Reducing weed seed production in herbicide resistant weeds with pre-harvest herbicide application
To determine efficacy of pre-harvest herbicide to reduce weed seed production in kochia and wild mustard.
Weeds with resistance to imidazolinone (Group 2) herbicides survive in-crop herbicide application, compete with the crop, and produce seeds that create future weed problems. Wiping non-selective herbicides onto surviving tall weeds above the crop could limit weed seed production. Two field experiments were conducted from 2015-2017 in central Saskatchewan, to determine the optimal timing of weed wiping (Experiment 1), and to identify effective herbicides and concentrations (Experiment 2), for control of wild mustard seed production in lentil. Experiment 1 was conducted over four site-years, and tested three 20% herbicide solutions containing dicamba, 2,4-D amine, or glyphosate, applied to wild mustard in progressively later weeks beginning at bolting. Experiment 2 was conducted at two sites in 2017, and screened 8 herbicide treatments (glyphosate, 2,4-D amine, mecoprop-P, Pixxaro A (halauxifen + fluroxypyr), Pixxaro A + glyphosate, 2,4-D + glyphosate, mecoprop + glyphosate) at 3 concentrations (12.5, 25, and 50% herbicide). Due to variable weed emergence Experiment 2 was conducted on tame mustard (Brassica juncea) and volunteer canola in addition to wild mustard.
For Experiment 1, applying dicamba, 2,4-D amine, and glyphosate to wild mustard reduced seed production by 51, 54, and 72%, respectively. Glyphosate also reduced mustard seedling emergence by 52% in the following spring, and resulted in similar or better lentil yield than no weed wiping. In contrast dicamba caused >50% yield loss, and therefore is not recommended for weed wiping in lentil. The optimal timing of weed wiping varied from year to year. In general weed wiping was most effective if it was conducted in the first or second week of wild mustard bolting. Weed wiping twice, in the first and third week of mustard flowering, was at least equally effective to wiping once at the optimal timing.
In Experiment 2 different herbicides and concentrations reduced weed seed production inconsistently, and to a lesser degree than in Experiment 1. Increasing the concentration lowered weed seed production for all herbicides on tame mustard, and five of seven herbicides on volunteer canola. Glyphosate, mecoprop, 2,4-D amine, 2,4-D amine + glyphosate, and mecoprop + glyphosate reduced mustard seed production by an average of 35%. Strong winds (approx. 50 km/hr) on the date of herbicide application may have played a role in the lower efficacy of weed wiping in Experiment 2 than Experiment 1. 2017 was an exceptionally windy year and we speculate that this may have increased herbicide loss through volatilization.