Leaching Rates and Preferential Flow of Selected Herbicides through Tilled and Untilled Soil

Abstract Pesticides can be transported to ground water more rapidly through preferential flowpaths than would be predicted from their physico‐chemical properties. The leaching rates of the herbicides 2,4‐D, bromoxynil, clopyralid, dicamba, diclofop, MCPA, and mecoprop were compared in this study on plots filled after harvest (conventional till, CT) and those that were not (fall tillage operation omitted, NT). The soil‐incorporated herbicides triallate and trifluralin were applied to the CT plot only. Herbicide was applied immediately prior to a leaching irrigation for salt removal, which represents a “worst‐case” scenario for pesticide leaching. Direct evidence of preferential flow was obtained when the herbicides, with the exception of triallate and trifluralin, were detected in the first water reaching the tile drains. Although the nonincorporated herbicides were transported preferentially at the same rate, the amounts transported depended on the solubility and adsorption coefficient ( K oc ) of the herbicide. Only 0.01% of the application of the least soluble herbicide, diclofop, was transported, compared with 0.46% of the most soluble herbicide, dicamba. Preferential flow was only slightly reduced by the tillage pass. The amounts of herbicide transported to the tile drain, however, were substantially reduced on the CT plot. The tillage effect was greatest for the more soluble and less strongly absorbed herbicides. There was no clear relationship between amounts transported in the year after application and reported persistence but herbicides with the longer half‐lives persisted in relatively greater amounts than the other herbicides.