Assessing the Relationship between Extensive ... - ACS Publications

Oct 9, 2007 - Assessing the Relationship between Extensive Use of Organochlorine Pesticides and Cooling Trend during the Mid-20th Century in the ...
1 downloads 0 Views 479KB Size
Research Assessing the Relationship between Extensive Use of Organochlorine Pesticides and Cooling Trend during the Mid-20th Century in the Southeastern United States J I A N M I N M A , * ,† Y I - F A N L I , † TOM HARNER,† AND ZUOHAO CAO‡ Air Quality Research Division, Science and Technology Branch, Environment Canada, 4905 Dufferin Street, Toronto, Ontario M3H 5T4, Canada, and Meteorological Service of Canada, Environment Canada, 4905 Dufferin Street, Toronto, Ontario M3H 5T4, Canada

In contrast to global warming, temperatures in the southeastern United States have exhibited a decreasing trend of up to 1-2 °C over the last century. We investigated the relationship between this cooling trend and the extensive use of organochlorine pesticidessparticularly dichloro-diphenyl-trichloroethane (DDT) and toxaphenes in the southeastern U.S. from the mid-1940s through the 1970s. Based on annual use and soil residue estimates, it is shown that enthalpies associated with the endothermic evaporation of pesticides from soil resulted in surface temperature decreases of up to -0.2 °C. This is the first study to show the inverse correlation between surface air temperature and pesticide use. These findings highlight the potential contribution of the extensive use of organochlorine pesticides to changes in the surface air temperature, especially in regions of intensive agriculture and pesticide use.

Introduction Observed climate change over the past century has been attributed to natural climate processes and human activities. The earth’s climate varies naturally for many reasons, including changes in the intensity of solar radiation, largescale perturbations in ocean thermal and flow structures, and eruption of volcanoes, to name only a few. There is an increasing body of evidence indicating that the 20th century climate change and warming trend is not due solely to natural climate forcing (1). Human activities are also influencing climate. The major anthropogenic impacts on climate include the increase in carbon dioxide levels associated with fossil fuel combustion (positive forcing); followed by increased atmospheric aerosols which exert a cooling effect (negative forcing); and land use changes (which can be positive or negative forcing). The observed global warming over the past 100 years has not been monotonic (i.e., consistent in one direction) and * Corresponding author tel: 1-416-739-4857; fax: 416-739-4288; e-mail: [email protected]. † Air Quality Research Division, Science and Technology Branch, Environment Canada. ‡ Meteorological Service of Canada, Environment Canada. 10.1021/es0617744 CCC: $37.00 Published on Web 10/09/2007

Published 2007 by the Am. Chem. Soc.

has not occurred uniformly across the globe. For instance, there was a period of overall cooling during the mid-20th century which contradicted the expected warming due to increased “greenhouse” gases emissions during the postWorld War 2 industrialization. The cooling trend was especially strong in the southeastern United States during the period 1950-1970s and still persists todayspossibly a remnant of the early pronounced period of cooling (2-5). This anomalous period of cooling has been studied extensively and attributed to natural climate variability and negative forcing associated with anthropogenic sulfate aerosols (2-6). However, a recent general circulation model (GCM) simulation that accounted for potential sulfate aerosol forcing did not demonstrate the cooling trend in the southeastern U.S. (7). A question may be raised: what other human activities may have contributed to the observed cooling? In this study we investigate the possibility that heavy use of organochlorine pesticides in the southern United States during the mid-20th century may have contributed to the observed cooling trend. It is postulated that the long-term endothermic (absorbing heat) evaporation of these heavily used chemicals (8, 9) caused a measurable cooling effect. This analysis is conducted using usage and soil residue trends for two organochlorine pesticides (OCPs) that dominated the market during this time: DDT (dichloro-diphenyltrichloroethane) and toxaphene.

DDT and Toxaphene Use and Soil Residue Trends The southeastern United States was the largest user of DDT and toxaphene in the world (8, 9), with peak consumption in 1966 and 1974, respectively. From 1945 until its ban in 1972, 6.13 × 105 t of DDT (about 40% of the global total) was used in the United States (mostly in the southeastern U.S.) (8, 10). During this time, it was typical for more than 1000 kg of DDT to be applied to a single 100-hectare cotton field over a 4-week period (11). During the 1960s and 1970s some uses of DDT were replaced with toxaphene which became the most heavily used pesticide in the United States until its restriction in 1982 (12). Toxaphene use in the United States is estimated at 4.9 × 105 t for the period 1947 to 1986 (approximately 40% of the global total, (9)). More than 85% of the toxaphene used in the U.S. was for cotton-growing, primarily in the southeast (12). Usage and soil residue inventories for DDT and toxaphene have been produced on a 1° × 1° spatial resolution (9, 13, 14). These estimates have been validated using measured air and soil concentrations (13, 15-17) and numerical simulations (18). Figure 1 shows the time series for annual use and soil residues for DDT and toxaphene in the U.S. for the period 1947-2000. Because of their stability (persistence) and lowvolatility, DDT and toxaphene accumulated in these soils with gradual revolatilization (emission) to the atmosphere over the next several decades, even after their use had ceased (15). Figure 1 also shows the smoothed annual U.S. air temperature anomalies for the period 1947-1998, compiled by the National Climate Data Center [NCDC, (19)] of the U.S. National Oceanic and Atmospheric Administration (NOAA). As seen from Figure 1, the temperature anomalies are inversely correlated to DDT and toxaphene soil residues. In fact the timing of the minimum in the temperature anomaly and the maximum in soil residues agree very well. This inverse correlation was further explored by relating the linear trends of toxaphene and DDT use and the GISS VOL. 41, NO. 21, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7209

FIGURE 1. Annual use and soil residues of DDT and toxaphene (kt) for the period 1947-2000 and smoothed U.S. annual temperature anomalies (°C) for the period 1947-1998. Temperature anomaly time series data was obtained from NCDC/NOAA at http://www.ncdc.noaa.gov/ oa/climate/online/doe/usa1×1.temp.

FIGURE 2. (a) Trends of annual toxaphene use (t yr-1) in the U.S. for the period 1947-1976. (b) GISS derived surface air temperature anomalies from March to August (oC) for the same period. surface temperature anomalies [Goddard Institute for Space Studies Surface Temperature Analysis (20)] for the period 1947-1976 [the year when the use of toxaphene in the southeastern U.S. started to decline (13)]. The time trends were analyzed using a simple linear regression

Y ) A + B(T) where T is time in years and the slope B indicates the average rate of change in the use of toxaphene (or DDT). 7210

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 21, 2007

Figure 2a shows the 30-year linear trend of toxaphene use in the U.S., highlighting its extensive use in six southeastern statessnamely, Arkansas, Louisiana, Mississippi, Alabama, Georgia, and South Carolina (14). The gridded maximum linear trend occurred in Alabama with an annual average of 44 t yr-1. For DDT, there is no significant trend over this same period (1947-1976) because of a sharp decrease in use after 1959. However, over the first 15 years (1945-1959) a maximum annually averaged trend of about

FIGURE 3. Linear correlation coefficients between toxaphene annual soil residues and GISS surface temperature anomalies for the period 1948 to 2000. The areas encircled by white contours indicate the regions where the correlation is significant at the 99% confidence level. 30 t yr-1 is observed for Alabama (not shown). Figure 2b illustrates the trends of the temperature anomalies from March to August. The spring/summer period was chosen since it represents the time of maximum application and emission from soil. The strong cooling trend in the southeastern U.S. is clearly seen and agrees well with the trends of increasing use of toxaphene (Figure 2a) and DDT (not shown).

TABLE 1. Linear Correlation Coefficients (r) and p-Values (p) (at 95% Confidence Level) Between the Soil Residues of Toxaphene and DDT and ht, the “Time Bias Corrected” Average Temperature in the Six States of the Southeastern U.S.; and the 30 Years Linear Trend of ht and Summed Trend of ∆T, the Temperature Drop Due to Volatilization of DDT and Toxaphene for 1947-1976 in the Six States of the Southeastern U.S. correlation between residue and ht

Linking Observed Temperature Declines to Use of Toxaphene and DDT Relationships between DDT and toxaphene soil residues and temperature are assessed further by linear correlation analysis using annual soil residues and GISS temperature anomalies from 1948 to 2000, shown in Figure 3. The spatial pattern of the correlation coefficients indicates a statistically significant relationship (g99% confidence, encircled by white contours) between toxaphene soil residues and the temperature anomalies in the southeastern U.S. These results show a pattern similar to the linear trend of toxaphene use and the temperature anomalies shown in Figure 2. Correlations between DDT soil residues and temperature anomalies (not shown) exhibit a similar spatial pattern. We have also estimated the linear correlation coefficients between the measured annual mean temperatures in each state of the U.S., which were derived from monthly “time bias corrected” average temperature (th) compiled by Karl et al. [(21), Climate Diagnostics Center (CDC) of NOAA], and the soil residues of toxaphene and DDT (averaged over each of the aforementioned 6 southeastern U.S. states) for the period 1947-1976. Results show good correlations (g99% confidence) with the exception of the state of Arkansas (Table 1). The reason for the poor correlation for Arkansas is not clear. Nonetheless, a link appears to exist between the mid20th century cooling and the extensive use of toxaphene and DDT in this region. Currently, because there is insufficient knowledge to assess the radiative forcing associated with OCPs due to their presence in the atmosphere, it is difficult to examine the projection and sensitivity of OCPs in a climate model as a radiative forcing. Alternatively, in the present study we postulate that the association between the use of OCPs and the cooling trend is not a coincidence and that the energy/heat drain associated with evaporation/volatilization of soil-bound toxaphene and DDT contributed to the lowering of surface temperatures in the southeastern U.S.

Endothermic Evaporation of Toxaphene and DDT The evaporation/volatilization of toxaphene and DDT (and any other semivolatile pesticide/chemical) from a surface is

toxaphene vs ht

Alabama Arkansas Mississippi Louisiana South Carolina Georgia

DDT vs ht

trend

r

p

r

p

ht

Σ∆T

-0.62 -0.36 -0.54 -0.59 -0.54 -0.57

0.0003 0.0572 0.0024 0.0007 0.0025 0.0011

-0.64 -0.24 -0.54 -0.57 -0.56 -0.65

0.0001 0.2067 0.0023 0.0016 0.0023 0.0001

-1.25 -0.53 -0.98 -0.94 -0.94 -1.11

-0.058 -0.023 -0.039 -0.045 -0.050 -0.035

an endothermic process. In other words, energy is required to allow the molecules to break free from the condensed state (adsorbed or absorbed onto/into surfaces as a liquid, solid, or supercooled liquid) and become gaseous. This energy is absorbed from the surrounding environment, resulting in a temperature drop. When these pesticides are applied, they become retained in soils, on the plant canopy (e.g., cotton) (22), and also on atmospheric aerosols. However, the heat loss associated with plant- and aerosolbound pesticides, even immediately after application when air concentrations are elevated (8, 9, 12, 13), is expected to have little effect on the annual average temperature and longterm changes in regional climate. This is because the chemical is dissipated relatively quickly by advection and degradation. Half-lives of DDT and toxaphene in the air and on canopies are about 7-9 days (23) and 15-19 days (24). A much larger and persistent quantity of pesticide is associated with the soil. Soil residues of DDT and toxaphene have increased an order of magnitude since their use began in the 1940s despite the fairly steady application rate (Figure 1). Thus evaporation from soil (emission) is a more continuous, longer-term process that may better explain the observed temperature anomalies. Toxaphene and DDT are semivolatile chemicals that undergo soil-air exchange. Soil-to-air transfer of sorbed chemical involves phase change from the supercooled liquid state (the state assumed by chemicals that are “dissolved” in soil organic matter) to the gas phase. This is analogous to energy requirements associated with octanol-air partitioning. Octanol is often used a model compound for representing VOL. 41, NO. 21, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7211

environmental organic phases such as soil, aerosols, and vegetation. This partitioning constant (KOA) and its temperature dependence (enthalpy, ∆HOA, kJ mol-1) are well-known for OCPs (25) and generally range in value between 60 and 100. To quantitatively assess the temperature effect associated with DDT and toxaphene volatilization, we assume that the chemicals are at thermal equilibrium with air in the near surface regime. Thus, the increase in the enthalpy associated with their volatilization (∆Hsoil-air) results in an equal loss of heat from the surface air at a constant pressure. This gives

Mpest∆Hsoil-air t -(mair CPair ∆T)

(1)

where Mpest is the mass of a pesticide and ∆T is the temperature drop due to volatilization; CPair is the heat capacity of dry air at constant pressure ()1.006 kJ kg-1 K-1); and mair is the mass of air associated with the surface compartment. In eq 1, pesticide masses are converted to moles using the molecular weights of DDT and toxaphene, 354.5 and 414 g mol-1, respectively. To estimate the change in annual surface temperature due to DDT and toxaphene volatilization in eq 1, we first calculate the energy required for volatilization by multiplying the mass (Mpest, moles) of chemical by its enthalpy of soilair transfer (∆Hsoil-air) which we know is similar to its octanolair partitioning enthalpies (i.e., ∆HOA), 88 kJ mol-1 for DDT and 86 kJ mol-1 for toxaphene (25). These enthalpies are only weakly dependent on air temperature and are therefore taken as constants. The amount of soil involved in soil-air transfer (emission) from the control volume of 1 m2 by 0.1 m deep is calculated from annual soil residue data, using an annual decay (loss) rate of 9% of total soil residues due to volatilization for DDT and 8% for toxaphene (i.e., 8-9% of soil residues are volatilized to the air each year). These decay rates are based on the results of a coupled OCPs atmospheric transport, soil/ air, and water/air exchange modelsCanMETOP (Canadian Model for Environmental Transport of Organochlorine Pesticides) (18, 27, 28)sthat was applied to DDT and toxaphene soil residues across North America. Previous studies have identified a depth of 0.1 m as representative of the soil control volume that corresponds to high soil residues (9). This layer is also well-mixed as a result of mechanical working of the soil, e.g., tilling (29). Although the volatilization or soil/air exchange of toxaphene and DDT takes place mostly in the “exchange layer” (0-1 cm depth), deeper soil layers help to maintain or “buffer” the concentration in the top layer (28) via upward diffusion and mixing. These deeper soil layers are included in the overall soil residue calculations to better describe the long-term soil residue trends. Figure 4a and b show the CanMETOP modeled annual loss rate of DDT and toxaphene soil residue concentrations averaged over 0.1-10 cm soil depth, respectively. For DDT, the annual loss rates in the southern U.S. are about 16% (Figure 4a). Given the annual degradation rate of 6.7% based on an assumed half-life in soil of 10 years (15), the annual decay rate of DDT in the southeastern U.S. soils due to volatilization based on an assumed half-life in soil of 10 years (15) is about 9%. For toxaphene, the annual loss rates in the southern U.S. are about 15% (8% due to volatilization and 6.7% due to degradation, Figure 4b). This calculation also assumes a soil degradation half-life of 10 years. The mass of air near the surface below 0.1 m height subject to the roughness length of 0.1 m over a cotton field was estimated over each cell at the 1° × 1° resolution subject to the air density 1.2 kg m-3. As shown later, the height of 0.1 m used in eq 1 can be extended to the entire surface boundary layer. Using the 1° × 1° gridded annual soil residue inventories for DDT and toxaphene over the period 1947 to 1976 and 7212

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 21, 2007

FIGURE 4. (a) Annual loss rate (%) of DDT soil residue concentrations averaged over the 0.1 - 10 cm soil depth layer, simulated by CanMETOP for the year 2000; (b) same as (a) but for toxaphene. (Note: daily meteorological data (windspeed, temperature, and precipitation) and DDT and toxaphene soil residue data in the year 2000 were used for the model simulations.) knowing that ∆Hsoil-air is similar to ∆HOA (25) we compute the linear trend of ∆T due to volatilization using eq 1. Figure 5 shows the combined effect for DDT and toxaphene. The results are statistically significant at the 0.01 level. The strongest temperature decrease is calculated for Alabama, the state with the greatest use of DDT and toxaphene during this period (13). These findings are confirmed by the measured temperature data (Table 1) for the 6 states that exhibited the greatest cooling trend. The 30-year cooling trends in the observed temperature, collected from the “time bias corrected” average temperature (21) in the top 6 states, range from -0.53 (Arkansas) to -1.25 °C (Alabama). Whereas the estimated decreasing trends of the temperature drop from eq 1 averaged over each state is much smaller and ranges from -0.023 (Arkansas) to -0.058 °C (Alabama). Despite this small contribution from the volatilization of toxaphene and DDT on the overall cooling trend, the results should not be overlooked because, as shown in Table 1, the soil residues of these OCPs exhibit significant inverse correlation with the measured trend of the surface air temperature in the southeastern U.S. As discussed previously, it is likely that the strong cooling trend in the southeastern U.S. was induced largely by natural climate variability with only minor, negative contributions from the volatilization of pesticides.

FIGURE 5. Sum of the linear trend of temperature drop ∆T (°C) due to volatilization of DDT and toxaphene computed by eq 1. Overall, the results of our analysis indicate that the volatilization of toxaphene and DDT from southeastren US soils does not account for (quantitatively) the observed cooling trend in this region. However, this negative contribution is important when we consider that it is likely of similar magnitude as the one contributed to by anthropogenic forcing from sulfate or organic aerosols (4, 30, 31). Furthermore, the negative contribution associated with pesticide volatilization will increase when contributions from all applied pesticides are considered.

Other Considerations Water Evaporation. It is very likely that the heat loss (temperature change) associated with the evaporation of water from surface soil and water bodies is orders of magnitude greater than the heat loss associated with volatilization of OCPs from soil. However, water evaporation responds to climate change and is regarded as a climate response (30)sas opposed to volatilization of toxaphene and DDT which induces a cooling effect on surface air temperature. Pesticide Applied as Solvent Mix and Irrigation. In the present study we did not attempt to investigate the contribution from the evaporation of solvents (that may have been used to disperse the pesticides during application) and/or water (applied during irrigation of crops) on the changes in the surface air temperature. As aforementioned, in its peak application, more than 1000 kg of DDT could be sprayed on a single 100-hectare cotton field (11). Similarly, irrigation occurs in a short period during the course of a year. Such events are likely to lower the local air temperature for only a short period of time and hence do not cause changes in the annually averaged air temperatures and long-term variation in the regional climate. Whereas evaporation of OCPs from soil (emission) is a more continuous, longerterm process. Besides, the possible roles of solvents and irrigations are complex and beyond the scope of this paper. Air Compartment Height. Intuitively, one might expect that the temperature decrease calculated from eq 1 would become smaller as the control volume height is increased from, say 0.1 to 10 m. This is not the case (see Supporting Information). The 0.1 m height was selected for two reasons. First, the 20th century cooling and warming trends in the southeastern U.S. (and other parts of the world) occurred near the surface of the earth and were measured by the surface air temperature below 2 m height. Second, soil-air exchange of OCPs is controlled by molecular diffusion across an

interfacial sublayer immediately above the surface [readers are also referred to the well-know two-film model (32)]. The exact height of this layer is site specific and dependent on the roughness length over the particular field. For a typical cotton field, the roughness length based on field data is ∼0.1 m (26). Above this height, turbulent diffusion dominates the vertical exchange and mixing of OCPs. The extension of the temperature decrease calculation to the atmospheric turbulent boundary layer is presented in the Supporting Information. In this part of the paper, we have demonstrated that for a given temperature drop of -0.2 K at 0.1 m computed by eq 1, the temperature perturbation is -0.14 K at z ) 10 m and -0.11 K at z ) 100 m. In summary, this analysis, based on toxaphene and DDT that were heavily used in the southeastern U.S., has demonstrated the potential contribution of pesticide use and subsequent long-term emission (or volatilization), to the cooling trend. Although this contribution was shown to be an order of magnitude lower than what is believed to be attributed to natural climate variability, these findings still merit further research on this topic to consider other regions where persistent pesticides were applied and including a broader suite of heavily used and persistent pesticides. One example is eastern China where DDT and technical hexachlorocyclohexane (HCH) (8, 33) were extensively used from the 1950s to the 1980s. Coincidently, the surface air temperature in this part of China during the summer season exhibited a weak cooling trend over the period 1950-2000 (34) that is inversely correlated with the use of DDT and technical HCH. Further study is planned.

Acknowledgments CRUTEM2 data is provided by the NOAA-CIRES Climate Diagnostics Center, Boulder, Colorado, at http://www.cdc. noaa.gov/. Special thanks to S. Venkatesh of Environment Canada and G. Zhang of Scripps Institution of Oceanography for their useful comments.

Supporting Information Available Extension of the temperature drop computed from eq 1 from the air compartment height of 0.1 m to the turbulent atmospheric boundary-layer. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Scott, P. A. Attribution of regional-scale temperature changes to anthropogenic and natural causes. Geophys. Res. Lett. 2003, 30, doi: 10.1027/2003GL017324. VOL. 41, NO. 21, 2007 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

7213

(2) Kiehl, J. T.; Brieglieb, B. P. The relative role of sulfate aerosols and greenhouse gases in climate forcing. Science 1992, 260, 311-314. (3) Tett, S. F.; Stott, B. P. A.; Allen, M. R.; Ingram, W. J.; Mitchell, J. F. B. Causes of twentieth century temperature change near the earth’s surface. Nature 1999, 399, 569-572. (4) Intergovernmental Panel on Climate Change. Climate Change 2001: The Scientific Basis; Houghton, J. T., Ding, Y., Griggs, D. J., Noguer, M., van der Linden, P. J., Dai, X., Maskell, K., Johnson, C. A., Eds.; Cambridge University Press: Cambridge, UK, 2001; 881 pp. (5) Karl, T. R.; Jones, P. D.; Knight, R. W.; Kukla, G.; Plummer, N.; Razuvayev, V.; Gallo, K. P.; Lindseay, J.; Charlson, R. J.; Peterson, T. C. A new perspective on recent global warming: Asymmetric trends of daily maximum and minimum temperatures. Bull. Am. Meteorol. Soc. 1993, 74, 1007-1023. (6) Saxena, V. K.; Menon, S. Sulfate-induced cooling in the southeastern U. S.: an observational assessment. Geophys. Res. Lett. 1999, 26, 2489-2402. (7) Knutston, T. R.; Delworth, T. L.; Dixon, K. W.; Held, I. M.; Lu, J.; Ramawwamy, V.; Schwarzkopf, M. D. Assessment of twentiethcentury regional surface temperature trends using the GFDL CM2 coupled models. J. Clim. 2006, 19, 1624-1651. (8) Voldner, E. C.; Li, Y. F. Global usage of selected persistent organochlorines. Sci. Total Environ. 1995, 160/161, 201-210. (9) Li, Y. F. Toxaphene in the United States. 1. Usage gridding. J. Geophys. Res. 2001, 106, 17919-17927. (10) U.S. EPA. DDT regulatory history: a brief survey (to 1975); prepared for the Committee on Appropriations of the U.S. House of Representatives by EPA; EPA-540/1-75-022; Washington, DC, July 1975. (11) Attaran, A.; Roberts, D. R.; Curtis, C. F.; Kilama, W. L. Balancing risks on the backs of the poor. Nat. Med. 2000, 6 (7), 729-731. (12) Hoh, E.; Hites, R. A. Sources of toxaphene and other organochlorine pesticides in North America as determined by air measurements and potential source contribution function analyses. Environ. Sci. Technol. 2004, 38, 4187-4194. (13) Li, Y. F.; Bidleman, T. F.; Barrie, L. A. Toxaphene in the United States. 2. Emissions and residues. J. Geophy. Res. 2001, 106, 17929-17938. (14) Li, Y. F.; MacDonald, R. Sources and pathways of selected organochlorine pesticides to the Arctic and the effect of pathway divergence on HCH trends in biota: A review. Sci. Total Environ. 2005, 342, 87-106. (15) Harner, T.; Bidleman, T. F.; Wideman, J.; Jantunen, L. M. M.; Parkhurst, W. J. Residues of organochlorine pesticides in Alabama soils. Environ. Pollut. 1999, 106, 323-332. (16) Hoff, R. M.; Muir, D. C. G.; Grift, N. P. The annual cycle of polychlorinated biphenyls and oranohalogen pesticides in air in Southern Ontario. 1. Air concentration data. Environ. Sci. Technol. 1992, 26, 266-275. (17) Cortes, D. R.; Bash, I.; Sweet, C. W.; Brice, K. A.; Hites, R. A. Temporal trends in gas-phase concentrations of chlorinated pesticides measured at the shores of the Great Lakes. Environ. Sci. Technol. 1998, 32, 1920-1927. (18) Ma, J.; Venkatesh, S.; Li, Y. F.; Daggupaty, S. M. Tracking toxaphene in the North American Great Lakes basin. 1. Impact of toxaphene residues in the U.S. soils. Environ. Sci. Technol. 2005, 39, 8123-8131.

7214

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 41, NO. 21, 2007

(19) Baker, C. B. Area averaged temperature time series for China, India, and the United States; NOAA; available at http:// www.ncdc.noaa.gov/oa/climate/online/doe/ usa1x1.temp. (20) Hansen, J. R. R.; Glascoe, J.; Sato, M. GISS analysis of surface temperature change. J. Geophys. Res. 1999, 104, 30997-31022. (21) Karl, T. R.; Williams, C. N., Jr.; Young, P. J.; Wendland, W. M. A model to estimate the time of observation bias associated with monthly mean maximum, minimum, and mean temperatures for the United States. J. Clim. Appl. Meteorol. 1986, 25, 145-160. (22) Willies, G. H.; McDowell, L. L.; Harper, L. A.; Southwick, L. M.; Smith, S. Seasonal disappearance and volitilization of toxaphene and DDT from a cotton field. J. Environ. Qual. 1983, 12, 80-85. (23) U.S. Environmental Protection Agency. Environmental Fate and Effects Division, Pesticide Environmental Fate One Line Summary: DDT (p,p′); Washington, DC, 1989. (24) Nash, R. G.; Beall, M. L.; Harris, W. G. Toxaphene and 1,1,1trichloro-2,2-bis (p-chlorophenyl) ethane (DDT) losses from cotton in an agroecosystem chamber. J. Agric. Food Chem. 1977, 25, 336-341. (25) Shoeib, M.; Harner, T. Using measured octanol-air partition coefficients to explain environmental partitioning of organochlorine pesticides. Environ. Toxicol. Chem. 2002, 21, 984990. (26) Kustas, W. P.; Choudhury, B. J.; Kunkel, K. E.; Gay, L. W. Estimate of the aerodynamic roughness parameters over an incomplete canopy cover of cotton. Agric. For. Meteorol. 1998, 46, 91-105. (27) Ma, J.; Daggupaty, S. M.; Harner, T.; Li, Y. F. Impacts of lindane usage in the Canadian Prairies on the Great Lakes ecosystem. 1. Coupled atmospheric transport model and modelled concentrations in air and soil. Environ. Sci. Technol. 2003, 37, 37743781. (28) Harner, T.; Bidleman, T. F.; Jantunen, L. M. M.; Mackay, D. Soil-air exchange model of persistent pesticides in the United State cotton belt. Environ. Toxicol. Chem. 2001, 20, 1612-1621. (29) Scholtz, M. T.; Voldner, E.; McMillan, A. C.; VanHeyst, B. J. A pesticide emission model (PEM) Part I: model development. Atmos. Environ. 2002, 36, 5005-5013. (30) Hansen, J. E.; Sato, M.; Lacis, A.; Ruedy, R.; Tegen, I.; Matthews, E. Climate forcing in the Industrial era. Proc. Natl. Acad. Sci. 1998, 95, 12753-12758. (31) Meehl, G. L.; Washington, W. M.; Wigley, T. M.; Arblaster, L.; Dai, A. Solar and greenhouse gas forcing and climate response in the Twentieth century. J. Clim. 2003, 16, 426-444. (32) Liss, P. S.; Slater, P. G. Flux of gases across the air-sea interface. Nature 1974, 247, 181-184. (33) Li, Y. F. Global gridded technical hexachlorocyclohexane usage inventories using a global cropland as a surrogate. J. Geophys. Res. 1999, 104, 23785-23797. (34) Hu, Z.; Yang, S.; Wu, R. Long-term climate variations in China and global warming signals. J. Geophys. Res. 2003, 108, doi: 10.1029/2003JD003651.

Received for review July 26, 2006. Revised manuscript received February 13, 2007. Accepted February 19, 2007. ES0617744