This novel approach provides clues to sources of pesticides found in the atmosphere.

TERRY F. BIDLEMAN AND RENEE L. FALCONER

     Last July, a negotiating committee of the United Nations Environmental Programme (UNEP) met for the first time to seek worldwide elimination of 12 persistent organic pollutants (POPs), 9 of which are OC pesticides: DDT, chlordane, heptachlor, aldrin, dieldrin, endrin, toxaphene, mirex, and hexachlorobenzene (HCB)--the latter is both a pesticide and an industrial byproduct. Criteria are under development that will address additional substances in the future. The ban is contentious for DDT, which is still used in tropical countries for control of malaria. Worldwide sales of chlordane and heptachlor were halted only in 1997, and use of existing stocks can be anticipated for some time.

     For the decision makers to effectively eliminate OC pesticides and other POPs from the world's atmosphere, they must know the origins of the compounds. Presently, sources that contribute to observed global atmospheric distributions are not fully understood. At issue is a central question: Are OC pesticides transported from regions where they are currently applied, or are they instead ghosts of the past--recycled by evaporation of residues from previously contaminated soils and bodies of water?

     One novel approach for investigating pesticide recycling from soil and water involves using enantiomers as tracers. This is feasible because several classes of insecticides and herbicides have members that are chiral--compounds having right- and left-handed molecular configurations, or enantiomers (see figure on top of next page). Four of the eight OC pesticides on the UNEP list (o,p'-DDT, chlordane, heptachlor, and toxaphene) are chiral.

     Pesticide enantiomers are useful as tracers of soil-air and water-air exchange processes for the following reasons. Although a few chiral pesticides are manufactured as single-enantiomer products, most are racemic mixtures having a (1:1) enantiomer ratio (ER). Enantiomers have the same physical and chemical properties. As a result, transport processes (leaching, volatilization, and atmospheric deposition) and abiotic reactions (hydrolysis and photolysis) do not discriminate between the enantiomers--such processes leave ERs unaffected. In contrast, metabolism of pesticides by microorganisms in water and soil and by enzymes in higher organisms often proceeds enantioselectively, leading to nonracemic residues and an alteration of the original ER (2-6) ( see sidebar on next page). By examining ERs, it is possible to differentiate new sources of OC pesticides from ghosts of the past. The technique provides a sensitive indicator of biological degradation and clues about the origins of pesticides found in the atmosphere. Work performed by ourselves and other investigators demonstrates the potential usefulness of this approach.

New sources, old sources
OC pesticides are still being applied or have been used in the recent past in many tropical and subtropical countries. This has led to especially high levels of some compounds in the ambient air of eastern and southern Asia (7). The conclusion that less developed countries are solely responsible for global contamination is, however, overly simplistic. Historically, large quantities of OC pesticides were also used in industrialized countries.

     A recent worldwide survey found that residues of HCB, dieldrin, and the technical chlordane component trans-nonachlor in tree bark, which integrates atmospheric exposure over a 3- to 5-year period, were positively correlated with a country's gross national product per person and Human Development Index (8). Tree bark sampled in the poorest countries generally showed the least amount of contamination, whereas the highest levels occurred in the United States and countries of northern Europe. No significant correlations were found between socioeconomic indicators and residues of hexachlorocyclohexanes (HCHs) or DDT.

     One explanation for the pesticide distribution picture provided by the tree bark survey is that emissions from previously contaminated soil and water may be entering the atmosphere in significant amounts. Soil, especially agricultural soil, is likely the largest reservoir of OC pesticides and a major source of emissions. For example, toxaphene was heavily used in the southern United States on cotton and soybeans before being deregistered in 1982. Toxaphene concentrations measured in South Carolina air during the mid-1990s were 10 times higher than levels in the Great Lakes region and showed no relationship to air transport direction, suggesting volatilization from regional soils (9). Moreover, DDT residues in air above soil at a California farm, where DDT had been applied 23 years previously (10), were 2 to 3 orders of magnitude higher than in the Great Lakes region (Environ. Sci. Technol. 1998, 32, 1920-1927).

     Long-term monitoring on the shores of the Great Lakes reveals that airborne pesticides are declining only slowly, with "virtual elimination" dates (when levels fall below detectability) ranging from 2010 to 2060 (Environ. Sci. Technol. 1998, 32, 1920-1927). Concentrations of the OC pesticides and other POPs in air and surface water are now nearly in steady state with respect to gas exchange (Environ. Sci. Technol. 1998, 32, 2216-2221). Evidently, certain processes are acting to stabilize atmospheric concentrations. This finding has significant consequences for cleanup of the lakes, because improvement of water quality is pegged to long-term atmospheric trends (11). Understanding how POPs migrate through the global environment is therefore critical to developing control strategies, and as a result, tracing the cycling processes of POPs among air, soil, water, and vegetation has become an area of growing importance (Environ. Sci. Technol. 1996, 30, 390A-396A). Methods that can investigate emissions from soil and water as well as discriminate between "new" and "old" sources of contamination contribute to our understanding of these issues.

     Techniques for directly measuring the flux of pesticides from soil are well established (Envion. Sci. Technol. 1993, 27, 121-128)(11), but they have been largely applied to determining postapplication volatilization rates rather than emission of in-place residues. Models can predict pesticide volatilization based on soil properties, residue levels, meteorological conditions, and physicochemical properties of chemicals; modeled and measured pesticide fluxes have agreed well in field trials (12). However, application of models to predicting emissions of formerly used pesticides is limited by the paucity of residue data for soils. Unlike programs for air and biota, there is not an organized effort to monitor soils. Another problem is that residues in agricultural soils are highly variable (3), which creates difficulties in assessing the reservoir of pesticides in soil and selecting representative concentrations for modeling emissions.

Identifying sources
Ratios of parent compounds to metabolites and proportions of components in technical mixtures have been used to infer sources. Total DDT residues in the upper slices of peat cores from the Great Lakes and eastern Canada regions contained a high proportion of parent DDT relative to the metabolite DDE, leading to the hypothesis that "new" DDT continued to be atmospherically transported from Mexico, Central America, and Caribbean countries for years after the 1972 U.S. ban (13). Atmospheric measurements taken in the mid-1990s at Integrated Atmospheric Deposition Network stations on the Great Lakes show DDT/DDE < 1.0 (Environ. Sci. Technol. 1998, 32, 1920-1927), suggesting that a substantial portion of today's DDT comes from recycled old residues. However, a problem with DDT/DDE ratios is that they are quite variable in soils. DDT/DDE was 0.5-2.0 and occasionally 5 in soils of the midwestern United States (3). Furthermore, DDE is about 6 times more volatile than DDT, making it difficult to interpret ratios in air relative to those in soil.

     Similar challenges confound identifying sources of HCH, which is one of the most widely used insecticides in the world (2, 14). The technical mixture contains 60-70% -HCH, 5-12% -HCH, 10-12% -HCH, and other isomers (2). Only -HCH is insecticidal, but the other isomers have toxic properties, particularly -HCH, which is more bioaccumulative and is a possible environmental estrogen (2). Canada, the United States, and European countries have banned technical HCH in favor of using pure -HCH (lindane), but large quantities of technical HCH were used in Asia throughout the 1980s and to a lesser extent into the 1990s (14). The atmospheric signal today consists of lindane superimposed on a background of technical HCH, and elevated ratios of -HCH/-HCH indicate episodic transport of lindane from regions of current use (2). A difficulty with interpreting this ratio is that the two isomers are removed from the atmosphere at different rates during transport, possibly due to differences in air-sea exchange or photolysis rates (2).

     Technical chlordane is another mixture containing trans-chlordane (TC), cis-chlordane (CC), trans-nonachlor (TN), and heptachlor as major components. In the United States, chlordane and heptachlor were used in agriculture until the mid-1970s and as termiticides until 1988 when registrations were cancelled. The proportions of TC:CC:TN in the ambient air of Columbia, S.C., where chlordane was used for termite control, were quite close to those in the technical product after accounting for differences in volatility (9). Residues in agricultural soil vary in TC:CC:TN (3) due to differences in rates of metabolism and physical dissipation. Selective removal of TC takes place during atmospheric transport, possibly by photochemical reactions, so that ratios of TC/CC are generally <1.0 in remote regions. In the Arctic, TC/CC undergoes seasonal variations from 0.6-1.0 in winter to 0.2-0.4 in summer (1), suggesting that the ratio of the two isomers might be an indicator of transport time in the atmosphere.

Enantiomers as tracers
Volatilization of Pesticides From Soil: The enantiomer profiles of nonracemic pesticide residues in soil and water are preserved upon volatilization, giving an "old source" signature that can be distinguished from freshly applied (racemic) pesticide. Once in the atmosphere, pesticides are subject to only nonbiological removal mechanisms, such as photolysis, hydroxyl radical attack, and atmospheric deposition--achiral processes that will not change the proportion of enantiomers. Moreover, since enantiomers have the same volatility and air-water partitioning properties, the ERs are not subject to fractionation effects as are isomer and parent/metabolite ratios.

     Enantiomers used as tracers provide clues to the origin of pesticide residues in ambient air. When air samples were collected above agricultural soils containing nonracemic residues of heptachlor, heptachlor epoxide, and -HCH, a good correspondence was found for the enantiomer profiles in soil and air (6). Heptachlor epoxide is produced by both photolysis of heptachlor and metabolism of heptachlor in soils and organisms. The heptachlor epoxide in ambient air from the Great Lakes region, South Carolina, and Alabama was distinctly nonracemic, with ERs ranging from 1.5 to 2.1 (6) (Environ. Sci. Technol. 1998, 32, 1546-1548). These are similar to soil values (3), indicating that heptachlor epoxide in air probably emanates from enantioselective epoxidation of heptachlor in soils followed by volatilization, rather than from photolysis of heptachlor (Environ. Sci. Technol. 1998, 32, 1546-1548). Sources of chlordane include volatilization of soil residues and emissions from termiticide usage. Chlordane in ambient air was racemic in Columbia, S.C., where the pesticide was used for termite control (6). Air sampled near the Great Lakes (4, 6) showed ERs that fell between the racemic termiticide signature and the nonracemic values of Cornbelt soil (3) (see figure on previous page), suggesting that both source types contribute chlordane to the Great Lakes.

     Exchange With Lakes and Oceans: A recent atmospheric budget for the Great Lakes shows that fluxes of several OC pesticides and PCBs are dominated by air-water gas exchange (Environ. Sci. Technol. 1998, 32, 2216-2221). Exchange of HCH isomers undergoes seasonal cycles in the Great Lakes in response to changes in water temperature, air concentrations, and stratification of the water column. The (+) enantiomer of -HCH is depleted in lake water resulting in a seasonally invariant ER = 0.85, while the ER of -HCH in air is controlled by volatilization (nonracemic) and long-range transport (racemic) contributions. When measurements were made over a year's time, a seasonal cycle could be seen in which ERs were nearly racemic (0.98-1.02) in winter, spring, and fall months and dropped to 0.91-0.94 in the summer, when volatilization had more effect on the enantiomer signature in air (6) (Environ. Sci. Technol. 1997, 31, 1940-1945) (see figure at right).

     The Arctic Ocean has been the recipient of atmospheric HCH loadings for a long time, and concentrations in surface water are higher than in temperate oceans and lakes due to the "cold condensation effect." (Environ. Sci. Technol. 1996, 30, 390A-396A). A 10-fold drop in atmospheric HCHs since 1980, caused by restrictions or bans of technical HCH usage in Asian countries (14), has led to reversal in the net gas exchange direction of -HCH in northern seas, from deposition in the 1980s to volatilization in the 1990s (15). The -HCH in surface water is nonracemic, being depleted in the (+) enantiomer in the western Arctic Ocean and the (-) enantiomer in the Bering and Chukchi Seas. The -HCH in air over ice-free regions showed the same enantiomer depletions as the surface water (6, 15).

     Such measurements provide direct evidence of the "two-way street" nature of gas exchange. They also call into question whether reliable estimates of air-water gas exchange for POPs can be made from air concentration data obtained at shoreline stations.

Next steps
Other potential applications of this approach include differentiating biological versus chemical breakdown in soils and sediments, investigating transport and degradation in surface and groundwater, and following selective accumulation and metabolism in food chain studies. The application of enantiomers in these areas would be enhanced by improving analytical methods for enantiomer separations; determining rates and mechanisms of enantioselective metabolism--in particular, microbial processes should receive attention, since their key role has been established (2); examining the role of field parameters on enantiomeric composition of residues (e.g., microbial populations, pH, temperature and concentrations of the chemical, organic carbon, and nutrients); defining spatial and temporal variability of pesticide concentrations and ERs in soil and water to allow data to be applied on a regional basis; and extending the use of enantiomer tracers to other chiral compounds such as atropisomeric polychlorinated biphenyls and currently used pesticides.

References

(1) Halsall, C. J.; Stern, G. A.; Barrie, L. A.; Fellin, P.; Muir, D. C. G.; Billeck, B. N.; Rovinsky, F. Ya.; Kononov, E. Ya.; Pastukhov, B. Environ. Pollut. 1998, 102, 51-62.

(2)  Willett, K. L.; Ulrich, E. M.; Hites, R. A. Environ. Sci. Technol. 1998, 32, 2197-2207.

(3)  Aigner, E. J.; Leone, A. D.; Falconer, R. L. Environ. Sci. Technol. 1998, 32, 1162-1168.

(4)  Ulrich, E.; Hites, R. A. Environ. Sci. Technol., 1998, 32, 1870-1874.

(5)  Wiberg, K.; Oehme, M.; Haglund, P.; Karlsson, H.; Olsson, M.; Rappe, C. Mar. Pollut. Bull. 1998, 36, 345-353.

(6)  Bidleman, T. F., et al. Environ. Pollut. 1998, 102, 43-49.

(7)  Iwata, H.; Tanabe, S.; Sakai, N.; Nishimura, A.; Tatsukawa, R. Environ. Pollut. 1994, 85, 15-33.

(8)  Simonich, S.; Hites, R. A. Environ. Sci. Technol., 1997, 31, 999-1003.

(9)  Bidleman, T. F.; Alegria, H.; Ngabe, B.; Green, C. Atmos. Environ. 1998, 32, 1849-1856.

(10)  Spencer, W. F.; Singh, G.; Taylor, C. D.; LeMert, R. A.; Cliath, M. M.; Farmer, W. J. J. Environ. Qual. 1996, 25, 815-821.

(11)  Makay, D.; Bentzen, E. Atmos. Environ. 1997, 31, 4045-4047.

(12)  Scholtz, M. T.; McMillan, A. C.; Slama, C.; Li, Y-F.; Ting, N.; Davidson, K. Pesticide Emissions Modelling. Development of a North American Pesticides Emission Inventory; Canadian Global Emissions Interpretation Centre, ORTECH Corporation: Mississauga, Canada, 1997.

(13)  Rapaport, R. A.; Urban, N. R.; Capel, P. D.; Baker, J. D.; Looney, B. B.; Eisenreich, S. J.; Gorham, E. Chemosphere 1985, 14, 1167-1173.

(14)  Li, Y-F.; Bidleman, T. F.; Barrie, L. A.; McConnell, L. L. Geophys. Res. Lett. 1998, 25, 39-41.

(15)  Jantunen, L. M.; Bidleman, T. F. J. Geophys. Res. 1996, 101, 28,837-28,846; corrections Ibid. 1997, 102, 19,279-19,282.

Terry F. Bidleman is a research scientist at the Atmospheric Environment Service in Downsview, Ontario, Canada. Renee L. Falconer is an associate professor in the Chemistry Department of Youngstown State University in Ohio.