Seasonal leaching and biodegradation of dicamba in turfgrass.

The leaching of surface-applied herbicides, such as dicamba (2methoxy-3,6-dichlorobenzoic acid), to ground water is an environmental concern. Seasonal changes in soil temperature and water content, affecting infiltration and biodegradation, may control leaching. The objectives of this study were to (i) investigate the leaching of dicamba applied to turfgrass, (ii) measure the degradation rate of dicamba in soil and thatch in the laboratory under simulated field conditions, and (iii) test the ability of the model EXPRES (containing LEACHM) to simulate the field transport and degradation processes. Four field lysimeters, packed with sandy loam soil and topped with Kentucky bluegrass (Poa pratensis L.) sod, were monitored after receiving three applications (May, September, November) of dicamba. Concentrations of dicamba greater than 1 mg L(-1) were detected in soil water. Although drying of the soil during the summer prevented deep transport, greater leaching occurred in late autumn due to increased infiltration. From the batch experiment, the degradation rate for dicamba in thatch was 5.9 to 8.4 times greater than for soil, with a calculated half-life as low as 5.5 d. Computer modeling indicated that the soil and climatic conditions would influence the effectiveness of greater degradation in thatch for reducing dicamba leaching. In general, EXPRES predictions were similar to observed concentration profiles, though peak dicamba concentrations at the 10-cm depth tended to be higher than predicted in May and November. Differences between predictions and observations are probably a result of minor inaccuracies in the water-flow simulation and the model's inability to modify degradation rates with changing climatic conditions.

Development of an enzyme-linked immunosorbent assay for the detection of dicamba.

A competitive indirect enzyme-linked immunosorbent assay (CI-ELISA) was developed to quantitate the herbicide dicamba (3,6-dichloro-2-methoxybenzoic acid) in water. The CI-ELISA has a detection limit of 2.3 microg L(-1) and a linear working range of 10--10000 microg L(-1) with an IC(50) value of 195 microg L(-1). The dicamba polyclonal antisera did not cross-react with a number of other herbicides tested but did cross-react with a dicamba metabolite, 5-hydroxydicamba, and structurally related chlorobenzoic acids. The assay was used to estimate quantitatively dicamba concentrations in water samples. Water samples were analyzed directly, and no sample preparation was required. To improve detection limits, a C(18) (reversed phase) column concentration step was devised prior to analysis, and the detection limits were increased by at least by 10-fold. After the sample preconcentration, the detection limit, IC(50), and linear working range were 0.23, 19.5, and 5-200 microg L(-1), respectively. The CI-ELISA estimations in water correlated well with those from gas chromatography-mass spectrometry (GC-MS) analysis (r(2) = 0.9991). This assay contributes to reducing laboratory costs associated with the conventional GC-MS residue analysis techniques for the quantitation of dicamba in water.

Development of an enzyme-linked immunosorbent assay for the detection of glyphosate.

A competitive indirect enzyme-linked immunosorbent assay (CI-ELISA) was developed to quantitate the herbicide glyphosate [N-(phosphonomethyl)glycine] in water. The ELISA has a detection limit of 7.6 microg mL(-1) and a linear working range of 10-1000 microg mL(-1) with an IC(50) value of 154 microg mL(-1). The glyphosate polyclonal antisera did not cross-react with a number of other herbicides tested but did cross-react with the glyphosate metabolite aminomethylphosphonic acid and a structurally related herbicide, glyphosine [(N,N-bis(phosphonomethyl)glycine]. The assay was used to estimate, quantitatively with accuracy and precision, glyphosate concentrations in water samples. Water samples were analyzed directly, and no sample preparation was required. To improve detection limits, water samples were concentrated prior to analysis, resulting in the increase of the detection limits by 100-fold. After the sample preconcentration step, the detection limit improved to 0.076 microg mL(-1) with an IC(50) value of 1.54 microg mL(-1), and a linear working range was 0.1-10 microg mL(-1). Glyphosate concentrations determined by ELISA correlated well with those determined by high-pressure liquid chromatography (r(2) = 0.99). This assay contributes to reducing the costs associated with conventional residue analysis techniques for the quantitation of glyphosate in water.

Pesticide and polychlorinated biphenyl residues in waters at the mouth of the Grand, Saugeen, and Thames Rivers, Ontario, Canada, 1986-1990.

Water samples were collected from the mouths of the three major agricultural watersheds, the Grand, the Saugeen, and the Thames (Ontario, Canada), between January 1986 and December 1990. Analyses were performed for 18 herbicides, 26 insecticides, and 4 fungicides in use in the basins. A total of between 425 and 474 samples were analyzed for each of the major groups of pesticides. Six herbicides, two insecticides and polychlorinated biphenyls (PCBs) were identified in surface water. Atrazine and its metabolite desethylatrazine were the most frequently found pesticide present in 340 of 474 samples or 72%; the metabolite was not always present with the parent compound. The second most frequently found pesticide was metolachlor which was identified in 30 of 474 samples or 6.3%. 2,4-D and cyanazine were present in 3.3% and 1.5% of the samples, respectively; alachlor, mecoprop, and simazine were present in 0.5% of the samples. Dicamba and metribuzin were present in single samples (0.2%). DDT, heptachlor epoxide, and PCB were identified in only single samples over the 5-year period. Between 342 and 2959 kg/annum of total atrazine were found passing the mouth of the three rivers and entering Lakes Erie or St. Clair between 1986 and 1990. The greatest loss was from the Thames River and the least from the Saugeen River. Between 1% and 2% of that applied in the watershed was lost at the mouth. Loadings of only two other pesticides to the rivers exceeded 5 kg in any one year, namely, metolachlor and 2,4-D. In the case of metolachlor, loadings ranged from less than 5 to 1,726 kg/annum, the highest being in the Thames and the lowest in the Saugeen River. 2,4-D exceeded a loading of 5 kg/annum in 1988 in the Grand River. Atrazine, cyanazine, and metolachlor were tracked across Lake St. Clair from the mouth of the Thames to the mouth of the Detroit River in 1987.

Triazine and chloroacetamide herbicides in Sydenham River water and municipal drinking water, Dresden, Ontario, Canada, 1981-1987.

Samples of raw river water from the Sydemham River, Ontario were collected 30 to 50 times per year between 1981 and 1987 along with paired samples of drinking water from the town of Dresden. Atrazine and its metabolite, deethyl atrazine, were found in 89 to 100% of the raw water over the seven year period. Alachlor was found only in 1982, 1984 and 1985 when 2 to 17% of raw waters were contaminated. Cancellation of the registration to use alachlor at the end of 1985 resulted in no residues being found in 1986 and 1987. Cyanazine was found in 3 to 29% (1982-87), metolachlor in 19 to 27% (1984-87) and metribuzin in 2 to 7% (1982-86) of raw river water. Comparison of those residues in raw with those in drinking water revealed that chlorination of river water had no effect in reducing herbicide concentrations. During 1985 the addition of up to 50 mg/L of powdered charcoal to raw water reduced residues to near or below detection limits for s-traizine and chloroacetamide herbicides. However, in 1986, with a reduced rate of 20 mg/L of charcoal herbicide residues were only slightly reduced and in 1987 with only 5 mg/L no reductions occurred.