causal relationships between emissions and ambient concentrations of trace metals in air in many parts of the world, and have thereby confirmed the over-riding importance of anthropogenic inputs on the regional and global atmospheric fluxes of the trace elements (see Nriagu and Davidson, 1986). In a recent extensive survey of the global distribution of Pb in the marine atmosphere, Volkening et al. (1988) found the highest concentrations near the highly industrialized areas such as the west coast of Europe and the coastal areas of eastern South America. The lowest concentrations were recorded over the Ekstrom ice shelf in Antarctica. A prior study of ice cores from Antarctica had found a recent 4-fold increase in Pb concentration (Boutron and Patterson, 1987) which would suggest that no place on earth is now free of lead pollution. It should be noted, however, that there has been a significant decrease in the Pb fluxes which can be attributed to the reduced use of leaded gasoline. As to be expected, industrial emissions have drastically changed the concentrations of trace elements in the rainfall. Assuming t hat 70% of all the emissions to the atmosphere (Table I) is deposited on the terrestrial environment, and t h a t the volume of rainfall on land is 1.1 x 1017 1
T A B L E II Concentrations (ng/1) of trace metals in atmospheric precipitation Element
Predicted a
Urban b
Rural b
Remote b
As Cd Cr Pb Hg Ni V
200 55 570 2200 36 545 730
< 1000 < 700 < 3000 < 25000 65 < 12000 < 6000
285 110 380 4500 20 ? 1200
19 8 50 220 2 ? 160
a See text for the calculation. b Compiled from the' literature.
TABLE III Global emissions of trace metals into the atmosphere, water and soils (in i000 metric tonnes per year) a Element
Air
Water
Soil
As Cd Cr Cu Hg Mn Mo Ni Pb Sb Se V Zn
19 7.6 30 35 3.6 38 3.3 56 332 3.5 3.8 86 132
41 9.4 142 112 4.6 262 11 113 138 18 41 12 226
82 22 896 954 8.3 1670 88 325 796 26 41 132 1372
a From Nriagu and Pacyna (1988).
(Berner and Berner, 1987), the average concentrations of the trace elements in bulk precipitation have been calculated and shown in Table II. It would seem reasonable t hat the calculated concentrations are comparable to those of rural areas and lie between the values observed in urban and remote locations. A notable feature in Table II is t hat the levels of trace metals in some urban areas now exceed the levels considered safe in drinking water. This should be a matter of some concern in developing countries where rain water is consumed by a large number of neople in the urban areas.
Pollution of lakes and rivers Estimates of the worldwide discharges of trace metal pollution into the aquatic environments are shown in Table III. It should be noted t hat industrial discharges of waste effluents are often localized so t h a t the atmosphere is the dominant contributor to the trace metal economy of many lakes in rural and remote areas. As a first approximation, I have assumed t hat atmospheric inputs were responsible for 20-30% of the trace metals in lakes in prehistoric times; the dominant fraction was derived from rock weathering. By assuming also t h a t 5-10% (a high fraction in fact) of the global emissions of trace metals from natural sources got into the lakes and river, the baseline concentrations of trace elements have
118
J.O. NRIAGU
T A B L E IV Calculated average (Total) concentrations of trace elements (ng/1) in lakes and rivers Element
Background concentration
Median
Industrial contribution
Median
As Cd Cr Cu Hg Mo Ni Pb Sb Se V Zn
11 - 31 2.4- 7.2 37 -110 29 - 86 1.2- 3.5 2.8- 8.5 29 - 87 10 - 35 4.7- 14 11 - 32 3.1- 9.2 58 -174
(21) (4.8) (74) (58) (2.4) (5.6) (58) (22) (9.4) (22) (6.2) (116)
15 - 46 1.7- 5.0 56 -170 36 -108 3.2- 9.6 3.8- 12 38 -115 15 - 46 3.1- 9.2 12 - 36 36-108 58 -173
(31) (3.4) (113) (72) (6.4) (7.9) (76) (31) (6.2) (24) (72) (115)
been estimated (Table IV). For comparison, the average pollution component of the trace metal burden in lakes and rivers have also been calculated (Table IV) assuming that 10-30% of the total anthropogenic discharges into the aquatic ecosystems goes into lakes and rivers with global volume of 1.3 x 10 i7 1 (Berner and Berner, 1987) and that the average retention time for trace metals is 4 months. The total concentrations of the trace metals in lakes and rivers are estimated as the sum of the baseline and pollution burdens. For most of the trace elements, the calculated baseline concentrations are similar to the levels found in the open ocean (see Wong et al., 1983). The very low baseline concentrations imply that small changes in pollutant metal inputs can significantly perturb the distribution of trace metals in fresh water ecosystems. The calculations (Table IV) suggest that the anthropogenic contributions now constitute about 50% or more the total trace metal burdens in lakes in rivers. The determination of the pollution component in lakes and rivers has been hampered by the fact that the available data bases are severely compromised by contamination artifacts. I am encouraged by the fact that the predicted concentrations are in general agreement with some of the most recent concentrations that have been obtained using the ultra-clean laboratory technique.
The concentrations of trace metals in rainfall in the urban and rural areas (Table II) are generally higher than the levels in lakes (Table IV). Apparently, lakes have developed effective self-cleaning processes that involve such factors as pH, water renewal time, complexation capacity, p a r t i c u l a t e concentration, biological processes (such metal methylation and assimilation), bioturbation of sediments, etc. Our studies have clearly demonstrated that the cycling of trace metals in lakes is strongly dominated by particulate dynamics in the water column (Wong and Nriagu, 1984; Nriagu and Wong, 1986). I believe t h a t the trace metal content of the seston (suspended particulates) may be a better indicator of the level of metal pollution than the concentration of the trace metals in the water. Several studies have shown that aquatic ecosystems respond rather quickly to changes in anthropogenic inputs of trace metals. For example, the surface ocean waters off the northeastern United States have shown a 30-40% decline in lead concentration due to the phase-out of leaded gasoline. The isotopic composition of Pb in the surface waters of the Great Lakes is very similar to the isotopic signature of the ambient aerosol (Flegal et al., 1989); the close correlation stems from the fact that the life-time of lead in the epilimnion is only a few days and the Pb concentration is controlled by recent atmospheric inputs. Also, there has been a drastic shift in the isotopic composition of Pb in Lake Erie sediments which reflects the declining role of atmospherically derived automotive lead (Mueller, 1988). The installation of a tall stack at the C u / N i smelters in Sudbury (Ontario) has reduced the rates of accumulation of trace metals in the local lake sediments (Nriagu and Rao, 1987), while the reduced consumption of leaded gasoline has engendered the recent decline in Pb accumulation in aquatic sediments which has been documented in many parts of the world (Levitt, 1988). Influence on m a r i n e trace m e t a l cycle
The current massive industrial emissions of trace metals have greatly increased the atmo-
HLrMAN I N F L U E N C E ON T H E GIX)BAL CYCLING OF T R A C E METALS
spheric fluxes of trace metals into the ocean (see above for the regional distribution of Pb in m a r i n e atmosphere). F u r t h e r m o r e , model calculations (above) suggest t h a t t h e river-borne delivery of trace metals to the oceans has p r o b a b l y doubled in recent times. To w h a t extent h a v e the n a t u r a l geochemical cycles been p e r t u r b e d b y the anthropogenic inputs? In addition to the mass balance considerations, there are several lines of evidence which show t h a t the anthropogenic inputs h a v e completely overwhelmed the n a t u r a l m a r i n e cycle of lead: (a) there is a 15-fold increase in Pb content of recent coral growth layers c o m p a r e d to those of a century ago; (b) the m a r k e d change in the f o r m of Pb profiles in ocean w a t e r column reflects the strong a t m o s p h e r i c source intensity; (c) a 26-fold difference in the Pb concentrations in mixed surface layers has been noted between the N o r t h Atlantic (polluted) and the S o u t h Pacific (relatively unpolluted); and (d) a measurable shift in the isotopic signature of Pb in surface waters towards the isotopic ratio of industrial lead has also been reported (see P a t t e r son and Settle, 1986, for detailed discussion of oceanic lead pollution). A recent analysis of the Atlantic Ocean sediments show a s h a r p increase in the Pb content of the surficial layer which has been a t t r i b u t e d to increased influx of anthropogenic lead (Veron et al., 1987). T h e h u m a n influence on the oceanic cycling of the other trace elements has not been fully assessed a l t h o u g h industrial activities are also transferring large quantities of these trace elem e n t s into the marine environment.
Contamination o/soils Soils are the u l t i m a t e sink for p o l l u t a n t metals in the terrestrial e n v i r o n m e n t (Table III). Assuming t h a t the discharges from industrial sources are dispersed uniformly over the land area (150 × 1012 m2), the annual rates of m e t a l addition to soils are estimated to v a r y from 0.1 g h a - 1 y r - 1 for Cd, to a b o u t 5 g h a - 1 y r - 1 for Pb, Cu and Cr, to over 6 g h a - 1 y r - i for Zn and Mn. T h e large background reservoir of trace metals in soils often obscures such high industrial load-
119
ings and increased concentrations are rarely observed except in u r b a n soils and near m a j o r point sources of pollution. Several studies h a v e d o c u m e n t e d an increase in the Cd content of soils following continued, long-term applications of p h o s p h a t e fertilizers to agricultural soils (see Adriano, 1986). Since the 1850s, an increase of 27-55% in the Cd burden of the soil plough layer (0-23 cm) has been reported a t the R o t h a m s t e d E x p e r i m e n t a l Station, a semi-rural location in s o u t h e a s t England (Jones and Johnston, 1989). Such a change in solid Cd burden of 1.9-5.4 g ha 1 y r - 1 has been a t t r i b u t e d solely to increased atmospheric deposition of this element (Jones and Johnston, 1989). T h e r e is some evidence t h a t large areas of J a p a n and central E u r o p e have become severely c o n t a m i n a t e d with c a d m i u m (Asami, 1984; Kloke et al., 1984). Increased concentration of Cd in soils is of h u m a n health concern because (a) there a p p e a r s to be a positive relationship between the Cd levels in soils and plants, and (b) crops represent the m a j o r source of dietary Cd for the general population.
References Aaby, B., Jacobsen, J. and Jacobsen, O.S. Dan. Geol. Unders. Arbog, 1978: 5-68. Adriano, D.C., 1986. Trace Elements in the Terrestrial Environment. Springer, New York, N.Y. Agricola, G., 1555. De re Metallica (Translated 1950, by H.C. Hoover and L.H. Hoover, Dover Publications, New York). Asami, T., 1984. In: J.O. Nriagu (Editor), Changing Metal Cycles and Human Health. Springer, Berlin, pp. 95-11. Berner, E.K. and Berner, R.A., 1987. The Global Water Cycle. Prentice-Hall, Englewood Cliffs, N.J. Boutron, C.F. and Patterson, C.C., 1987, J. Geophys. Res., 92: 8454-8464. Davis, R.B., Norton, S.A., Hess, C.T. and Brakke, D.F., 1983. Hydrobiologia, 103: 113-123. Del Mar, A., 1880. History of Precious Metals. Bell, London. Flegal, A.R., Nriagu, J.O., Coale, K.H. and Niemeyer, S., 1989. Nature, 339: 455-458. Jones, K.C. and Johnston, A.E., 1989. Environ. Pollut. 57: 199-216. Kloke, A., Sauerbeck, D.R. and Vetter, H., 1984. In: J.O. Nriagu (Editor), Changing Metal Cycles and Human Health. Springer, Berlin, pp. 113-141. Jaworowski, Z., Bilkiewicz, J., Dobosz, E. and Wodkiewicz, L., 1975. Environ. Pollut., 9: 305-315. Lee, J.A. and Tallis, J.H., 1973. Nature, 245: 216-218. Lenihan, J., 1988. The Crumbs of Creation. Hilger, Bristol.
120 Livett, E., 1988. Advanc. Ecol. Res., 18: 65-176. MARC, 1985. Historical Monitoring. Monit. Res. Assess. Cent., Chelsea College, Univ. London. Martin, M.H., Coughtrey, P.J. and Ward, P., 1979. Proc. Bristol Nat. Soc., 37: 91-97. Mueller, E.L., The Variation in Lead Isotope Ratios in Lake Erie Sediments. Thesis. Dept. Physics, Univ. Toronto. Murozumi, M., Chow, T.J. and Patterson, C.C., 1969. Geochim. Cosmochim. Acta, 33: 1247-1294. Nriagu, J.O., 1979. Nature, 279: 409-411. Nriagu, J.O., 1988. Environ. Pollut., 50: 139-161. Nriagu, J.O., 1989. Nature, 338: 47-49. Nriagu, J.O. and Davidson, C.I. (Editors), 1986. Toxic Metals in the Atmosphere. Wiley, New York, N.Y. Nriagu, J.O. and Wong, H.K.T. 1986. Water, Air and Soil Pollut., 31: 999-1006.
J.O. NRIAGU Nriagu, J.O. and Pacyna, J.M., 1988. Nature, 333: 134-139. Nriagu, J.O. and Rao, S.S., 1987. Environ. Pollut., 44: 211218. Patterson, C.C. and Settle, D.M., 1987. SEAREX Newsl., 10 (1). Veron, A., Lambert, C.E., lsley, A., Linet, P. and Grousset, F., 1987. Nature, 326: 278-281. Verta, M., Tolonen, K. and Simola, H., 1989. Sci. Total Environ., 87: 1-18. Volkening, J., Baumann, H. and Heumann, K., 1988. Atmos. Environ., 22: 1169-1174. Wong, H.K.T. and Nriagu, J.O., 1984. Trace Substances in Environ. Health, 18: 416-426. Wong, C.S., Boyle, E.A., Bruland, K.W., Burton, D. and Goldberg, E.D. (Editors), 1983. Trace Metals in Sea Water. Plenum, New York, N.Y.