The presence of volunteer canola is becoming a significant agro-ecological concern, given the large-scale use of herbicide-tolerant varieties in some areas. Our goal was to estimate the frequency and persistence of volunteer canola in Québec cropping systems by surveying fields that included a single canola crop since 1995. A survey was conducted in 131 fields in the main canola-growing areas of Québec: in the Saguenay-Lac Saint-Jean region and the Québec City–La Pocatière area. Volunteer canola plants were counted in 0.25-m2 quadrats every 10 m along a W pattern, and every 15 m along the margins of 88 fields. Volunteer canola plants were found in 90% of the fields surveyed and in a wide range of crops, including cereal, corn, and soybean. Average densities of 4.9 and 3.9 plants/m2 were found 1 yr after canola production in fields and field margins, respectively. Volunteer canola densities decreased significantly over time. However, volunteer plants were still present at low densities 4 and 5 yr after production. Dense stands of volunteer canola were found before postemergence herbicide application in no-till fields (9.8 ± 4.1 plants/m2), suggesting that, contrary to what was suggested in the literature, seeds could become dormant in no-till as well as in tilled systems. A small proportion of the volunteer canola plants observed in no-till fields near Québec City and Ottawa included plants that had overwintered, either originating from fall-germinated seedlings, harvested adult plants that had grown new leaves before the onset of winter, or spring regrowth from the base of unharvested adult plants from experimental plots. The presence and persistence of low densities of volunteer canola may not have been a cause of concern until now. However, producers should be made more aware of the potential short-and long-term problems associated with potential gene flow between different herbicide-tolerant canola (HT canola) varieties and also between HT canola and related weed species.

Herbicide-resistant canola dominates the canola market in Canada. A multiyear field experiment was conducted at three locations to investigate the effect of time of weed removal (two-, four-, or six-leaf canola) and herbicide rate (50 or 100% recommended) in three herbicide-resistant canola systems. Weeds were controlled in glufosinate-resistant canola (GLU) with glufosinate, in glyphosate-resistant canola (GLY) with glyphosate, and in imidazolinone-resistant canola (IMI) with a 50:50 mixture of imazamox and imazethapyr. Canola yields were similar among the three canola cultivar–herbicide systems. Yields were not influenced by 50 vs. 100% herbicide rates. Timing of weed removal had the greatest effect on canola yield, with weed removal at the four-leaf stage giving the highest yields in most cases. Percent dockage was often greater for GLU and IMI than for GLY. In comparison with the other treatments, dockage levels doubled for GLU after application at 50% herbicide rates. The consistency of monocot weed control was usually greater for GLY than for GLU or IMI systems. However, weed biomass data revealed no differences in dicot weed control consistency between IMI and GLY systems. Greater dockage and weed biomass variability after weed removal at the six-leaf stage or after low herbicide rates suggests higher weed seed production, which could constrain the adoption of integrated weed management practices in subsequent years.