Milwaukee, WI, as a Source of Atmospheric PCBs to Lake Michigan

Dec 3, 2004 - A field study of atmospheric PCBs in Milwaukee, WI, U.S.A. was conducted on the shore of Lake Michigan. We believe this is the first rep...
18 downloads 12 Views 327KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 57-63

Milwaukee, WI, as a Source of Atmospheric PCBs to Lake Michigan DAVID M. WETHINGTON, III† AND KERI C. HORNBUCKLE* Department of Civil and Environmental Engineering and IIHR-Hydroscience and Engineering, The University of Iowa, Iowa City, Iowa 52242

A field study of atmospheric PCBs in Milwaukee, WI, U.S.A. was conducted on the shore of Lake Michigan. We believe this is the first report of atmospheric PCBs in Milwaukee, although PCBs are well-known to contaminate the sediments of the Milwaukee River and Outer Harbor. Concentrations of PCBs collected during the June 2001 study are similar to concentrations in other urban-industrial areas and higher than PCBs in background air. The average ΣPCB (sum of 88 congener groups) gas-phase concentration in Milwaukee was 1.9 ng m-3 ( standard deviation 0.78 ng m-3. The average and standard deviation for the particulate-associated PCBs are 0.05 ( 0.02 ng m-3. Particulate-phase PCBs account for less than 5% of the total atmospheric concentration. PCBs in Milwaukee air are a source of PCBs to Lake Michigan. Calculated net gas exchange fluxes predicted for the Milwaukee sampling period ranged from -60 to -400 ng m-2 d-1, where net deposition is indicated by the negative sign. Calculated particleassociated PCB deposition ranged from 80 to 500 ng m-2 d-1. Most of the particle-phase deposition flux is a result of coarse particle deposition and decreases rapidly with distance from shore. Under typical meteorological conditions, particle-associated PCBs depositional flux to the lake surface decreases by 90% within 40 km. For net gasexchange, the flux reaches zero at about the same distance. At greater distances, particle-phase PCB deposition is negligible, and PCBs are volatilizing at a higher rate than they are being deposited. We calculated that Milwaukee air contributes about 120 kg of PCBs to Lake Michigan each year. This is about 10 times larger than the discharge of PCBs from the Milwaukee River.

Introduction Polychlorinated biphenyls (PCBs) contaminate the environment in the city of Milwaukee, WI, as a result of historical use and discharge of PCBs from various industries, such as packing plants, breweries, tanneries, machine shops, and foundries. These industries benefited from Milwaukee’s centralized Midwestern location on the shore of Lake Michigan, which provided a commerce link to the east via the Great Lakes, and numerous rail lines as a gateway to the West. Post World War II, Milwaukee’s industrial economy reached its peak, with tens of thousands of employees working in heavy industries during the 1960s (1). At this time, PCBs * Corresponding author phone: (319)384-0789; fax: (319)335-5660; e-mail: [email protected]. † Current address: U.S. Army Corps of Engineers, Chicago District, 111 N Canal Street, Suite 600, Chicago, IL 60606. 10.1021/es048902d CCC: $30.25 Published on Web 12/03/2004

 2005 American Chemical Society

were common additives to cutting fluids, electrical transformers, light ballasts, and lubricants, so it is not surprising that these highly recalcitrant organic compounds are found in the Milwaukee area. Studies since the early 1970s have identified the Milwaukee River Watershed as severely contaminated by PCBs (2-4). As a result, the Milwaukee River now contributes about 12 kg of PCBs, annually, to Lake Michigan through direct tributary discharge (5). This is a major source, contributing about 10% of the total tributary PCB load, excluding Green Bay (5). (The Fox River is a major source of PCBs to Green Bay, but a small portion of this is exported to Lake Michigan (6).) The level of contamination in Milwaukee prompted the International Joint Commission and the U.S. Environmental Protection Agency to designate a 22 square-mile area of the Milwaukee River Estuary as one of 43 Areas of Concern (AOC) in the Great Lakes. PCBs are of concern because they are endocrine disruptors (7, 8), neurotoxins (9), and immunodepressants (10). Some of these human biological effects are likely at PCB levels measured in Great Lakes fish (11-13). As a result of these potential hazards, state and federal agencies have established fish consumption advisories for Lake Michigan. Several studies have indicated that atmospheric sources, especially from industrial regions, may contribute significantly to the input of PCBs to surface waters (14-17). However, there are few reports of atmospheric concentrations of PCBs for Great Lakes cities other than Chicago (16, 18-22) and Toronto (23, 24). To address these concerns, a field study was conducted in an industrial region of Milwaukee, close to Milwaukee Harbor and the shore of Lake Michigan. The major objectives of the study were to determine the magnitude of atmospheric concentrations and the potential for Milwaukee air to be a source of gas-phase and particulate PCBs to Lake Michigan.

Materials and Methods Field sampling was conducted at the Wisconsin Aquatic Technology and Environmental Research (WATER) Institute in Milwaukee, WI, on June 19-24, 2001. The Institute is located in a semi-industrial area several miles due south of the Milwaukee downtown area and sits on the shore of Lake Michigan just inside the Milwaukee Harbor. All samples were collected on the roof of the Institute, approximately 20 m above water level. High volume air samplers (Hi-Vols, Tisch Environmental, Cleves, OH), modified to hold heavy gauge aluminum cartridges packed with XAD-2 resin, were employed to measure and quantify the gas- and particulate-phase concentrations of atmospheric PCBs. A pair of Hi-Vols, 2 m apart, operated in tandem during each sampling period. During the 5-day campaign, Hi-Vol samples were changed three times each day: two 6-h samples were collected during the day, followed by one 12-h sample, which was collected at night. A slack tube manometer (Tisch Environmental, Cleves, OH) was used to calibrate the samplers at the start and end of each collection period. During each sampling period, wind speed and direction was averaged on 15-minute intervals via the on-site meteorological equipment (CSAT-3 Sonic Anemometer, Campbell Scientific, Logan, UT). The Hi-Vols pumped approximately 0.30 m3 min-1 (average flow rate) of air through a glass fiber filter (GFF; 20.3 cm × 25.4 cm, Gelman Sciences, Ann Arbor, MI), followed by a cartridge containing 40 g of XAD-2 resin (20-60 mesh size, Supelco, Bellefonte, PA), arranged in series. Prior to field use, the GFFs were individually wrapped in aluminum foil and baked at 450 °C for a minimum of 12 h. The filters were VOL. 39, NO. 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

57

for PCBs. Each sample was concentrated to 100 µL and injected with internal standards 2,4,6-trichlorobiphenyl (IUPAC No. 30) and 2,2′,3,4,4′,5,6,6′-octachlorobiphenyl (IUPAC No. 204). The samples were analyzed for PCB congeners on a Hewlett-Packard 6890 Gas Chromatograph (GC), equipped with a 63Ni microelectron capture detector (µECD) and a DB-5 capillary column (60 m × 0.25 mm i.d. × 0.25 µm, 5%-phenyl 95%-polymethylsiloxane, Model No. 122-5062, J&W Scientific Inc, Folsam, CA). Quality was evaluated using side-by-side replicate sampling, surrogate standard recoveries, and field blanks (Table 2). All samples were corrected for surrogate recoveries. PCB masses were not blank-corrected. The average relative percent difference (RPD) between the side-by-side sampling was 11% for gas-phase ΣPCBs (sum of congeners) and 35% for particulate-phase ΣPCBs collected using the Hi-Vols. The average PCB concentration of both the trip blanks and the solvent blank values are very similar, reporting at 0.015 ( 0.006 ng m-3 and 0.017 ( 0.008 ng m-3, respectively. In addition, there was no statistical difference in the values of the blank PCB concentrations between the XAD-2 and GFF sampling media. PCBs were not detected on the MOUDI substrates at levels higher than the field blanks.

TABLE 1. Fifty Percent Aerodynamic Cutoff Diameters (D50) for Each MOUDI Stage at 13 L min-1 Flow Ratesa stage

cut size (µm)

mean deposition velocity (cm s-1)

IN 1 2 3 4 5 6 7 8 9 10

51 32 20 12 8.0 4.8 2.8 1.7 0.94 0.53 0.30

52 15 5.0 2.0 1.1 0.54 0.18 0.061 0.020 0.0065 0.0024

a Particle deposition velocities are representative of average meteorological conditions recorded during this study.

allowed to cool, then desiccated, and weighed. The XAD-2 resin was preextracted as described elsewhere (25). The resin is packed in combusted aluminum cartridges, wrapped in aluminum foil, placed in an aluminum canister, and sealed in a plastic bag. Filters and XAD cartridges were stored at -10 °C until sampling. Particle size distributions of aerosol particles were determined using two Micro-Orifice Uniform Deposit Impactors (Model 110, MSP Corporation, Minneapolis, MN) or MOUDIs. The MOUDIs were positioned 1 m apart and operated in tandem. Both MOUDIs were run for 24-h periods during each of the 5 consecutive days. Flow rates were measured by connecting a rotometer (Model E700, Matheson Gas Equipment Technology Group, Montgomeryville, PA) to the inlet of the MOUDI before and after each sample run. Particle size cuts (aerodynamic diameters) were determined as a function of flow rate from a calibration algorithm provided by Pryor et al. (1999) and listed in Table 1. The range of particles collected in Milwaukee ranged from 0.3 to about 50 µm. All air samples were collected in duplicate. A total of 45 samples were collected and analyzed. Of these, 26 consisted of both a particulate-phase (GFF) and gas-phase (XAD-2) sample, 10 were trip blanks (5 GFF and 5 XAD-2), and 9 were solvent (lab) blanks. Analysis for a suite of 88 individual or coeluting PCB congeners was conducted for all samples. The methods used to extract, clean up, concentrate, and fractionate the atmospheric samples are consistent with those outlined the Lake Michigan Mass Balance Methods Compendium, Volume 2 (26). The basic procedure is described here, in brief. The XAD-2 resin and GFF were extracted for 24 h with 1:1 acetone:hexane solution, using Soxhlet systems. The MOUDI substrates (combusted aluminum foil) were extracted by sonication in 1:1 acetone:hexane solution. All samples, trip blanks, and solvent blanks were spiked with 3,5-dichlorobiphenyl (IUPAC No. 14), 2,3,5,6-tetrachlorobiphenyl (IUPAC No. 65), and 2,3,4,4′5,6-hexachlorobiphenyl (IUPAC No. 166) as recovery standards. The extracts were concentrated and eluted through a cleanup column of 3% deactivated silica gel with aliquots of hexane, dichloromethane, and methanol. The hexane aliquot was analyzed

Results PCB Concentrations. The average ΣPCB gas-phase concentration in Milwaukee was 1.9 ng m-3 ( standard deviation of 0.78 ng m-3 (Figure 1). This PCB gas-phase concentration is similar to other urban areas and higher than background levels. For example, the Integrated Atmospheric Deposition Network reports a gas-phase concentrations of 0.62 ng m-3, 2.7 ng m-3, and 1.6 ng m-3 for three samples collected in Chicago in the same month, June 2001 (Ron Hites, Indiana University, personal communication). Concentrations of ΣPCBs in Baltimore in 1996 ranged from 0.02 to 3.4 ng m-3 (17). PCB concentrations in suburban New Jersey in 19971999 ranged from 0.086 to 2.3 ng m-3 (27). Gas-phase PCB concentrations measured during the Milwaukee study are about eight times higher than atmospheric concentrations in air collected over Lake Michigan. Miller reports measurements of gas and particulate-phase PCBs over southern Lake Michigan for 1999 and 2000. These samples are the most comparable over-water values to compare to the Milwaukee values reported here and are detailed in Figure 1. In Miller’s study, over-water gas-phase ΣPCB concentration averaged 0.23 ng m-3 ( standard deviation 0.22 ng m-3 (28). The effect of wind direction on PCB concentrations, PCB congener distributions, and particle size distributions was examined in detail and reported elsewhere (29). In summary, no strong relationships between wind direction and airborne PCBs were observed. Although the ΣPCB concentrations varied almost a factor of 5, winds at this site were very dynamic, and few of our samples were collected during consistent winds from any one direction. PCB congener distributions were consistent for all samples, and the normalized congener distributions exhibited little variability. The atmospheric signal strongly resembles Aroclor 1242, which is the major PCB technical mixture contaminating the sediments of Milwaukee Harbor (4, 30). Using an Aroclor

TABLE 2. Quality Control Statistics for Hi-Vol and MOUDI Samplesa % surrogate recovery sample type

media

n

14

65

166

%

n

Hi-Vol Hi-Vol MOUDI

XAD GFF Alum.

34 32 28

94.8 ( 13.4% 91.4 ( 4.7% 108.8 ( 13.6%

87.8 ( 11.4% 81.7 ( 6.4% 120.0 ( 12.9%

89.2 ( 6.9% 92.9 ( 3.3% 102.1 ( 14.9%

10.8 ( 12.9 34.5 ( 30.5 ND

13 12 14

a

58

RPD

9

ND ) ambient PCBs not detectible on MOUDI plates higher than field blank values.

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 1, 2005

FIGURE 1. Concentrations of gas-phase ΣPCBs and particle-associated ΣPCBs measured in air samples collected in Milwaukee (this study) and collected on the R/V Lake Guardian in southern Lake Michigan (28). The sampling dates are indicated on the horizontal axis. The data noted with (/) were collected in the southwestern region of the lake, near Milwaukee and Chicago. The paired bars for the Milwaukee samples represent the two replicate samples collected during each sampling period. ‘PM’, ‘Night’, and ‘AM’ refer to the time of day the Milwaukee samples were collected.

estimation method evaluated by Sather et al. (31) and Aroclor distributions from Frame et al. (32), we found the atmospheric PCBs to consist of 81% Aroclor 1242 indicating little weathering between the source(s) and the sampling location. The mean and 95% confidence limits for the measured gasphase PCB congener distribution are provided in Figure S1 and Table S1 of the Supporting Information. The lack of variability in the congener distribution pattern and the lack of correlation with wind speed suggest that our site is close to or in the middle of major sources of atmospheric PCBs. Particulate-phase PCB concentrations in Milwaukee are much lower than gas-phase PCB concentrations (Figure 1). The average and standard deviation for the particulateassociated PCBs is 0.05 ( 0.02 ng m-3. Particulate-phase PCBs account for less than 5% of the total atmospheric concentration. Although particulate-phase PCBs are lower than gasphase samples, they are still higher than particulate-phase PCBs over Lake Michigan. Concentrations of particulatephase PCBs over the lake are about half that measured in Milwaukee. Particle Concentrations and Size Distributions. Total Suspended Particle concentrations (TSP) were comparable between the MOUDI and the Hi-Vols. TSP measured from the sum of the MOUDI plates ranged from 113 to 233 µg m-3, while TSP measured from the Hi-Vol samples ranged from 24 to 224 µg m-3. All were measured in duplicate: the difference in the five MOUDI TSP replicate samples was 11% with a standard deviation of 5%; the difference in the 13

Hi-Vol replicate samples was 18% with a standard deviation of 13%. The particle size distributions measured in the Milwaukee samples were all bimodal and enriched with particles of sizes near 2 µm and coarse particles near 12 µm (aerodynamic diameters). Several samples also included very large particles (>14 µm). Figure 2 displays two different days of replicate MOUDI samples that exhibit these categorical traits. The coarse particles are characteristic of an industrial area and are rarely observed in air collected over the lake. During the AEOLOS study in southern Lake Michigan, investigators found that particle distribution profiles were enriched in particles smaller than 10 µm except when winds were directly from the Chicago-Gary industrial region (33, 34). Gas-Exchange. Emissions of airborne PCBs in Milwaukee may result in higher concentrations over Lake Michigan and enhanced deposition of PCBs to the lake. In fact, Miller reports higher concentrations of airborne PCBs over the lake near Milwaukee (28). These data are indicated in Figure 1. These data are inconclusive, however, without a model to quantify the impact of the elevated concentrations on air/water exchange. The concentrations in Milwaukee are comparable to those found in Chicago, where models of net deposition have shown that Chicago air is an important source of the PCBs to Lake Michigan (16, 35, 36). Net gas exchange of PCBs in Lake Michigan at Milwaukee was estimated using a modified two-film gas exchange model described elsewhere (21, 37, 38). The net flux of gas-phase VOL. 39, NO. 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

59

FIGURE 2. Plots of two representative particle size distributions at the Milwaukee sampling site, plotted as concentration (µg m-3), normalized to the logarithmic bin width, versus the aerodynamic diameter. The data in the left plot were obtained on June 20, while the data the right plot were obtained on June 21. Duplicate samples are presented in each plot.

TABLE 3. Gas-Phase PCB Deposition Velocities PCB homologue

average kol ( standard deviation (cm s-1)

dichlorobiphenyl trichlorobiphenyl tetrachlorobiphenyl pentachlorobiphenyl hexachlorobiphenyl heptachlorobiphenyl octachlorobiphenyl nonachlorobiphenyl

0.00087 ( 0.00012 0.00072 ( 0.00012 0.00055 ( 0.00010 0.00028 ( 0.00006 0.00037 ( 0.00008 0.00026 ( 0.00006 0.00010 ( 0.00002 0.00002 ( 0.000004

At a 5 m/s wind speed, the flux of gas-phase PCBs from Milwaukee is about 3000 times larger than the gross depositional flux of PCBs from air to the water. Dry Particle Deposition. Deposition of PCBs associated with atmospheric particles may be responsible for a large portion of total atmospheric deposition of PCBs to the Great Lakes and to Lake Michigan in particular (45-47). This flux is a function of the particle-associated contaminant concentration, Cp in ng m-3, multiplied with a particle deposition velocity, vd in m d-1 (eq 2).

F p ) vd C p PCBs is equal to the gross volatilization minus the gross deposition. Net flux across the air-water interface is calculated using mass transfer coefficients describing the airand water-side boundary layers and the bulk air and water PCB concentration gradient (eq 1).

Fgas,net ) kolCw - kol

CgRT ) volatilization - deposition H (1)

The overall mass transfer coefficient, kol, was calculated utilizing empirical correlations for water evaporation and CO2 exchange dependent on temperature and wind speed (39, 40). Average values of kol for each PCB homologue are listed in Table 3. Temperature (T, Kelvin) was held constant at 285 K for all computations (consistent with the June lake water temperatures), while the average wind speed varied on a per-sample basis. Henry’s Law Coefficients (H, atm m3 mol-1 K-1) were corrected to 285 K (41-43). Dissolved-phase PCB concentrations (Cw) are from the U.S. EPA Lake Michigan Mass Balance sampling site MB21 as reported by Miller, 1999 (44), located approximately 10 miles offshore of the city of Milwaukee. Six samples were collected and analyzed by the EPA and range from 87 pg L-1 to 240 pg L-1. The average congener-specific concentrations were used in this calculation. Gas fluxes, Fg, were calculated on a per-congener basis, then summed, and reported as a net flux out of the lake. Over the sampling campaign, PCBs from the Milwaukee urban atmosphere are predicted to deposit into Lake Michigan. Daily gas fluxes predicted for the Milwaukee sampling period ranged from -60 to -396 ng m-2 d-1, with an average of -210 ng m-2 d-1. This is similar to fluxes reported for the Chicago area (21, 36). The flux of PCBs to water is small compared to the total amount of PCBs in the air. For example, emissions of PCBs from Milwaukee can be estimated by multiplying the mean concentration times the mean wind speed. This is equivalent to the flux of atmospheric PCBs across a vertical plane at the Lake Michigan shoreline. 60

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 1, 2005

(2)

Particle deposition velocities are a function of particle size. Many modeling efforts have assumed an average deposition velocity for all atmospheric particles. This practice is commonplace primarily because particle size data are difficult to obtain and varies widely in different types of atmospheres. Even when the particle size distribution is known, the deposition velocities themselves can be difficult to predict because they vary with meteorological conditions such as wind speed, humidity, and temperature. We estimated the potential impact of particle-bound PCB deposition to the lake surface using the Williams model (48) with modifications described by Pryor et al. (49). Details of the application of this model to the Milwaukee samples are described elsewhere (29), and key parameters are listed in Table 4. The Pryor model predicts deposition velocity (vd) as a function of three components: turbulent transfer velocity, gravitational settling, and transfer velocity across the laminar surface layer. The model considers gravitational settling for both dry and wet particles, the wet aerosols resulting from hygroscopic growth near the water boundary layer. Other than particle size, a sensitivity analysis found that the wind speed had the greatest effect on particle deposition, significantly increasing the rates of ultrafine particle deposition with increased wind speed. We were not able to efficiently collect these particles, however, so we cannot see this effect. Water temperature, particle density, and saturation ratio resulted in very little change in the overall profile of deposition velocity versus particle size. The calculated deposition velocity as a function of particle size is similar to that predicted by Caffrey et al. (33), in their study of dry deposition of elemental tracers in southern Lake Michigan. Previous estimates of PCB dry particle deposition assumed a deposition velocity of 0.15 to 0.2 cm s-1 (21, 46, 50, 51), and the Pryor model suggests this range is appropriate for particles of size near 3 µm, which is a reasonable estimate for nonurban air. It is not a very good estimate for Milwaukee air.

TABLE 4. Range of Values Used in Particle Deposition Velocity Calculations

range

measured wind speed at 10 m Uref (m s-1)

friction velocity u* (m s-1)

air temp Tair (°C)

water temp Tw (°C)

particle density, Gparticle (kg m-3)

saturation ratio, S

2.5-5.0

0.050-0.300

14-24

10-14

2000-3000

0.95-0.995

TABLE 5. Particle-Associated Dry Deposition PCB Fluxes location Chicago w/ large particles (>10 µm) Chicago w/out large particles Chicago Chicago Over-Lake Offshore Chicago Lake Michigan Lake Michigan Tainan City, Taiwan Milwaukee, WI

predicted PCB flux (ng m-2 d-1)

source (21)

530 15 210 2800-9700 79 2.5 0.55 4730 80-500

(21) (47) (55) (47) (54) (51) (56) this study

PCB deposition to nearby water associated with dry particles was calculated for 10 particle sizes using the measured mean particle-associated PCB air concentration from the Hi-Vol GFFs and measured meteorological conditions. The largest fluxes are predicted for the largest particle sizessthose particles larger than 30 µm and during the highest wind speeds. Fluxes of PCBs on particles into Lake Michigan from the city of Milwaukee are predicted to range in magnitude from 80 to 500 ng m-2 d-1, with an average of 250 ng m-2 d-1. The deposition fluxes calculated for Milwaukee are similar to other estimates of dry particle deposition in Lake Michigan (Table 5) but lower than that of Holsen et al. (45) who measured direct deposition of Chicago PCBs to coated plates. The importance of particle-associated PCB deposition decreases with distance away from shore. The rate of decrease is a function of wind speed from the city over the lake and particle size. We demonstrate this effect on PCB deposition using a first-order decay model. The model assumes a firstorder loss of particles of size i over Lake Michigan as a function of distance from Milwaukee (x), deposition velocity for particle of size i (vdi), wind speed (u) and mixing height (L). The concentration of particles of size i at Milwaukee is Cp,i,0, and the concentration of particles of size i over the lake x distance from Milwaukee is Cp,i.

(

Cp,i ) Cp,i,0 exp -

)

vdix uL

(3)

Coarse particles are rapidly deposited: Figure 3 shows the relative PCB concentration for each size class of particles as they travel over water away from Milwaukee. For particles in the submicron range, little or no loss of PCBs can be observed. By 20 km from shore, the distribution of particles has changed so fewer large particles are present, and therefore less particle-associated PCB deposition occurs. By 20 km from shore, the flux of PCBs of dry particles decreases to less than 25%. Gas Phase, Particulate Phase, and Tributary Sources of PCBs. It is interesting to compare atmospheric deposition to tributary discharge as sources of PCBs from Milwaukee. The Milwaukee River is estimated to provide about 12 kg of PCBs each year to Lake Michigan (5). To arrive at a comparable figure for atmospheric deposition, we integrated the predicted depositional flux over distance from Milwaukee, assuming that gas-phase PCB concentrations decrease

exponentially with distance. The rate of decrease in gasphase concentrations with distance was estimated from the results of Green et al. (52). Green et al. determined the annual rate of decrease using over 400 gas-phase PCB measurements collected over 18 months in the Lake Michigan region, including over the lake near Chicago. This rate of change with distance reflects the average over the 18 month period and is therefore not a function of daily meteorological conditions. This report suggests a distance of half concentration of about 20 km from the city (eq 4)

Cg ) Cg,o exp

(-0.693x 20000 )

(4)

where x is the distance from Milwaukee (m), Cg is the predicted gas-phase PCB concentration at distance x, and Cg,o is the measured gas-phase PCB concentration at Milwaukee. This distance also is consistent with measurements of gas-phase PCBs over Lake Michigan near Chicago (16, 53, 54). The decrease in gas-phase concentrations is a result of mixing with cleaner air, not from loss by deposition. That is, the calculated loss by gas exchange is less than 1% of the observed decrease in concentrations. The net gas exchange flux was calculated as a function of distance from Milwaukee and is illustrated in Figure 4. The point of zero net exchange occurs at about 40 km. Beyond 40 km, the lake is volatilizing PCBs faster than it is absorbing PCBs. For a radial distance of 40 km or 2500 km2, net gas deposition is about 100 kg yr-1. The estimate of area is a rough estimate of annual area of influence. It is not appropriate for any particular day, as the area of influence is a function of many meteorological factors (36). Particle-associated PCB fluxes decrease with distance from Milwaukee as a function of mixing with cleaner air and particle settling. Therefore the estimate of total deposition must include both causes of decreasing concentrations and subsequent decrease in deposition with distance. We assumed the same effect of mixing for particles as for gases (eq 4). The resulting deposition flux as a function of distance from Milwaukee, x in meters:

Fp,i ) vd,iCp,i exp

(

) (

)

-vp,ix -0.693 exp uL 20000

(5)

The equation was applied to 10 particle sizes and used average meteorological conditions measured in Milwaukee. The model predicts that about half of the total particle deposition occurs within 10 km of Milwaukee (Figure 4). It also predicts near-zero particle-associated PCB deposition fluxes beyond approximately 40 km. This outcome is only reasonable for the conditions examined here and assumes no other sources of particles to the lake. Additionally, this model is intended to describe annual trends rather than predict gas-phase PCB fluxes for a specific day. Both particle- and gas-phase methods used average meteorological conditions measured in Milwaukee and assume that the June 2001 measured atmospheric concentrations are representative of the annual PCB concentrations in Milwaukee. (Unpublished IADN datas Ron Hites Indiana University personal communications show that the average measured concentration of PCBs in Chicago in 2001 is not significantly different than the June VOL. 39, NO. 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

61

FIGURE 3. Predicted decrease in particulate PCB concentrations by particle size in air over Lake Michigan at increasing distances from Milwaukee. Conditions are representative of those measured or estimated for Milwaukee June conditions: wind speed 5 m s-1; mixing height 500 m; aerodynamic diameters for MOUDI deposition plates.

Literature Cited

FIGURE 4. Net gas exchange (black region) and dry particle deposition of atmospheric PCBs to Lake Michigan as a function of distance from Milwaukee. Negative values indicate depositional fluxes. 2001 PCB concentration at the 95% confidence level.) When integrated for the 40 km radius from Milwaukee, the total particle-associated PCB deposition is about 17 kg per year. The total estimated atmospheric PCB deposition (∼120 kg net deposition or ∼150 gross deposition per year) from Milwaukee is about 10 times larger than the direct discharge of PCBs via flow from the Milwaukee River.

Acknowledgments We appreciate the assistance of Val Klump and the staff at the Great Lakes WATER Institute. Bill Eichinger, IIHRHydroscience and Engineering, supplied the meteorological equipment deployed during our study. This work was funded in part by the Center for Global and Regional Environmental Research (CGRER), IIHR-Hydroscience and Engineering, and the University of Iowa Environmental Health Science Research Center (Grant P30 ES05605).

Supporting Information Available Mean and 95% confidence limits for the measured gas-phase PCB congener distribution. This material is available free of charge via the Internet at http://pubs.acs.org. 62

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 1, 2005

(1) Gurda, J. The Making of Milwaukee; Milwaukee County Historical Society, Burton & Mayer, Inc.: Brookfield, WI, 1999. (2) Steuer, J. S.; Fitzgerald, S. A.; Hall, D. W. Distribution and Transport of Polychlorinated Biphenyls and Associated Particulates in the Milwaukee River System, Wisconsin, 1993-95; United States Geological Survey, WRIR-99-4100, 1999. (3) Wisconsin Department of Natural Resources. State of the Milwaukee River Basin; PUBL-WT-704-2001, 2001. (4) Rachdawong, P.; Christensen, E. R.; Chi, S. Source identification of PCBs in sediments from the Milwaukee Harbor Estuary, USA. Water. Sci. Technol. 1998, 37, 199-206. (5) U.S. Environmental Protection Agency, Lake Michigan Lakewide Managment Plan (LaMP) 2004 Status Report. http:// www.epa.gov/glnpo/lakemich/2004update/. (6) U.S. Environmental Protection Agency, Lake Michigan Mass Balance Study Loadings Report. http://www.epa.gov/glnpo/ lmmb/results/loadpcbs.html. (7) Walkowiak, J.; Wiener, J. A.; Fastabend, A.; Heinzow, B.; Kramer, U.; Schmidt, E.; Steingruber, H. J.; Wundram, S.; Winneke, G. Environmental exposure to polychlorinated biphenyls and quality of the home environment: effects on psychodevelopment in early childhood. Lancet 2001, 358, 1602-1607. (8) Kester, M. H. A.; Bulduk, S.; Tibboel, D.; Meinl, W.; Glatt, H.; Falany, C. N.; Coughtrie, M. W. H.; Bergman, A.; Safe, S. H.; Kuiper, G.; Schuur, A. G.; Brouwer, A.; Visser, T. J. Potent inhibition of estrogen sulfotransferase by hydroxylated PCB metabolites: A novel pathway explaining the estrogenic activity of PCBs. Endocrinology 2000, 141, 1897-1900. (9) Schantz, S. L.; Widholm, J. J.; Rice, D. C. Effects of PCB exposure on neuropsychological function in children. Environ. Health Perspect. 2003, 111, 357-376. (10) Weisglas-Kuperus, N.; Patandin, S.; Berbers, G. A. M.; Sas, T. C. J.; Mulder, P. G. H.; Sauer, P. J. J.; Hooijkaas, H. Immunologic effects of background exposure to polychlorinated biphenyls and dioxins in Dutch preschool children. Environ. Health Perspect. 2000, 108, 1203-1207. (11) Schantz, S. L.; Gasior, D. M.; Polverejan, E.; McCaffrey, R. J.; Sweeney, A. M.; Humphrey, H. E. B.; Gardiner, J. C. Impairments of memory and learning in older adults exposed to polychlorinated biphenyls via consumption of Great Lakes fish. Environ. Health Perspect. 2001, 109, 605-611. (12) Schantz, S. L.; Sweeney, A. M.; Gardiner, J. C.; Humphrey, H. E. B.; McCaffrey, R. J.; Gasior, D. M.; Srikanth, K. R.; Budd, M. L. Neuropsychological assessment of an aging population of Great Lakes fisheaters. Toxicol. Ind. Health 1996, 12, 403-417. (13) Schantz, S. L.; Gardiner, J. C.; Gasior, D. M.; Sweeney, A. M.; Humphrey, H. E. B.; McCaffrey, R. J. Motor function in aging Great Lakes fisheaters. Environ. Res. 1999, 80, S46-S56. (14) Bamford, H. A.; Ko, F. C.; Baker, J. E. Seasonal and annual airwater exchange of polychlorinated biphenyls across Baltimore Harbor and the northern Chesapeake Bay. Environ. Sci. Technol. 2002, 36, 4245-4252.

(15) Park, J. S.; Wade, T. L.; Sweet, S. Atmospheric deposition of organochlorine contaminants to Galveston Bay, Texas. Atmos. Environ. 2001, 35, 3315-3324. (16) Zhang, H. X.; Eisenreich, S. J.; Franz, T. R.; Baker, J. E.; Offenberg, J. H. Evidence for increased gaseous PCB fluxes to Lake Michigan from Chicago. Environ. Sci. Technol. 1999, 33, 2129-2137. (17) Offenberg, J. H.; Baker, J. E. Influence of Baltimore’s urban atmosphere on organic contaminants over the northern Chesapeake Bay. J. Air Waste Manage. 1999, 49, 959-965. (18) Simcik, M. F.; Eisenreich, S. J.; Lioy, P. J. Source apportionment and source/sink relationships of PAHs in the coastal atmosphere of Chicago and Lake Michigan. Atmos. Environ. 1999, 33, 50715079. (19) Offenberg, J. H.; Baker, J. E. PCBs and PAHs in southern Lake Michigan in 1994 and 1995: Urban atmospheric influences and long-term declines. J. Great Lakes Res. 2000, 26, 196-208. (20) Offenberg, J. H.; Baker, J. E. Polychlorinated biphenyls in Chicago precipitation: Enhanced wet deposition to near-shore Lake Michigan. Environ. Sci. Technol. 1997, 31, 1534-1538. (21) Miller, S. M.; Green, M. L.; Depinto, J. V.; Hornbuckle, K. C. Results from the Lake Michigan mass balance study: Concentrations and fluxes of atmospheric polychlorinated biphenyls and trans-nonachlor. Environ. Sci. Technol. 2001, 35, 278-285. (22) Cotham, W. E.; Bidleman, T. F. Polycyclic aromatic hydrocarbons and polychlorinated biphenyls in air at an urban and rural site near Lake Michigan. Environ. Sci. Technol. 1995, 29, 27822789. (23) Helm, P. A.; Bidleman, T. F. Current combustion-related sources contribute to polychlorinated naphthalene and dioxin-like polychlorinated biphenyl levels and profiles in air in Toronto, Canada. Environ. Sci. Technol. 2003, 37, 1075-1082. (24) Gingrich, S. E.; Diamond, M. L. Atmospherically derived organic surface films along an urban-rural gradient. Environ. Sci. Technol. 2001, 35, 4031-4037. (25) Hornbuckle, K. C.; Smith, G. L.; Miller, S. M.; Eadie, B. J.; Lansing, M. B. Magnitude and origin of polychlorinated biphenyl (PCB) and dichlorodiphenyltrichloroethane (DDT) compounds resuspended in southern Lake Michigan. J. Geophys. Res.-Oceans 2004, 109, Art. No. C05017 MAY 05018. (26) U.S. EPA Lake Michigan Mass Balance Study (LMMB) Methods Compendium, Volume 2: Organic and Mercury Sample Analysis Techniques; United States Environmental Protection Agency: 1997. (27) Brunciak, P. A.; Dachs, J.; Gigliotti, C. L.; Nelson, E. D.; Eisenreich, S. J. Atmospheric polychlorinated biphenyl concentrations and apparent degradation in coastal New Jersey. Atmos. Environ. 2001, 35, 3325-3339. (28) Miller, S. M. The Effects of Large-Scale Episodic Sediment Resuspension on Persistent Organic Pollutants in Southern Lake Michigan, Ph.D. Dissertation, University of Iowa, Iowa City, IA, 2003. (29) Wethington, D. M. Atmospheric concentrations and vertical fluxes of persistent organic pollutants from Milwaukee, Wisconsin, M.S. Thesis, University of Iowa, Iowa City, IA, 2002. (30) Rachdawong, P.; Christensen, E. R. Determination of PCB sources by a principal component method with nonnegative constraints. Environ. Sci. Technol. 1997, 31, 2686-2691. (31) Sather, P. J.; Newman, J. W.; Ikonomou, M. G. Congener-based aroclor quantification and speciation techniques: A comparison of the strengths, weaknesses, and proper use of two alternative approaches. Environ. Sci. Technol. 2003, 37, 5678-5686. (32) Frame, G. M.; Wagner, R. E.; Carnahan, J. C.; Brown, J. F.; May, R. J.; Smullen, L. A.; Bedard, D. L. Comprehensive, quantitative, congener-specific analyses of eight aroclors and complete PCB congener assignments on DB-1 capillary GC columns. Chemosphere 1996, 33, 603-623. (33) Caffrey, P. F.; Ondov, J. M.; Zufall, M. J.; Davidson, C. I. Determination of size-dependent dry particle deposition velocities with multiple intrinsic elemental tracers. Environ. Sci. Technol. 1998, 32, 1615-1622. (34) Offenberg, J. H.; Baker, J. E. Aerosol size distributions of polycyclic aromatic hydrocarbons in urban and over water atmospheres. Environ. Sci. Technol. 1999, 33, 3324-3331. (35) Simcik, M. F.; Zhang, H. X.; Eisenreich, S. J.; Franz, T. P. Urban contamination of the Chicago coastal Lake Michigan atmosphere by PCBs and PAHs during AEOLOS. Environ. Sci. Technol. 1997, 31, 2141-2147. (36) Hornbuckle, K. C.; Green, M. L. The impact of an urban-industrial region on the magnitude and variability of persistent organic

(37)

(38)

(39)

(40)

(41)

(42)

(43)

(44)

(45)

(46)

(47)

(48) (49)

(50)

(51)

(52)

(53)

(54)

(55) (56)

pollutant deposition to Lake Michigan. Ambio 2003, 32, 406411. Achman, D. R.; Hornbuckle, K. C.; Eisenreich, S. J. Volatilization of Polychlorinated Biphenyls from Green Bay, Lake Michigan. Environ. Sci. Technol. 1993, 27, 75-87. Hornbuckle, K. C.; Sweet, C. W.; Eisenreich, S. J. Seasonal variations in air-water exchange of polychlorinated biphenyls in Lake Superior. Environ. Sci. Technol. 1994, 28, 1491-1501. Wanninkhof, R. H.; Bliven, L. F. Relationship between gas exchange, wind speed, and radar backscatter in a large windwave tank. J. Geophys. Res. 1991, 96, 2785-2796. Schwarzenbach, R. P.; Gschwend, P. M.; Imboden, D. M. Environmental Organic Chemistry; Wiley-Interscience: New York, 1993. Brunner, S.; Hornung, E.; Santl, H.; Wolff, E.; Piringer, O. G.; Altschuh, J.; Bruggemann, R. Henry’s law constants for polychlroinated biphenyls: experimental determination and structure-property relationships. Environ. Sci. Technol. 1990, 24, 1751-1754. Bamford, H. A.; Poster, D. L.; Baker, J. E. Temperature dependence of Henry’s law constants of thirteen polycyclic aromatic hydrocarbons between 4 degrees C and 31 degrees C. Environ. Toxicol. Chem. 1999, 18, 1905-1912. Bamford, H. A.; Poster, D. L.; Baker, J. E. Henry’s law constants of polychlorinated biphenyl congeners and their variation with temperature. J. Chem. Eng. Data 2000, 45, 1069-1074. Miller, S. M. Spatial and Temporal Variability of Organic and Nutrient Compounds in Atmospheric Media Collected During the Lake Michigan Mass Balance Study, M.S. Thesis, State University of New York at Buffalo, Buffalo, NY, 1999. Holsen, T. M.; Noll, K. E.; Liu, S. P.; Lee, W. J. Dry Deposition of Polychlorinated-Biphenyls in Urban Areas. Environ. Sci. Technol. 1991, 25, 1075-1081. Hoff, R. M.; Strachan, W. M. J.; Sweet, C. W.; Chan, C. H.; Shackleton, S. M.; Bidleman, T. F.; Brice, K. A.; Burniston, D. A.; Cussion, S.; Gatz, D. F.; Harlin, K.; Schroeder, W. H. Atmospheric Deposition of Toxic Chemicals to the Great Lakes: A Review of Data Through 1994. Atmos. Environ. 1996, 30, 3505-3527. Franz, T. P.; Eisenreich, S. J.; Holsen, T. M. Dry deposition of particulate polychlorinated biphenyls and polycyclic aromatic hydrocarbons to Lake Michigan. Environ. Sci. Technol. 1998, 32, 3681-3688. Williams, R. M. A model for the dry deposition of particles to natural water surfaces. Atmos. Environ. 1982, 16, 1933-1938. Pryor, S. C.; Barthelmie, R. J.; Geernaert, L. L. S.; Ellermann, T.; Perry, K. D. Speciated particle dry deposition to the sea surface: results from ASEPS ’97. Atmos. Environ. 1999, 33, 2045-2058. Strachan, W. M. J.; Eisenreich, S. J. Mass balancing of toxic chemicals in the Great Lakes: The role of atmospheric deposition; Report to the International Joint Commission, 1988. Eisenreich, S. J.; Strachan, W. M. J. Estimating Atmospheric Deposition of Toxic Substances to The Great Lakes - an Update -; Great Lakes Protection Fund and Environment Canada, 1992. Green, M. L.; Depinto, J. V.; Sweet, C.; Hornbuckle, K. C. Regional spatial and temporal interpolation of atmospheric PCBs: Interpretation of Lake Michigan mass balance data. Environ. Sci. Technol. 2000, 34, 1833-1841. Simcik, M. F.; Franz, T. P.; Zhang, H. X.; Eisenreich, S. J. Gasparticle partitioning of PCBs and PAHs in the Chicago urban and adjacent coastal atmosphere: States of equilibrium. Environ. Sci. Technol. 1998, 32, 251-257. Pirrone, N.; Keeler, G. J.; Holsen, T. M. Dry Deposition of Semivolatile Organic-Compounds to Lake- Michigan. Environ. Sci. Technol. 1995, 29, 2123-2132. Holsen, T. M. Dry deposition of polychlorinated biphenyls in urban areas. Environ. Sci. Technol. 1991, 25, 1075-1081. Lee, W. J.; Lewis, S. J. L.; Chen, Y. Y.; Wang, Y. F.; Sheu, H. L.; Su, C. C.; Fan, Y. C. Polychlorinated biphenyls in the ambient air of petroleum refinery, urban and rural areas. Atmos. Environ. 1996, 30, 2371-2378.

Received for review July 15, 2004. Revised manuscript received October 14, 2004. Accepted October 15, 2004. ES048902D

VOL. 39, NO. 1, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

63