Journal of Environmental Quality 30:768-775 (2001)
© 2001 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
TECHNICAL REPORT
Atmospheric Pollutants and Trace Gases
Atmospheric Movements of Lindane (
-Hexachlorocyclohexane) from Canola Fields Planted with Treated Seed
D.T. Waitea,
N.P. Gurprasadb,
J.F. Sproullb,
D.V. Quiringa and
M.W. Kotylakc
a Environment Canada, 300-2365 Albert St., Regina, SK, S4P 4K1, Canada
b Environment Canada, EP Labs, 5320-122 St., Edmonton, AB, T6H 3S5, Canada
c Univ. of Regina, Faculty of Engineering, 3737 Wascana Parkway, Regina, SK, S4S 0A2, Canada
Corresponding author (Don.Waite{at}ec.gc.ca)
Received for publication February 3, 2000.
 |
ABSTRACT
|
|---|
Lindane (
-hexachlorocyclohexane [-HCH]) is used as an insecticide in many countries. Concentrations of -HCH have been found in air, water, soil, snow, and tissue samples throughout the world and concerns have been raised for its potential effects on human and ecosystem health. In Canada, -HCH is primarily used as a treatment on canola (Brassica napus L) seed with an estimated 455.3 Mg applied in 1997 and 510.4 Mg in 1998. The purpose of this study was to measure -HCH volatilization from fields planted with treated canola seed. Atmospheric dry and wet deposition and soil samples were collected for two growing seasons (1997 and 1998) from a canola field planted with treated seed. Atmospheric concentrations as high as 16.1 and 7.4 ng m-3 were measured at 1 m above the canola field compared with maximum concentrations of 2.9 and 2.7 ng m-3 measured above a grass field located 2 km away (1997 and 1998, respectively). On the basis of measurements made in this study it was estimated that between 12 and 30% of the -HCH applied as canola seed treatment may volatilize and be released to the atmosphere. This would create an atmospheric loading of 66.4 to 188.8 Mg for the 6-wk period following planting, estimated from the quantity of seed sown on the Canadian prairies in 1998. Dry deposition rates and rain concentrations as high as 2203 ng m-2 d-1 and 170 ng L-1 were measured adjacent to the canola field.
Abbreviations: GC–MS, gas chromatography with mass spectrometric detection • -HCH, -hexachlorocyclohexane (lindane)
|
INTRODUCTION
|
|---|
LINDANE (-HCH) is used as an insecticide in many countries with an estimated 20400 Mg applied in 1990 (Li et al., 1996). Concentrations of -HCH have been reported in the atmosphere of continental Europe, Scandinavia, the northern Atlantic Ocean (Haugen et al., 1998), and the arctic (Hargrave et al., 1997) ranging from 0.006 to 3.5 ng m-3 (Haugen et al., 1998). Concerns have been raised on the human and ecosystem health effects of -HCH because of its potential to accumulate in the fatty tissues of animals and its suspected action as a carcinogen, teratogen, fetotoxin, and reproductive toxin (e.g., USEPA, 1980; Sang et al., 1999; Beard and Rawlings, 1999; Silvestroni and Palleschi, 1999). Li et al. (1996) estimated that 284.5 Mg of -HCH were applied in Canada in 1990, where it is primarily used as a treatment on seeds or as a soil treatment (Pest Management Regulatory Agency, 1999). -HCH, often in combination with some fungicides, is frequently applied to canola seed, generally before it is purchased by the farmer. Canola is primarily grown in the prairie region of Canada (the provinces of Alberta, Saskatchewan, and Manitoba), with the greatest amount grown in Saskatchewan. Canola hectarage has increased from 2551500 ha (1986) to 4819500 ha (1997) and 5402575 ha (1998) (Statistics Canada, 1999), with approximately 95% of the seed pretreated with lindane at a rate of 15.3 g per kg of seed. Seeding rates generally range from 5 to 8 kg ha-1 and the seeds are planted at an approximate 2.5-cm depth in the soil (Thomas, 1999). Using an average seeding rate of 6.5 kg seed ha-1, it can be estimated that lindane use on the Canadian prairies was approximately 455.3 (1997) and 510.4 Mg (1998).
In 1994, -HCH concentrations of 1.3 ng m-3 were measured in the air near Regina, Saskatchewan (Waite et al., 1999). These concentrations were high compared with those reported in many other parts of the world. At that time there was little canola cultivation in that part of the province, canola being generally grown in the cooler and wetter northern region of the grain belt. The purpose of the present study was to determine if (i) significant quantities of -HCH volatilize from canola fields, (ii) the quantities volatilizing from canola fields could influence regional atmospheric concentrations, and (iii) regional volatilization of -HCH, from canola growing areas of the Canadian prairies, influences concentrations found in the Regina area.
Concentrations of -HCH were measured at various heights above a canola field (C site) planted with lindane-treated seed and above a neighboring grassy field (G site) that had never been cultivated or treated with lindane. The movement of -HCH from the field was estimated and extrapolated to atmospheric losses that might occur throughout the Canadian prairies.
|
METHODS
|
|---|
Sampling Sites and Methods
Samples were collected, in 1997 and 1998, from two sites located in the black soil zone in southcentral Saskatchewan (Fig. 1). The C site (canola site) was located on a small (0.5 ha) canola field that was seeded to -HCH-treated canola for the first time in 1997. The farmer had, in previous years, incorporated carbofuran (2,3-dihydro-2,2-dimethylbenzofuran-7-yl methylcarbamate) pellets for root insect control. The canola was planted at a depth of approximately 3 cm. The G site (grassy site), located on a grassy field in an abandoned farmyard that had never been planted to crops or treated with -HCH, was approximately 2 km west–southwest of the C site. In 1997, ambient air samples were collected at 100 and 300 cm aboveground, at both sites, beginning on 27 May, 1 d after the canola seed was sown on the C site. Weekly composite samples were collected for 5 wk following sowing. In 1998, ambient air samples were collected at 30, 100, and 300 cm at the C site and at 100 and 300 cm on the G site, beginning on 12 May, 1 d after canola was sown on the C site and continuing for 6 wk until 23 June. Additional 1-wk samples were collected at both sites and at all heights, from 18 to 25 Aug. 1998, immediately after the crop was swathed.
The methods used for sampling and analyzing ambient air for -HCH were described by Waite et al. (1999) and will be briefly summarized here. Ambient air was sampled using a high-volume air sampler (Model PS1, General Metal Works, Village of Cleves, OH) collecting ambient concentrations of -HCH on a sampling unit consisting of a glass fiber filter followed by a polyurethane foam (PUF) plug and XAD-2 resin cartridge (Rohm and Haas, West Hill, ON, Canada). The PUF samplers were calibrated to aspirate 2100 m3 during the 7-d sampling period. Brorstrom-Lunden et al. (1994) reported no breakthrough of -HCH at higher flow rates (20 m3 h-1) and similar sample volumes. Ambient air samples were collected at 30-, 100-, and 300-cm heights during this study. The stock Model PS-1 sampler collects samples at the 100-cm height. To collect the 300-cm height samples, the samplers were mounted on a 200-cm-high platform. A special 30-cm sampler was built, following the exact same sampling geometry of the PS-1, to permit collection at that height.
Weekly composite dry and wet atmospheric deposition samples, at a 1-m height, were collected from both sites from 12 May to 23 June 1998, using a unique sampler described by Waite et al. (1999). This sampler collects dry atmospheric deposits on a sheet of water that passes across a horizontal 0.5- x 1.0-m stainless steel tray. Any particles that deposit to the water and gases that partition into it are swept into a reservoir contained within the body of the sampler. Water is recirculated from the reservoir back to the surface of the sampler via a composite glass fiber and XAD-2 resin sampling cartridge that removes particles and dissolved organic compounds from the water. This material accumulates in the sampling cartridge, representing a composite dry deposition sample for the period for which the sampler is operated; in the case of this study for 7 d. At the end of the sampling period the cartridge is removed, extracted, and analyzed, as described in the following section on extraction and analytical procedures. In the event of rainfall, a sensor is triggered, the water flow ceases, and rain water falling on the stainless steel collecting tray is diverted into a separate collecting container via a series of valves. In this study the rain sample container was emptied at the end of the 7-d sampling period and represented a composite rain sample for that time duration. Using this sampler, then, it is possible to collect separate dry and wet deposition samples. Rain volume was measured separately with a standard rain gauge.
Soil samples, two at each site, were collected on 3 June and 9 July 1997 as follows. The vegetation, if present, was removed by cutting and a rectangle 25 x 12.5 cm was marked on the soil. The top 3 cm of soil was removed with an acetone-rinsed shovel and the soil sample stored in a new, acetone-rinsed jar, which was capped with acetone-rinsed foil and a lid. In 1998, soil samples were taken at each site with a 2-cm-diam. Oakfield corer (Oakfield Apparatus Co., Oakfield, WI). The results were inconsistent, probably because the -HCH was located on the seed in discrete rows rather than being evenly dispersed on the soil. The core sampler may have missed or oversampled the seed rows resulting in extreme variability of the data. For this reason, the 1998 soil data cannot be used.
Weekly average wind speed and rainfall were measured at the Environment Canada Langham atmospheric station located in the center of Redberry Lake, 39 km east of the C site (M. Arts, Environment Canada, Saskatoon, SK, personal communication, 1997, 1998).
Extraction and Analytical Procedures
Ambient air samples collected on glass fiber filters plus PUF/XAD-2 sampling units were extracted in a Soxhlet apparatus with acetone for at least 8 h. The extracts were concentrated to 5 mL by rotary evaporation and solvent-exchanged to cyclohexane. The rate of recovery based on six spiked sampling units (PUF/XAD-2 resin cartridge plus glass fiber filter) was 105 ± 14% (standard deviation).
The dry deposition samples collected on XAD-2 resin columns were eluted with 1000 mL of acetone. The acetone eluant was concentrated by rotary evaporation until only water from the column remained. The water was transferred to a separatory funnel and extracted with three separate portions of 60 mL dichloromethane, draining the organic phase through cotton wool. The dichloromethane extracts were concentrated to 5 mL by rotary evaporation and solvent-exchanged to cyclohexane. The rate of recovery based on five spiked XAD-2 resin columns was 110 ± 4% (standard deviation).
Soil samples were not dried before extraction. For soil analysis, a 20-g aliquot of each sample was extracted with 50 mL of aqueous acidic acetonitrile (75% acetonitrile, 22% water, 3% acetic acid) on a wrist-action shaker for 1 h. The mixture was centrifuged and 25 mL of the supernatant was transferred to a separatory funnel and combined with 100 mL of 5% sodium carbonate in water. The solution in the separatory funnel was acidified with sulfuric acid and then extracted with three separate portions of 30 mL dichloromethane. The dichloromethane extract was drained through cotton wool, concentrated to 5 mL by rotary evaporation, and solvent-exchanged to cyclohexane. The soil results were based on dry weight, with percent water determined on a separate portion of sample. The rate of recovery based on six spiked soil samples was 119 ± 29% (standard deviation).
Rain samples were acidified with sulfuric acid and extracted with three separate portions of 60 mL dichloromethane. The organic phase was drained through cotton wool, concentrated to 5 mL by rotary evaporation, and solvent-exchanged to cyclohexane. The rate of recovery based on six spiked water samples was 115 ± 5% (standard deviation).
Sample extracts were analyzed by gas chromatography with mass spectrometric detection (GC–MS) using a Hewlett–Packard (Palo Alto, CA) GC–MS system including Model 7673 automatic sampler, Model 5890 gas chromatograph, and Model 5970 or 5971 mass spectrometer. The GC columns used were of the type 5% phenyl–95% methylpolysiloxane (either a Restek [Bellefont, PA] Rtx-5MS or a Rose Scientific [Edmonton, AB, Canada] RH-5MS), 30 m in length with a 0.25-mm internal diameter and a 0.25-micron film thickness. The carrier gas was helium and the injection mode was splitless.
The GC temperature zones and program were as follows: injector 300°C, initial column temperature 80°C for 1 min, ramp 20°C min-1 to 140°C, hold at 140°C for 6 min, ramp at 10°C min-1 to 300°C, hold at 300°C for 5 min. The column transfer line to the mass spectrometer was at 300°C and the mass spectrometer was operated in the selected ion mode monitoring ions 181, 183, and 219, which are characteristic of lindane. The GC–MS data were acquired by a computer equipped with Hewlett–Packard Chemstation software. Sample extract concentrations were determined using external standard calibration of the GC–MS system.
Standard material for calibration was obtained from reputable suppliers such as Supelco and ChemService.
|
RESULTS AND DISCUSSION
|
|---|
Atmospheric Concentrations of -HCH
In 1997, concentrations of -HCH in the air above the canola field (C site) ranged from 3.7 to 16.1 ng m-3 at the 100-cm height and 1.9 to 9.2 ng m-3 at the 300-cm height (Fig. 2). At the C site the highest concentrations were always found at 100 cm, indicating that the field was the primary source of the -HCH. The highest concentrations were measured during the second week after seeding (3–10 June). For the following 3 wk the atmospheric concentrations steadily declined. At the G site, where no canola was planted, similar atmospheric concentrations of -HCH were measured at both the 100- and 300-cm heights for the entire 5-wk sampling period. They ranged from 2.9 to <0.1 ng m-3 for both heights, remaining fairly consistent for the first 4 wk and declining to below the level of quantitation for the fifth week. The similarity between the 100- and 300-cm sample concentrations indicated that the air column was relatively well mixed at this location and that there was no source of -HCH in the soil beneath the samplers. The concentrations measured at the G site may represent a background level for noncultivated land in this region.
Atmospheric samples were collected, in 1998, at three heights (30, 100, and 300 cm) at the C site and at two heights (100 and 300 cm) at the G site for 6 wk following the planting of canola at the C site. An additional set of samples was collected, from both sites, during the week of 18–25 August. The pattern of concentrations of -HCH at the C site was similar to that found in 1997 with the highest concentrations measured in air samples taken closest to the ground (30 cm) at the C site for 5 of the 6 wk (Fig. 3). Proportionately lower concentrations were measured, for those same weeks, at the 100- and 300-cm heights. The highest overall concentrations were found during the second week of sampling. The absolute concentrations measured in 1998 were, however, lower than those found in 1997. The maximum concentrations measured at 100 cm were 16.1 ng m-3 (1997) and 7.4 ng m-3 (1998). Atmospheric concentrations of -HCH measured at the G site, in 1998, were also lower than those found in 1997, ranging from 2.7 to <0.1 ng m-3 (Fig. 3). Concentrations at 100 cm at the G site were slightly higher than at 300 cm. In the week of 18–25 August atmospheric concentrations of -HCH at both sites were <0.1 ng m-3.
Atmospheric concentrations of -HCH measured in 1998 (Fig. 3) were generally lower than those found in 1997, possibly because of field moisture conditions. In 1997, field moisture conditions were good as a result of a snowfall several days before the canola was planted. While soil moisture measurements were not made, periodic rainfall during the sampling period maintained field moisture, producing good crop growth. In 1998, however, the field was dry at planting time and significant rain did not occur until the sixth week of sampling, resulting in stunted crop growth compared with 1997. Information in the literature suggests that moisture levels can affect the rate of volatilization of pesticides from soil, especially under very dry conditions. Ehlers et al. (1969a)(b), for example, showed that -HCH volatilization was negligible in soils with 1% moisture content and increased proportionately with moisture content up to 3%. Similar effects of water addition to dry soils have been reported for the volatilization of the soil-incorporated herbicides triallate [S-2,3,3-trichloroallyl di-isopropyl(thiacarbamate)] and trifluralin [(
,,-trifluoro-2,6-dinitro-N,N-dipropyl-p-toluidine)] (Grover, 1991; Grover et al., 1988) and for metolachlor [2-chloro-6'-ethyl-N-(2-methoxy-1-methylethyl)acet-o-toluidide] (Prueger et al., 1999). The dry conditions prevalent for the first 5 wk of sampling in 1998, may explain the lower atmospheric concentrations of -HCH found compared with those concentrations found in 1997. Rainfall during the sixth week of sampling in 1998 may have resulted in the increased concentration of -HCH found during that week.
Calculation of Atmospheric Loading of -HCH
The concentrations of -HCH shown in Tables 1 and 2 were obtained from 7-d composite samples and, thus, represent average atmospheric concentrations for a full week. It is possible to use these data (Fig. 3) to estimate the quantity of -HCH present above the C site for the entire 1998 sampling period, attributable to volatilization from that site, by using a simple compartment model and the following assumptions. (i) The concentration of -HCH in the 30-cm sampler was assumed to represent the concentration in a column of air from the soil surface to a height halfway between the 30- and 100-cm samplers, that is, a compartment height of 65 cm. (ii) Similarly, the concentration in the 100-cm sampler was assumed to represent the column of air from 65 to 200 cm, halfway between the 100- and 300-cm samplers, that is, a compartment 135 cm high. (iii) The 300-cm sampler was assumed to represent the concentration from 200 to 400 cm, a compartment height of 200 cm (Fig. 4). Averaging the weekly quantities of -HCH measured in 1998, it may be calculated that base (0–65 cm), middle (65–200 cm), and top (200–400 cm) compartments of the air column contained 31.4, 42.2, and 26.4%, respectively, of the -HCH present in the total 400-cm-high air column. The week of 9 to 16 June 1998 was not used in this calculation because the pattern was atypical of the other weeks sampled. For that week, higher concentrations of -HCH were measured at 100 cm than at 30 cm.
View this table:
[in this window]
[in a new window]
|
Table 1. Quantities of -HCH (ng) in a 1-m2 air column, calculated from atmospheric concentrations measured at the C site in 1997 and 1998 and extrapolated to a height where atmospheric concentrations reach background concentrations. The quantities reported for the 0–65 cm portion of the column for 1997 are based on concentrations extrapolated from 1998 data (see text for explanation).
|
|
View this table:
[in this window]
[in a new window]
|
Table 2. Quantities of -HCH transported horizontally, as a result of wind action, by a 1-m2 column of air at the C site, 1997 and 1998. For purposes of this table, Sampling Week 1 represents the first week of sampling for each year. Data from 18–25 Aug. 1998 were below the limit of quantitation and were not included in this table. Quantities have been corrected for background concentrations as measured at the G site.
|
|
View larger version (28K):
[in this window]
[in a new window]
|
Fig. 4. Method of calculating atmospheric loading of -HCH including examples of log normal curves applied to the data for two weeks in 1998.
|
|
Atmospheric concentrations of -HCH were only measured at two heights, 100 and 300 cm, in 1997. Because similar patterns of concentrations were found in 1997 and 1998, the 1998 data were used to estimate the concentrations that might have occurred at 30 cm in 1997. It was assumed that the top two compartments, 65 to 200 cm and 200 to 400 cm, would contain 68.6% of the -HCH, the bottom compartment containing 31.4% (see above). Based on this assumption, the probable concentrations of -HCH at the 30-cm height were calculated for the 1997 data.
The concentrations of -HCH at each height were corrected for regional background levels by subtracting the average concentration measured at the G site for each respective sampling week. The corrected weekly atmospheric concentrations (ng m-3) were plotted on a log normal curve (Fig. 4) following the findings of Prueger et al. (1999), who reported that log normal curves accurately described atmospheric concentrations of metolachlor found above treated fields. The R2 values for the curves calculated for our data ranged from 0.86 to 0.99, with only two R2 values below 0.95, indicating that the log normal curve provided a good fit for the data. The log normal curve was extrapolated to a height where the atmospheric concentration of -HCH equaled half of the minimum quantitation level (0.05 ng m-3), which was taken to represent zero concentration (Fig. 4). The quantity of -HCH in the total air column, 1 m2 in area, was then calculated using a four-compartment model, the height of the upper compartment being determined by the log normal curve. The data generated are shown in Table 1.
Volatilization Losses of -HCH from Canola Fields
The data in Table 1 represent static quantities of -HCH present for the sampling week because they are based on average air concentrations measured in 7-d composite samples. That is, these concentrations represent the total quantity of -HCH collected for 1 wk divided by the total volume of the air sample, with the samplers in continuous operation. The air above the field was, however, not static. It is difficult to calculate the quantity of -HCH moving off of the site without actual measurements of the flux or volatilization of -HCH from the soil. A rough estimate can be made by using the wind speed as an estimate of the horizontal vector of transport. Vertical movement through eddy dispersion and diffusion is not estimated in this manner and, thus, the amount of -HCH being removed from the site may be underestimated. Average weekly wind speeds, measured at a meteorological station approximately 39 km east of the site were 3.5 to 5.7 m s-1 (12.6 to 20.5 km h-1) in 1997 and 3.5 to 5.3 m s-1 (12.5 to 19.2 km h-1) in 1998. At a wind speed of 3.5 m s-1, 3.5-m2 columns of air would be horizontally displaced in 1 s and 2116800 columns would be displaced in a 7-d period. If the air column contained 32.2 ng of -HCH (week of 27 May–3 June 1997, when this wind speed was measured), it can be calculated that approximately 68.2 mg of -HCH would be transported horizontally from the 1-m2 area. The quantities of -HCH horizontally removed from the study site as a result of wind action were calculated for both 1997 and 1998 (Table 2). An estimated 388.3 and 158.6 mg of -HCH were transported horizontally from 1 m2 of the C site in 1997 and 1998, respectively. If this is further extrapolated to an area 1 ha in size (100 x 100 m), approximately 38.83 and 15.86 g of -HCH would be horizontally transported from the leading edge of the field.
Canola seed was planted on the C site at a rate of 7.9 to 8.9 kg ha-1, compared with the rate recommended by the Canola Council of Canada of 5 to 8 kg ha-1. Assuming an average seeding rate at the C site of 8.4 kg ha-1, at the recommended seed treatment rate of 15.3 g per kg of seed, it can be calculated that -HCH was applied at a rate of 128.5 g ha-1. Horizontal atmospheric transport, as calculated above, would account for losses of 30 and 12% of applied -HCH from the field for a 5-wk period in 1997 and a 6-wk period in 1998, respectively. No comparable studies involving volatilization of lindane from planted seed were found in the literature. Waymann and Rudel (1995) reported losses of 12 to 39% of soil-applied -HCH within 24 h of application. The experiments were conducted in a wind tunnel with the insecticide sprayed onto the soil surface. Volatilization increased with higher wind velocities and was also affected by the application dose rate (higher at lower dose rates) and soil surface area (higher volatilization when smaller surface areas were tested). Grover et al. (1988) reported volatilization of 27% of the herbicide trifluralin incorporated at a depth of 5 cm into a wheat (Triticum aestivum L.) field. Prueger et al. (1999) reported a 22% loss through volatilization of metolachlor broadcast-applied to a corn field. Approximately 95% of this loss occurred within the first 12 h after spraying. The estimated losses of -HCH resulting from volatilization from canola fields, then, are within the range of results reported by other authors. It must be remembered, however, that the -HCH is applied to the canola seed coat and the seed planted belowground. The -HCH is not directly applied to the soil surface. Upon germination of the seed, this coat may, in fact, be raised to or above the soil surface and be exposed directly to the atmosphere. This, then, is a somewhat different situation from the soil-incorporated (trifluralin) or surface-sprayed (-HCH and metolachlor) pesticides described by the other authors.
Replicate soil samples were collected from both C and G sites in 1997 by scooping the soil from an area 12.5 x 25 cm to a depth of approximately 3 cm. The average concentration of -HCH from the C site, immediately after planting (3 June), was 185 ng g-1 (dry weight). On 9 July, at the end of the sampling period, the average concentration was 85 ng g-1 (dry weight). This represents a difference of 54%. This is higher than the losses calculated above (30%) for horizontal atmospheric movement from the site. The difference may have been due to microbial degradation of -HCH (Engst et al., 1977, 1979), leaching below the soil sampling zone (Ehlers et al., 1969a,b), or to vertical atmospheric losses not accounted for by the measurements made in this study.
Regional Volatilization of -HCH
The canola-growing region of the Canadian prairies represents a large geographic area and may experience widely varying local weather. It is possible for one part to experience severe drought and another very wet conditions. The two study years represented both wet (1997) and dry (1998) situations and the data gathered may be indicative of the high and low extremes of -HCH volatilization from canola fields. When results from the study site are extrapolated to the canola-growing region of the Canadian prairies, it can be estimated (Table 3) that from 59.2 to 168.5 Mg and from 66.4 to 188.8 Mg of -HCH volatilized during 1997 and 1998, respectively, from this area. The canola-growing region of the Canadian Prairies has traditionally been north of the Regina area. Volatilization losses of 12 to 30% of the quantities of -HCH applied may be transported southward, by the prevailing northwest prairie winds, to the southern parts of the prairie region. This could account for the relatively high atmospheric concentrations of -HCH measured in the Regina area in 1994 (Waite et al., 1999).
View this table:
[in this window]
[in a new window]
|
Table 3. Calculated loss of -HCH from Canadian prairie canola fields using an average seeding rate of 6.5 kg ha-1, assuming that 95% of the canola seed was treated with -HCH and estimating losses based on the high (1997) and low (1998) values measured at the C site (corrected for background concentrations).
|
|
Dry and Wet Atmospheric Deposition of -HCH
Dry atmospheric depositions of -HCH were measured at both sites in 1998 (Table 4) using a sampler that simulates a pond surface (Waite et al., 1999). Quantities ranging from 853 to 2203 ng m-2 d-1 were measured at the C site for the first 3 wk after planting. For the final 3 wk of sampling at that site and for all 6 wk at the G site the dry deposition rates were below the quantitation level of 20 ng m-2 d-1. Combining the values above the quantitation limit with the ambient air concentrations at 1 m for the same weeks it is possible to calculate dry deposition velocities (Vd, cm s-1) using the equation:
where X (m d-1) = dry deposition (ng m-2 d-1)/ambient air at 100 cm (ng m-3), 86400 is a constant that converts days to seconds, and 100 converts meters to centimeters.
The dry deposition rates were 0.44, 0.35, and 0.38 cm s-1 for the weeks of 12–19 May, 19–26 May, and 26 May–2 June, respectively. This may be compared with the slightly lower Vd of 0.29 cm s-1 reported for -HCH at a site near Regina, SK, in 1994 (Waite et al., 1999). Eisenreich et al. (1981) reported dry deposition rates ranging from 0.04 to 1.0 cm s-1 for PCBs, with the lower rates for those compounds primarily present in the atmosphere as gases. Waite et al. (1999) reported Vd values of 1.5 and 1.2 cm s-1 for the herbicides 2,4-D [(2,4-dichlorophenoxy)acetic acid] and dicamba [3,6-dichloro-o-anisic acid], respectively, which are more susceptible than -HCH to adsorption onto particles. No dry deposition samples collected at the G site in 1998 contained quantifiable amounts of -HCH.
Concentrations of -HCH in wet deposition (rain, Table 5) were determined from samples collected using the dry–wet atmospheric deposition sampler. These concentrations were generally similar for both sites and ranged from <10 to 200 ng L-1. Concentrations <100 ng L-1 were reported from rain collected in the Regina, SK area in 1994 (Waite et al., 1999). Chevreuil et al. (1989) found average concentrations of up to 90 ng L-1 in bulk precipitation and Villeneuve and Cattini (1986) 30 ng L-1 in precipitation in France. Van Jaarsveld et al. (1997) published data on concentrations of -HCH in precipitation in Scandinavia, Europe, Russia, and the USA. They listed a range of concentrations of from 2.9 (Norway) to 43 ng L-1 (Hungary) but indicated large-scale background concentrations to be about 0.3 ng L-1. In comparison, concentrations of 0.001 to 0.936 ng L-1 were found in Bermuda (Knapp et al., 1988). The concentrations measured in Saskatchewan are relatively high but were collected only during the period when relatively high concentrations of -HCH were present in the air. Atmospheric washout of hydrocarbons by rainfall has been reported (e.g., Dickhut and Gustafson, 1995) and may have played a role in this. The relatively low rainfall during the sampling period may also have contributed to the high -HCH concentrations. High rainfall would dilute the chemical washed out of the local air mass while low rainfall might wash out the chemical without diluting it with additional rain. The annual average concentration might, then, be considerably lower than the weekly values reported here.
|
CONCLUSIONS
|
|---|
Significant quantities of -HCH volatilize from canola fields planted with treated seed. Volatilization begins immediately after planting, reaches its maximum rate during the second week, and decreases subsequently with time. The rate of volatilization may be greater in wet conditions than in dry. The quantities of -HCH released from treated canola seed to the atmosphere during the growing season contribute significantly to regional atmospheric concentrations. The importance of this source of -HCH to other North American and global ecosystems is currently under study. Atmospherically transported -HCH can enter aquatic and nontarget terrestrial ecosystems through the processes of dry and wet deposition.
|
ACKNOWLEDGMENTS
|
|---|
We gratefully acknowledge E. Cabalo and D. Chau, Environment Canada, Edmonton, AB, for laboratory analyses and L. Quinnett-Abbott, I. Kohut, and G. Gunther for field work. M. Arts, Environment Canada, Saskatoon, SK, provided weather information from the Langham, SK station. Funding for this study was provided by K. Puckett, Environment Canada, Downsview, ON, and D. Donald, Environment Canada, Regina, SK. We also thank L. Kohut and J. Worotniak for permitting us to collect samples on their farm land.
|
REFERENCES
|
|---|
- Beard, A.P., and N.C. Rawlings. 1999. Thyroid function and effects on reproduction in ewes exposed to the organochlorine pesticides lindane or pentachlorophenol (PCP) from conception. Toxicol. Environ. Health A 58:509–530.
- Brorstrom-Lunden, A., A. Lindskoog, and J. Mowrer. 1994. Concentrations and fluxes of organic compounds in the atmosphere of the Swedish west coast. Atmos. Environ. 28:3605–3615.
- Chevreuil, M., A. Chesterikoff, R. Letolle, and L. Granier. 1989. Atmospheric pollution and fallout by PCBs and organochlorine pesticides (Ile de France). Water Air Soil Pollut. 43:73–83.
- Dickhut, R.M., and K.E. Gustafson. 1995. Atmospheric washout of polycyclic aromatic hydrocarbons in the southern Chesapeake Bay region. Environ. Sci. Technol. 29:1518–1525.
- Ehlers, W., W.L. Farmer, W.F. Spencer, and J. Letey. 1969a. Lindane diffusion in soils: II. Water content, bulk density, and temperature effects. Soil Sci. Soc. Am. Proc. 33:505–508.
- Ehlers, W., J. Letey, W.F. Spencer, and W.L. Farmer. 1969b. Lindane diffusion in soils: I. Theoretical considerations and mechanism of movement. Soil Sci. Soc. Am. Proc. 33:501–504.
- Eisenreich, S.J., B.B. Looney, and J.D. Thornton. 1981. Airborne organic contaminants in the Great Lakes ecosystem. J. Am. Chem. Soc. 15:30–38.
- Engst, R., R.M. Macholz, and M. Kujawa. 1977. Recent state of lindane metabolism. Residue Rev. 68:59–90.[Medline]
- Engst, R., R.M. Macholz, and M. Kujawa. 1979. Recent state of lindane metabolism. Part II. Residue Rev. 70:71–95.
- Grover, R. 1991. Nature, transport and fate of airborne residues. p. 89–118. In R. Grover and A.J. Cessna (ed.) Environmental chemistry of herbicides. Volume II. CRC Press, Boca Raton, FL.
- Grover, R., A.E. Smith, S.R. Shewchuk, A.J. Cessna, and J.H. Hunter. 1988. Fate of trifluralin and triallate applied as a mixture to a wheat field. J. Environ. Qual. 17:543–550.[Abstract/Free Full Text]
- Hargrave, B.T., L.A. Barrie, T.F. Bidleman, and H.E. Welch. 1997. Seasonality in exchange of organochlorines between arctic air and seawater. Environ. Sci. Technol. 31:3258–3266.
- Haugen, J.-E., F. Wania, N. Ritter, and M. Schlabach. 1998. Hexachlorocyclohexanes in air in southern Norway. Temporal variation, source allocation, and temperature dependence. Environ. Sci. Technol. 32:217–224.
- Knap, A.H., K.S. Binkley, and R.S. Artz. 1988. The occurrence and distribution of trace organic compounds in Bermuda precipitation. Atmos. Environ. 22:1411–1423.
- Li, Y.F., A. McMillan, and M.T. Scholtz. 1996. Global HCH usage with 1° x 1° longitude/latitude resolution. Environ. Sci. Technol. 30:3525–3533.
- Pest Management Regulatory Agency. 1999. Special review of pest control products containing lindane. Special Review Announcement SRA99-01. Pest Management Regulatory Agency, Health Canada, Ottawa, ON, Canada.
- Prueger, J.H., J.L. Hatfield, and T.J. Sauer. 1999. Field-scale metolachlor volatilization flux estimates from broadcast and banded application methods in central Iowa. J. Environ. Qual. 28:75–81.
- Sang, S., S. Petrovic, and V. Cuddeford. 1999. Lindane—A review of toxicity and environmental fate. World Wildlife Fund Canada, Toronto, ON, Canada.
- Silvestroni, L. and S. Palleschi. 1999. Effects of organochlorine xenobiotics of human spermatozoa. Chemosphere 39:1249–1252.[Medline]
- Statistics Canada. 1999. Estimated area, yield, production, average farm price for canola-rapeseed—Canada and provinces by crop years beginning in 1943. CANSIM Matrix 3558. SDSS 3401 STC (22_022) [Online]. Available at http://www.statcan.ca (verified 4 Dec. 2000).
- Thomas, P. 1999. Canola growers manual—Grow with canola. Canola Council of Canada [Online]. Available at http://www.canola-council.org (verified 4 Dec. 2000).
- USEPA. 1980. Lindane. Position Document 2/3. Office of Pesticides and Toxic Substances, Washington, DC.
- Van Jaarsveld, J.A., W.A.J. Van Pul, and F.A.A.M. De Leeuw. 1997. Modeling transport and deposition of persistent organic pollutants in the European region. Atmos. Environ. 31:1011–1024.
- Villeneuve, J.P., and C. Cattini. 1986. Input of chlorinated hydrocarbons through dry and wet deposition to the western Mediterranean. Chemosphere 15:115–120.
- Waite, D.T., A.J. Cessna, N.P. Gurprasad, and J. Banner. 1999. A new sampler for collecting separate dry and wet atmospheric depositions of trace organic chemicals. Atmos. Environ. 33:1513–1523.
- Waymann, B., and H. Rudel. 1995. Influence of air velocity, application dose, and test area size in the volatilization of lindane. Int. J. Environ. Anal. Chem. 58:371–378.
TOP
ABSTRACT
INTRODUCTION
