Runoff, persistence, and runoff modelling of bromoxynil octanoate, diclofop-methyl and atrazine from three Manitoba soils
Kenny, Dorothea F.
A study was conducted to determine the losses of residues of the herbicides bromoxynil octanoate, diclofop-methyl and atrazine from treated fields through surface runoff. Field sites used had a runoff collection system in place. The plots had a 9% slope, were 22.1 m long and 4.6 m wide. Three sites were chosen for the experiment, near Miami, Roseisle and Whitewater Manitoba on Gretna clay, Leary sandy loam and Ryerson sandy clay loam, respectively. The active ingredients bromoxynil octanoate and diclofop-methyl were applied as Hoegrass II to the wheat plots and Aatrex, with the active ingredient atrazine, was applied to the corn plots. In addition to runoff samples, soil samples were taken following runoff events to determine the amount of each chemical remaining in the soil. Runoff samples from the plots were extracted with dichloromethane, derivatized by methyrating, cleaned up on a Florisil column and analyzed by gas chromatography using either an electron capture detector or a nitrogen phosphorus detector. Total losses of these herbicides in the runoff water over a field season ranged from 0.05% to 0.83%. Highest losses of bromoxynil octanoate and atrazine in runoff were found from the Leary sandy loam. The highest losses of diclofopmethyl were from the Ryerson sandy clay loam. Soil persistence of bromoxynil octanoate, bromoxynil, diclofop-methyl, diclofop acid and atrazine were determined at each of the three sites. Soil samples were extracted with acetonitrile and analyzed using gas chromatography. The loss of the readily hydrolyzed bromoxynil octanoate and diclofop-methyl was detected shortly after application to the phenol and acid form respectively. Persistence was greater at all sites for atrazine than for either of the other applied chemicals. Higher atrazine residues in the Leary sandy loam and the Ryerson sandy clay loam were found than in the Gretna clay. Estimations of runoff losses of each of the compounds was determined using the Simulator for Water Resources in Rural Basins (SWRRB) runoff model. Although higher concentrations of pesticide loss were computed, observed trends between actual results and those through modelling were clearly visible.