Herbicide loss in runoff: effects of herbicide properties, slope, and rainfall intensity

Abstract

Runoff potential of 5 herbicides (acetochlor, atrazine, hexazinone, pendimethalin, and terbuthylazine) was evaluated in a small-scale study under simulated rainfall on a cultivated Hamilton clay loam soil. At 24 h after herbicide application, rainfall events of different intensities were simulated to 0.5-m² field plots with 20% (70, 88, and 111 mm/h) and 30% (60, 70, and 80 mm/h) slope, respectively. The objective of this study was to compare the behaviour of pesticides covering a range of properties under identical hydrodynamic conditions. Sediment amounts and herbicide concentrations were determined in the runoff samples. As the transported sediment amounts were not sufficient for chemical analyses, herbicide residues attached to sediment were estimated using K_d values determined locally for the soil. Whereas pendimethalin concentrations followed no noticeable pattern, the concentrations for the other herbicides were highest in the first runoff samples, and decreased exponentially with further rain. Results show that herbicides were primarily transported in their dissolved form with the exception of pendimethalin. Slope affected cumulative runoff, sediment, and herbicide losses significantly (P < 0.05). The impact of increased rainfall intensity on runoff initiation followed a similar trend, but herbicide losses from plots exposed to different intensities were not always significant. Losses dissolved in runoff from plots with 20% slope were ≤1% of the applied herbicide, whereas on plots with 30% slope the maximum recorded loss was 65%. Here, losses for all herbicides ranged between 1 and 7% at 60 mm/h and 8 and 65% at 80 mm/h. Exports of herbicides with moderate solubility were negatively correlated with their K_d values and their water solubility.