| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Environment Canada, 300-2365 Albert Street, Regina, SK, Canada S4P 4K1
b Research Centre, Agriculture and Agri-Food Canada, Lethbridge, AB, Canada T1J 4B1 (present address: National Water Research Institute, 11 Innovation Boulevard, Saskatoon, SK, Canada S7N 3H5)
c Deceased (formerly at Research Station, Agriculture and Agri-Food Canada, Regina, SK, Canada)
d Retired (formerly at Research Station, Agriculture and Agri-Food Canada, Regina, SK, Canada)
e Faculty of Engineering, University of Regina, Regina, SK, Canada S4S 0A2
* Corresponding author (Don.Waite{at}ec.gc.ca).
Received for publication June 23, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
,,-trifluoro-2,6-dinitro-N,N-isopropyl-p-toluidine] were measured in the atmosphere, bulk atmospheric deposits, surface film, and dugout (pond) water at two sites near Regina, Saskatchewan, during 1989 and 1990. All five herbicides were detected in air and surface film and all but trifluralin were detected in the bulk atmospheric deposits and dugout water. Trifluralin was most frequently detected in air (79% of samples) whereas bromoxynil was present in maximum concentration (4.2 ng m–3). MCPA was present in maximum levels in bulk atmospheric (wet plus dry) deposits (2350 ng m–2 d–1), surface film (390 ng m–2), and dugout water (330 ng L–1), whereas dicamba was most frequently detected in surface film (47%) and dugout water (97%). The highest quantities of the herbicides tended to be present during or immediately after the time of regional application.
Abbreviations: PUF, polyurethane foam
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In 1990, almost 10000 Mg of these five herbicides were sold in Canada, with approximately 44% of this amount used in Saskatchewan (Environment Canada, unpublished data, 1995). Use of these herbicides, in decreasing order, was MCPA > trifluralin > diclofop > dicamba > bromoxynil. All five herbicides have been found in Saskatchewan surface waters and in the atmosphere (Grover, 1991; Grover et al., 1997a; Waite et al., 1992). Bromoxynil, dicamba, and diclofop have also been reported in bulk atmospheric deposition samples (Waite et al., 1995). More recent pesticide use data are not available so it is not possible to determine whether use of these herbicides has increased or decreased since 1990. However, all are currently still in common use on the Canadian Prairies and are consistently detected in atmospheric samples (unpublished data, 2002, 2003).
Movement of pesticides from the sites of application to nontarget regions represents an economic loss to farmers because the chemicals are transported away from cropland and the species they are designed to control. It also represents a threat to nontarget areas such as wetlands and other water bodies, such as farm dugouts, which, in Saskatchewan, serve as nesting, rearing, and resting areas for major North American waterfowl populations and related species. Farm dugouts (ponds) are common on agricultural lands of the Canadian Prairies where more than 100000 have been constructed since 1935 (Prairie Farm Rehabilitation Administration, 1995). Typically, these dugouts receive water from snowmelt and rainfall runoff, or infiltration from shallow (surficial) aquifers. These water bodies are generally slightly alkaline (pH = approximately 8) with concentrations of dissolved organic carbon ranging from 4 to 40 mg L–1 (Sketchell et al., 1993). The hardness of dugout waters is low, usually ranging from 50 to 500 mg L–1 equivalents of calcium carbonate. Generally ranging in volume from 2 to 5 million liters, they provide a valuable and often sole source of water for on-farm activities, which may include potable and household water use, livestock watering, and irrigation. Transport of pesticides into dugouts may pose a health risk to families consuming this water. Grover et al. (1997a) reported the presence of several herbicides, including the five discussed in the present paper, in Saskatchewan dugouts. In some cases, herbicides not present in the dugout water following spring snowmelt runoff were detected in mid-summer, thus implicating atmospheric transport as a possible vector of contamination.
The objectives of the study were to determine the quantities of these five herbicides in the water and surface film of two dugouts and in the associated ambient air and bulk (wet plus dry) atmospheric deposits, and to investigate possible links between atmospheric concentrations and bulk deposits and their relationships with quantities in surface film and water in relation to the physicochemical properties (Table 1) of the herbicides.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Samples
Weekly ambient air, bulk deposition, water, and surface film samples were collected at both dugouts in 1989 from the week commencing 13 May to the week ending 30 August. In 1990, samples were collected only at Dugout 1 from the week commencing 7 May to that ending 10 July and from the week commencing 7 August to the week ending 18 September.
Ambient Air
Sample Collection
Duplicate weekly air samples were collected using high-volume air samplers (Model PS-1; General Metals Works, Village of Cleves, OH) designed to collect airborne organic chemicals at 1 m above the ground. The air flow through the samplers (8.33 m3 h–1) was calibrated in the field using a Sierra-Misco (Berkeley, CA) orifice head and was monitored daily during operation by integral Venturi/Magnahelic assemblies. The samplers were operated for a 12-h period daily (0500–1700 h) for 7 d each week for a total air flow of about 700 m3 wk–1.
The sampling unit in the high-volume PS-1 samplers consisted of a 102-mm-diameter borosilicate microfiber filter (Micro Filtration Systems, Dublin, CA) in front of a polyurethane foam (PUF)/XAD-2 resin composite cartridge. The cartridge contained 25 mL of XAD-2 adsorbent resin (Caledon Laboratories Ltd., Georgetown, ON, Canada) sandwiched between 50-mm (upstream) and 25-mm (downstream) foam (polyester-type PUF) plugs. The PUF/XAD-2 resin sandwich was housed in a 60-mm-i.d. by 125-mm-long glass cylinder with a stainless steel wire mesh support screen. The filter and the composite cartridge collected particulate and vapor components of the air, respectively. Before assembly of the sampling cartridges for field use, the PUF plugs and the XAD-2 adsorbent were solvent-cleaned by Soxhlet extraction and monitored for background interference.
In a subsequent study (Grover et al., unpublished data, 1995), fortified PUF/XAD-2 resin cartridges were continuously aspirated at a flow rate of 12 m3 h–1 for 7 d to determine possible breakthrough of several herbicides including bromoxynil, dicamba, diclofop, MCPA, and trifluralin. The total volume of air that passed through the PUF/XAD-2 resin cartridge was 2016 m3 or approximately three times the volume collected in the present study. Under these conditions, there was 40 and 18% breakthrough of bromoxynil and trifluralin, respectively, and no breakthrough of dicamba, MCPA, or diclofop. Thus, in the present study, breakthrough may have resulted in some underestimation of the vapor concentrations of bromoxynil and trifluralin in the atmosphere.
Sample Extraction
The glass fiber filters and PUF/XAD-2 resin cartridges were extracted separately. The filters were Soxhlet-extracted with acetone (300 mL) for 6 h and the assembled PUF/XAD-2 resin cartridges with acetone (700 mL) for 16 h. The individual acetone extracts were concentrated to <10 mL using a rotary evaporator, and transferred to 50-mL graduated centrifuge tubes along with four 4-mL acetone rinses of the rotary evaporator flask. The individual extracts were further concentrated to <1 mL with a stream of nitrogen and then methylated with ethereal diazomethane. The methylated extracts were subjected to Florisil (4 mL, deactivated with 5% water) column cleanup and the column was eluted with 0.5% acetone in hexane. The eluate (9–30 mL) was concentrated to less than 1 mL using a stream of dry nitrogen gas and then taken to volume (1 mL) with hexane before gas chromatographic analysis.
Quality Assurance–Quality Control
The glass fiber filters were fortified with 10 µg of each herbicide by applying an acetone solution of the herbicides over the surface of the filters in two 0.5-mL portions. The filters were placed in a fume hood for 15 min after each application to permit evaporation of the acetone. The upstream PUF plug of the PUF/XAD-2 resin cartridge was similarly fortified with 10 or 1 µg of each herbicide by applying 1 mL of acetone solution of the herbicides to the upstream surface of the foam plug and then placing the cartridge in a fume hood for 1 h to permit evaporation of the acetone. The amount of each herbicide applied to the filter or the cartridge was equivalent to a 700-m3 sample of air with a concentration of 14.3 and 1.4 ng m–3 of each herbicide for the 10- and 1-µg fortification levels, respectively. Recoveries (%) are shown in Table 2. The minimum quantification level was taken to be 40 pg m–3 of air.
|
Bulk Deposits
Sample Collection
Bulk atmospheric deposition samples were collected over 7-d periods using duplicate pan samplers, as described by Waite et al. (1995). Each sampler consisted of a 1- by 1-m sheet metal (galvanized steel) pan, with 15-cm-high sides and a centrally located drain hole, supported on a wooden enclosure, which elevated the sampler surface about 60 cm above ground level. During rainfall events, the pans drained directly into 22.7-L glass carboys, carrying the rainfall sample as well as dry surface deposits that may have accumulated during the sampling period. The glass carboys were wrapped with aluminum foil to eliminate photolysis losses. Rain gauges were located beside the samplers and precipitation was recorded weekly. When there was <4 mm of rainfall during the sampling period, the pan was rinsed with 4 L of distilled water to obtain the bulk deposition sample for analysis. When rainfall exceeded 4 mm during the sampling period, the total bulk deposition sample was collected from the carboy, without rinsing the collector pan, and the volume recorded.
Sample Extraction
After thawing and shaking to suspend any sediment, bulk deposition samples (unfiltered 500-mL aliquots) were analyzed as described previously (Cessna et al., 1985).
Trifluralin.
The water sample was adjusted to pH = 12 with 20% NaOH solution (1 mL) and extracted twice with hexane (50 mL). The hexane extracts were passed through anhydrous sodium sulfate (15 mL), followed by a 15-mL hexane wash of the sodium sulfate. The combined hexane extracts were concentrated to approximately 1 mL using a rotary evaporator and then subjected to Florisil column cleanup and concentrated to 1 mL (as described above) before gas chromatographic analysis.
Bromoxynil, Dicamba, Diclofop, and MCPA.
The aqueous phase was acidified to a pH of <2 by addition of concentrated H2SO4 (1 mL). Sodium chloride (100 g) was added and the mixture was extracted twice with diethyl ether (50 mL). The combined ether extracts were passed through anhydrous sodium sulfate (15 mL) and, together with a 25-mL diethyl ether wash of the sodium sulfate, concentrated just to dryness using a rotary evaporator. Acetone (25 mL) was added and the solution evaporated just to dryness to azeotropically remove any residual water. The residue was dissolved in diethyl ether, methylated with ethereal diazomethane, and subjected to Florisil column cleanup and concentrated to 1 mL (as described above) before gas chromatographic analysis.
Quality Assurance–Quality Control
Bulk deposition samples were fortified with each herbicide at 1.0 and 0.1 µg L–1 by the addition of 1.0 and 0.1 µg of the herbicides dissolved in 1 mL of methanol. Unfortified samples were analyzed to determine background concentrations. Recoveries (%) are shown in Table 2. Minimum quantification levels were taken as one-half of the lowest fortification level, that is, 50 ng L–1. In the case of the bulk deposition samples, 50 ng L–1 was equivalent to a deposition rate of 200 ng m–2 or 29 ng m–2 d–1 when the sample was obtained during a week without rain and the sample consisted of a 4-L rinse of the sampler. Sample volumes of >4 L would result in the minimum deposition rate being greater than 29 ng m–2 d–1.
Bulk deposition samples were extracted in batches of eight along with one fortified sample and one solvent blank. Bulk deposition samples for fortification were collected from a pan sampler located on the Agriculture and Agri-Food Canada Research Station near Regina during one single precipitation event.
Dugout Water
Sample Collection
Water samples were collected from a boat directly into new 4.5-L amber glass bottles by plunging (to minimize collection of surface film) the bottle to a depth of 0.25 m, where it was held, and then holding it while it filled with 1 to 2 L of water. The bottles were sealed with aluminium foil, capped, and stored on their sides at –20°C until analysis.
Sample Extraction and Quality Assurance–Quality Control
Sample extraction and quality assurance–quality control procedures were as described for the bulk deposits.
Surface Film
Sample Collection
Weekly sampling was performed from a boat positioned centrally in each dugout. Surface film was sampled, following the method of Muir et al. (1991) and Waite et al. (2000), using a 20- by 20-cm Pyrex glass plate with a handle fastened to one side by means of silicon caulking. The lower plate surface was touched, horizontally, onto the surface of the water and then lifted and held vertically, one corner down, over a glass funnel placed in a 500-mL, wide mouth, amber glass bottle. Using a wash bottle, the glass plate was rinsed, into the funnel, with approximately 25 mL of dichloromethane (CH2Cl2). Five such subsamples were collected for each composite sample. The bottles were sealed with aluminium foil, capped, placed in a cooler, and returned to the laboratory where they were stored at –20°C until analysis. Two composite samples were collected, for each sampling day, from each dugout.
In 1990, two sets of duplicate surface film samples were collected from Dugout 1. One set was collected as described above and the second set by using two rafts to stabilize the dugout surface and quench wave action. The rafts were constructed of 55 by 100 by 5 cm (thick) rigid foam insulation board in which five holes, 25 by 25 cm, were cut. The walls of the holes were lined with glass but were left open to the air and the water. Surface film samples were collected, as described above, by dipping the glass sampling plate sequentially into each hole. The rafts were anchored in the center of the dugout.
Sample Extraction
Extraction methods for surface film samples have been described in detail by Waite et al. (2000) and will be summarized here. Surface film samples consisted of approximately 125 mL of CH2Cl2 and 35 mL of water.
Trifluralin.
Each sample was transferred to a 250-mL separatory funnel and the water volume brought to 100 mL with deionized (nanopure) water. After adjustment of the pH to 13 to 14 by the addition of 20% sodium hydroxide solution (1 mL), an additional 25 mL of CH2Cl2 was added, the flask shaken for 2 min, and the CH2CL2 layer passed through anhydrous sodium sulfate (30 mL) to remove residual water. The aqueous phase was extracted with a second portion of CH2Cl2 (25 mL) and the CH2Cl2 layer also passed through the sodium sulfate. The combined CH2Cl2 extracts, together with a 25-mL CH2Cl2 rinse of the sodium sulfate, were concentrated to approximately 2 mL using a rotary evaporator, transferred to a 50-mL graduated centrifuge tube along with four 2-mL rinses of the evaporating flask, and then concentrated to approximately 2 mL using a gentle stream of dry nitrogen gas. Isooctane (2 mL) was added and the volume again reduced to 2 mL. This step was repeated two more times. The isooctane residue was then subjected to Florisil column cleanup and the column eluate concentrated (as described above) before gas chromatographic analysis.
Bromoxynil, Dicamba, Diclofop, and MCPA.
The pH of the aqueous phase was adjusted to <2 by addition of concentrated H2SO4 (1 mL). Sodium chloride (20 g) was added and the acidified aqueous phase was extracted twice with diethyl ether (50 mL). The ether extracts were passed through anhydrous sodium sulfate (15 mL) and, together with a 25-mL diethyl ether rinse of the sodium sulfate, concentrated just to dryness using a rotary evaporator. Acetone (25 mL) was added and the solution evaporated just to dryness to azeotropically remove any residual water. The residue was dissolved in diethyl ether, methylated with ethereal diazomethane, and subjected to Florisil column cleanup, and the column eluate was concentrated to 1 mL before gas chromatographic analysis as described above.
Quality Assurance–Quality Control
Surrogate surface film samples (approximately 25 mL) were fortified with a mixture of 0.5 µg of each herbicide dissolved in 100 µL of acetone (equivalent to 2500 ng m–2) and then extracted as described above. The surrogate surface film was natural surface film collected from a local dugout using the same "glass plate" technique used in the sampling program. Herbicide concentrations present in the surrogate samples were subtracted before calculating recoveries (Table 2). The minimum quantification level was taken as 5 ng per sample for surface film samples. The area measured by a surface film sample was 2000 cm2 (five subsamples of 20 by 20 cm each), so the minimum quantification level could also be expressed as 25 ng m–2.
Gas Chromatography
Gas chromatographic analysis of all extracts was performed using a Hewlett-Packard (Palo Alto, CA) Model 5890 gas chromatograph (GC) equipped with a 63Ni electron capture detector. The gas chromatographic system was operated using the GC column and operating conditions described previously (Grover et al., 1997a).
All herbicide residue detections were confirmed using a Hewlett-Packard Model 5890 GC interfaced to the Model 5970 mass selective detector (MSD) operated in the selected ion monitoring mode. The GC–MSD system used the same GC column and operating conditions used above and the capillary interface between the GC and MSD was maintained at 280°C throughout each analysis. Before analysis, sample extracts were concentrated to 100 µL.
Farmer Survey
In each year, farmers within a 10-km2 area around Dugout 1 and a 2.6-km2 area around Dugout 2 were questioned to determine herbicides applied within those areas.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Calculation of Mean Concentrations and Quantities
For the purpose of calculating mean herbicide concentrations in dugout water, bulk deposition, and high-volume air samples and mean quantities in surface film samples, nondetections were given a zero value and were included in the total number of samples when calculating the mean. Confirmed detections that were less than the corresponding limit of quantification were assigned values equal to one-half the respective limit of quantification. The limits of quantification were such that the signal to noise ratios for all five herbicides were greater than three. None of the values greater than or equal to their respective limit of detection was corrected for recoveries.
Ambient Air
Factors Affecting Herbicide Inputs to the Atmosphere
The presence of pesticides in the atmosphere, as a result of application drift, post-application vapor losses, and wind erosion of soil, depends in large part on the magnitude of their vapor pressures, Henry's law constants, and field half-lives (indication of resistance to photolysis, microbial degradation, and hydrolysis). Diclofop and bromoxynil are commercially formulated as esters, MCPA as an ester or an amine salt, and dicamba as an amine salt. These herbicides may be present in environmental matrices as an ester or amine salt, or as the free acid/phenol following hydrolysis of the ester or amine salt. Thus, the partitioning of these four herbicides between the atmospheric and aquatic components of the environment is determined not only by the physicochemical properties of the esters, salts, and acids, but also on the rates at which the esters and amine salts are hydrolyzed to the respective acids. At the time that this study was performed, bromoxynil was commercially formulated as the octanoate or as a mixture of the octanoate plus butyrate; MCPA as mixed butyl esters, the iso-octyl ester, or dimethylamine salt; diclofop as the methyl ester; and dicamba as the dimethylamine salt. The variability of the dissipation of the various esters and amine salts of these herbicides from plant and soil surfaces and of trifluralin from soil is summarized in Table 3. In addition, each of the applied herbicides and their respective hydrolysis products are subject to different rates of photolytic and microbial degradation. Thus, the relative abundance of each of these herbicides in the atmosphere and the chemical form in which it is present is the expression of complex interactions of several environmental fate processes, as well as the relative amounts of the herbicides used on a regional basis.
|
|
|
|
|
|
|
bromoxynil > diclofop > MCPA (Table 3), indicating that a portion of each of these herbicides, with exception of MCPA, was present in the atmosphere associated with particles. The particulate fraction of dicamba, bromoxynil, and diclofop may have resulted from desiccation of spray droplets or atmospheric transport of wind-eroded soil from treated fields but may be underestimated because high-volume air samplers used in this study tend to undersample atmospheric particles (Waite et al., 2002a). Pesticides associated with atmospheric particles have higher atmospheric deposition velocities than those in the vapor phase (Waite et al., 2002b) and this is a contributing factor to their effect on aquatic ecosystems. The mean and maximum concentrations for the acidic herbicides reported here were less than concentrations of 2,4-D measured in the same study (Waite et al., 2002a) but greater proportions of dicamba and bromoxynil were associated with particles. Atmospheric concentrations of dicamba have not previously been reported from the Canadian Prairies. In 1989, bromoxynil and MCPA were applied in the areas around both dugouts, whereas only MCPA was applied in the vicinity of Dugout 1 in 1990. This may account for the detection of lower concentrations of bromoxynil in 1990 (Fig. 3), but the herbicide was still used regionally and these lower concentrations may represent regional background levels. Interestingly, atmospheric concentrations of MCPA were generally much lower than those of the other three herbicides even though MCPA was applied around both dugouts in 1989 and around Dugout 1 in 1990.
Bulk Deposits
The glass carboys in which the bulk atmospheric deposits were collected were wrapped with aluminium foil and located under the sampler, which shaded the sample from direct sunlight. The temperature of the bulk deposit samples was neither monitored nor controlled and it is possible that some biodegradation or vapor loss of the herbicides occurred during the 7-d sampling period. Vapor loss may also have occurred from dry deposits on the surface of the pan sampler. Thus, bulk deposition rates reported here may be somewhat underestimated. In sampling periods with no rainfall, bulk deposition samples would consist solely of dry (particulate) deposition.
Trifluralin was not detected in any bulk deposition from either dugout or in either year (Table 5) even though it was more frequently present in the atmosphere and in larger quantities than the other four herbicides (Table 3). This is in contrast to triallate, a herbicide with similar vapor pressure, which was found in 44 to 54% of bulk deposition samples in the same study (Waite et al., 2002a). This may be because triallate is more water-soluble than trifluralin [4 mg L–1 vs. 0.3 mg L–1, respectively (Tomlin, 2000; Table 1)] and may be more likely to partition into falling rain.
|
The herbicides were, in general, present in bulk deposits from mid-May to mid-July, the period of local application (Fig. 4 and 5). Bulk deposition during this period occasionally coincided with the presence of each herbicide in the atmosphere or with high rainfall, but these relationships, as noted for triallate and 2,4-D during the same study (Waite et al., 2002a), were not consistent. The possibility that a portion of the bulk deposits of each herbicide may have been transported into the study area from remote regions when local contributions were low may account for this inconsistency. For example, rainfall during the spring application period may have deposited chemicals transported from other regions as well as those scavenged from the immediate vicinity. The interaction of long-range transport versus the scavenging of local chemicals by rainfall has been discussed by Barrie (1992) and appears to explain, at least in part, the data presented in this current paper.
It should also be noted that, although atmospheric particles greater than 50 µm in diameter are not sampled efficiently by the high-volume air samplers used in this study, they may deposit to bulk deposition samplers. Herbicides adsorbed to surface soil may enter the atmosphere attached to wind-eroded soil particles and be collected in the bulk samplers. Thus, a herbicide may be present in a bulk deposition sample but not in the coinciding high-volume atmospheric sample. In the current study, this may explain, in part, why there appeared to be no correlation between atmospheric concentrations and bulk deposition rates of herbicides.
Surface Film
In contrast to atmospheric concentrations and bulk deposits, which represent average values for 7-d sampling periods, surface film samples are representative only of amounts of herbicide present at the end of the sampling period. This is because, even though atmospheric deposition into surface film may have occurred throughout the previous 7-d period, dissipation from the surface film may have occurred through partitioning into the water column, photolytic or microbial degradation, and vapor loss to the atmosphere. With the exception of trifluralin, the mean quantities of the other four herbicides in surface film were similar (Table 6).
|
The frequency of detection of the acidic herbicides in surface film, in decreasing order, was dicamba > diclofop > bromoxynil > MCPA (Table 6). Detected in 47% of the samples, only trace amounts of dicamba were detected in surface film at both dugouts and these detections occurred during the spring spraying season. During the same period in 1990, quantifiable amounts coincided, in part, with bulk atmospheric depositions; however, the correlation was not significant, as was also noted for 2,4-D in the same study (Waite et al., 2002a). Bromoxynil, diclofop, and MCPA were also detected primarily during the spring application period and their occurrence in surface film was also not well correlated with rates of bulk deposition. The highest quantity of any herbicide detected in surface film, 390 ng m–2 of MCPA, was measured in 1990 and occurred when bulk deposition of MCPA was also high (830 ng m–2 d–1) the preceding week. These high values may reflect drift from the application of MCPA that occurred in mid-June in the vicinity of the dugout.
Dugout Water
Trifluralin was not detected in water samples from either dugout in either year. This may indicate that amounts deposited into surface film volatilized back into the atmosphere rather than partitioning into the dugout water.
The acidic herbicides were detected in the water samples from both dugouts in both years in the following decreasing order: dicamba > MCPA > diclofop > bromoxynil (Table 7). Detected in 97% of samples, the appearance of dicamba in early spring may represent transport during the spring snowmelt runoff. Waite et al. (1992) reported dicamba concentrations as high as 410 ng L–1 in spring runoff from snowmelt and Grover et al. (1997a) attributed its presence in dugout water to spring runoff. In Dugout 1 in 1989, dicamba concentrations were highest in the spring and early summer and decreased to trace concentrations thereafter (Fig. 4) and this temporal pattern is similar to that of 2,4-D for the same study (Waite et al., 2002a). Only trace concentrations of dicamba were detected in the recently dredged Dugout 2.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Possible correlation between atmospheric bulk deposition of herbicides and their presence in surface film may be obscured by loss from surface film by volatilization, photolytic and biological degradation, or by partitioning into the underlying dugout water. The same factors could similarly obscure any correlation between atmospheric bulk deposition and herbicide concentrations in the dugout water. This correlation could be further confounded by release of herbicides from underlying sediments or partitioning of herbicides into sediments (not sampled in the present study). In addition, both surface film and dugout water were sampled once per week and, because of the above-discussed processes, may not represent the quantities of herbicides originally deposited to them.
Finally, a further factor affecting possible correlations among ambient air herbicide concentrations, bulk deposition rates, and concentrations in surface film and dugout water is the dependence of the environmental movement of the herbicides and their persistence in the various environmental compartments (air, surface film and water) on their physicochemical properties. For the acidic herbicides, this is confounded by the rates at which their esters and amine salts are hydrolyzed to the respective acids because the acids have significant differences in physical properties such as vapor pressure and water solubility (Table 1), or chemical properties such as susceptibility to photolytic and microbial degradation. Perhaps the significance of these changes and their relation to studies such as the present one might be better understood if the chemical analysis of field samples for acidic herbicides was expanded to include both esters and parent acids.
Although well-established correlations among ambient air herbicide concentrations, bulk deposition rates, and concentrations in surface film and dugout water were not evident in this study, the following observations about the five herbicides can be made.
Trifluralin
Although not applied locally, trifluralin was the most frequently detected herbicide in atmospheric samples at both dugouts in both years and its presence probably represents atmospheric transport from the more northern prairie agricultural region. Its detection mainly in the PUF/XAD-2 resin cartridges (present in only one filter sample) suggests that it was in the atmosphere primarily as a gas (Fig. 1). This, together with its relatively high vapor pressure and low water solubility, may explain why it was not detected in any bulk atmospheric deposition or water samples and only in one surface film sample. Since trifluralin is readily photolyzed in air and water (Grover et al., 1997b), photodegradation may have also contributed to its absence in these media.
Dicamba
Dicamba was also primarily present in atmospheric samples as a gas (70% of PUF/XAD-2 resin cartridges versus 17% of filters; Table 3) and tended to be detected most frequently during its spring application period. It was also detected in bulk atmospheric deposits, surface film, and dugout water samples. In a 4-yr study, Waite et al. (1995) reported the presence of dicamba in 58% of bulk atmospheric deposits in a small Saskatchewan watershed with a maximum daily deposition rate of 1900 ng m–2 d–1, comparable with that (1800 ng m–2 d–1) reported in the present study. Hill et al. (2002) also reported dicamba in bulk deposits from Alberta but the data were event based, with sample collection occurring at the end of each rainfall rather than weekly as was done in the present study, so results could be not compared. Dicamba has also been found in Saskatchewan surface waters and farm dugouts with wide ranges in concentrations and frequencies of detection (Waite et al., 1992; Grover et al., 1997a; Cessna and Elliott, 2004).
Bromoxynil
The proportion of bromoxynil present in the gas and particulate phases was similar to that of dicamba (66% of PUF/XAD-2 resin cartridges versus 19% of filters; Table 3) and it also was detected in bulk deposition, surface film, and dugout water samples. In a 4-yr study, Rawn et al. (1999) reported maximum atmospheric concentrations from a small watershed in southern Manitoba of 1.6 ng m–3 (vapor) and 0.44 ng m–3 (particulate), comparable with values in the present study. In the same study, bromoxynil was also found in rainfall in concentrations as high as 200 ng L–1 and the periodicity of occurrences in atmospheric samples and rainfall seemed to follow local use patterns, as was observed in the present study. Hill et al. (2002) also reported bromoxynil concentrations in bulk atmospheric deposits from Alberta. In their watershed study, Waite et al. (1995) reported maximum bulk deposition rates of 2500 ng m–2 d–1 compared with 700 ng m–2 d–1 in the present study.
Diclofop
Although less frequently detected in atmospheric samples (15% of PUF/XAD-2 resin cartridges versus 2% of filters; Table 3), most likely because of the relatively low vapor pressures of diclofop and its methyl ester (Table 1), diclofop was present in 27% of bulk atmospheric deposition samples with a maximum deposition rate of 320 ng m–2 d–1. It was previously detected by Waite et al. (1995) in 53% of bulk deposition samples from a small Saskatchewan watershed with a maximum bulk deposition rate of 1300 ng m–2 d–1. Diclofop was also found, in the present study, in surface film and dugout water samples. The maximum concentration in water was 36 ng L–1, which is much lower than that (4880 ng L–1) reported by Waite et al. (1992) from their watershed study and 3600 ng L–1 reported by Grover et al. (1997a) from Saskatchewan dugouts. The lower frequencies of occurrence and lower maximum concentrations in water may reflect decreased usage of diclofop.
MCPA
MCPA was seldom detected in atmospheric samples (4%; Table 3) even though it was applied locally around both dugouts in 1989 and Dugout 1 in 1990. The lower detection frequency than that of diclofop may reflect not only its relatively low vapor pressure and that of its iso-octyl ester (Table 1), but also its shorter persistence on plant and soil surfaces (Table 4). In their 4-yr Manitoba study, Rawn et al. (1999) reported MCPA in atmospheric samples primarily in periods of local use. The maximum concentration detected in that study was 13 ng m–3 compared with 0.39 ng m–3 in the present study. MCPA was, however, found in 24% of the bulk deposition samples with the highest maximum and mean concentrations (Table 5) of the five herbicides reported here. Rawn et al. (1999) reported MCPA in rain samples and Hill et al. (2002) detected MCPA in bulk atmospheric deposition samples from Alberta. Unfortunately, because of different methods of reporting, the data from these two studies cannot be directly compared with those of the present study.
MCPA was also present in surface film and dugout water samples in higher maximum quantities or concentrations than those of the other herbicides presented here (Tables 6 and 7). Grover et al. (1997a) reported concentrations of MCPA in 21 Saskatchewan dugouts ranging from not detected to 1970 ng L–1, with confirmed trace concentrations in 35 to 70% of the seasonal samples collected. The concentrations and frequency of detection reported in the present study are somewhat lower than those of Grover et al. (1997a).
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-1,2,3,4,5,6-hexachlorocyclohexane. Anal. Chim. Acta 467:245–252.Related articles in JEQ:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Agronomy Journal | Crop Science | |||
| Vadose Zone Journal | Journal of Plant Registrations | ||||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal |
