As faba bean acreage started to expand in Saskatchewan, a better understanding of how weeds can best be managed was a focal point of research at the University of Saskatchewan’s Plant Science Department. From 2016 to 2021, Saskatchewan Pulse Growers funded 13 faba bean weed control research projects at Professor Chris Willenborg’s lab.

One of the first studies developed an understanding of the critical period of weed control (CPWC). This is defined as the period in crop growth where weeds must be controlled to prevent yield loss. Research took place at the Kernen Research Farm near Saskatoon, Sask., and the Western Applied Research Corp (WARC) located at Scott, Sask., from 2018 through 2021.

Twelve different weed removal timings were compared, as well as a weed-free control and a weed-infested control. CDC Snowdrop was seeded into plots that had been treated with a glyphosate + bromoxynil or carfentrazone pre-seed application.

• The CPWC was found to be from the five to nine nodes (three to seven leaves) growth stages. This means that the crop must be kept weed-free until the seven-leaf stage.
• Additionally, applying a pre-emerge residual herbicide can also reduce early-season weed competition.

Screening herbicides for tolerance
Willenborg conducted six studies that investigated tolerance to various pre- and post-emergent herbicides. In one study, he found that faba bean tolerance to pyridate, a Group 6 herbicide sold under the tradename Tough 600EC, was unacceptable when applied at the two to three-node stage.

• Pyridate at all rates tested caused significant yield losses of 17 to 36 per cent.

Another study looked at tolerance to pre-seed herbicides from 2016 and 2017 at Kernen, and at Goodale Research Farm and Kernen in 2018 through 2020. Products were applied at 1x and 2x rates and included pyroxasulfone (Zidua), sulfentrazone (Authority), pyroxasulfone plus carfentrazone (Focus), pyroxasulfone plus sulfentrazone (Authority Supreme), pyroxasulfone plus saflufenacil (Zidua + Heat), flumioxazin (Valtera), pyroxasulfone plus flumioxazin (Fierce) and saflufenacil (Heat).

Overall, most of the herbicides did not impact crop establishment, development, or yield, but there were some exceptions. In 2018, there was some initial injury from the application of Zidua + Heat, Valtera (2X rate) and Zidua + Valtera when applied pre-emergence on the sandier soils at Goodale, but no injury was observed for any products in 2019.

In 2020 at both Goodale and Kernen, Focus caused greater than 10 per cent injury for faba bean at 1x and 2x alone or when tank mixed with Valtera at pre-plant or pre-emergence. At the 2x rate, pre-plant Focus significantly reduced faba bean seed yield by 11 per cent at Kernen and by 13 per cent at Goodale compared to the hand-weeded untreated check.

• These results suggest that Focus may not be a good fit as a pre-emergent herbicide for faba bean production, particularly under low organic matter, sandy soils.

Nine post-emergent herbicides were evaluated at Kernen in 2016 and 2017 and were applied at the four-leaf stage at 1x and 2x rates. Products screened included topramezone (Armezon), bentazon (Basagran), fomesafen (Reflex), Armezon + Basagran, cloransulam-methyl, cloransulam-methyl plus bentazon, Reflex + Basagran, fluthicacet-methyl (Cadet) and fluthiacet-methyl plus Basagran. Basagran is the only registered post-emergent product on faba bean.

• Based on the overall results of this trial, none of the unregistered products would be considered as new options for post-emergent weed control in faba bean.

A tolerance screening trial to layering pre- and post-emergent herbicides was conducted in 2017 and 2018 at Scott, Kernen and the Northeast Agriculture Research Foundation at Melfort.

Six pre-emergent herbicide treatments, including pyroxasulfone (Zidua), ethafluralin (Edge), saflufenacil (Heat), flumioxazin (Valtera), pyroxasulfone + sulfentrazone (Authority Supreme) and pyroxasulfone + saflufenacil (Zidua and Heat) were applied on their own. In addition, these pre-emergent herbicides were also applied in a layered combination with imazamox/ bentazon (Viper) and imazamox/imazethapyr/sethoxydim (Odyssey). Viper and Odyssey were also each applied with no pre-emergent application.

• Pre-emergent application provided significant reductions in broadleaf weed densities prior to the post-emergent application.  
• Depending on the weed spectrum, the combined pre- plus post-emerge application did not always result in higher yields than post-emerge herbicides applied alone.  

“We believe that growers have to consider the long-term economic benefits of this layering strategy in terms of delaying herbicide resistance rather than just short-term economic returns,” says Willenborg.

Recropping trials
Willenborg also ran three recropping trials looking at the impacts of residual herbicide carryover on faba bean production. In the first trial, clopyralid (Lontrel), quinclorac (Accord), flucarbazone-sodium (Everest), pyrasulfotole + bromoxynil (Infinity), dicamba (Banvel II) and metsulfuron-methyl (Ally) were applied at different rates to spring wheat the previous year in 2016 at Kernen and Scott. Post-harvest 2,4,D was also applied as a treatment following wheat harvest. The trial was repeated in 2017 at Melfort, Scott and Kernen.

Across the three years, faba bean injury from herbicide residues was inconsistent. Most herbicides caused some herbicide injury initially under the right environmental conditions.

• Accord and Lontrel residues at higher rates are more likely to cause damage to faba bean and to reduce yield, especially under dry, low pH and low soil organic matter soil conditions.
• Growers should avoid growing faba bean on fields where these two Group 4 products have been applied the previous year.

A second recropping study looked at variety tolerance to residual herbicides. Lontrel (300 g ai/ha), Accord (200 g ai/ha), Everest (40 g ai/ha) and Ally (9 g ai/ha) were fall-applied at 2x the rate to chemfallow in 2016 and 2017 at Kernen. Three low-tannin varieties, CDC Snowdrop, Snowbird, and Tabasco and three tannin varieties, CDC SSNS-1, Malik and Taboar were grown on the herbicide-treated land the following spring in 2017 and 2018.

• There were no differences between faba bean varieties in crop establishment, biomass or yield for the herbicides assessed.
• Lontrel produced high crop injury and yield loss.
• Everest residues resulted in higher levels of crop injury than both Accord and Infinity.

A third residue study looked at faba bean tolerance to Group 2 and Group 4 herbicide residue in comparison to field pea tolerance and was conducted at Kernen in 2017. The intent was to see if field pea re-cropping restrictions could also be applied to faba bean. CDC Snowdrop faba bean and CDC Meadow field pea were grown on plots that had increasing rates of Lontrel and Everest applied the previous year.

• Injury to both crops was much greater from Lontrel than Everest.
• The response of field pea and faba bean to Everest was similar.
• Faba bean was more sensitive to Lontrel residues than field pea.

Integrated weed management improves weed control
Field studies were conducted at Kernen in 2017, 2018 and 2019, and Melfort in 2019. The study was done to see if higher seeding rates along with pre-emergent herbicides could reduce the need for a post-emergent herbicide application.

 Seeding rate was 2.5, 5, 7.5, 10, 12.5 and 15 seeds per square foot (25, 50, 75, 100 and 125, 150 seeds/m2). Herbicide applications compared were glyphosate alone pre-seed, glyphosate with Authority Supreme (PRE) and glyphosate with Authority Supreme followed by Viper ADV post-emergent (PRE+POST).

• Layering glyphosate with Authority Supreme followed by Viper ADV post-emergent generally provided the best weed control and yield at all seeding rates.
• A seeding rate of five seeds/ft2 is recommended based on this study, but if glyphosate was the only burn down herbicide applied, a seeding rate of 12.5 seeds/ft2 was needed to optimize yield.

 Another integrated weed management study investigated seeding date, seeding rate, row spacing and mechanical weed control combined with pre-emergent herbicides with differing lengths of residual control. The study was conducted at Kernen from 2017 through 2019.

Three levels of IWM were compared. The low IWM level was late seeding (May 30 to June 7), wide row spacing of 23 inches (60 cm), regular seeding rate of 4.5 plants/ft2 (45 seeds/m2) and no mechanical rotary hoeing weed control. Medium IWM included an average seeding date (May 18–24), 12-inch (30 cm) row spacing, regular seeding rate of 4.5 plants/ft2 (45 seeds/m2) and no rotary hoeing. High IWM included early seeding (May 5–10), 12-inch row spacing, high seeding rate of 9.5 plants/ft2 (95 seeds/m2) and mechanical rotary hoeing.

 The four PRE residual herbicide treatments consisted of glyphosate alone or tank-mixed with a short residual herbicide, Heat (saflufenacil), moderate residual, Valtera herbicide (flumioxazin), and season-long residual Authority Supreme (sulfentrazone + pyroxasulfone).

• The results of this study showed that all three of the soil residual herbicides improved weed control.
• Additionally, by increasing from low to medium intensities of IWM, significant improvements to both weed control and crop yields were observed.

Layering for weed control
A trial conducted at Kernen and Melfort from 2017 through 2020 assessed whether a residual pre-emergent application allows for a broader application timing window for post-emergent herbicide application.

Pre-emergent herbicide treatments included a glyphosate tank-mix with either Aim (carfentrazone), Heat or Valtera. Aim has no residual, while Heat and Valtera provide some residual control of some broadleaf weeds. Post-emergent applications included Odyssey or Viper at either the two or six-leaf stage (four and eight node stages).

• Heat and Valtera applications had lower weed biomass than Aim.
• There was no difference in weed biomass between the two and six-leaf stages.
• Layering pre- and post-emerge herbicides increased crop biomass but not yield.

Faba bean desiccation
Research was conducted to determine the best timing and herbicide treatments on dry down and faba bean yield. The study was carried out at Kernen in 2018 and 2019, and Scott from 2018 through 2020. Herbicide applications included glyphosate, Heat (saflufenacil) and glyphosate + Heat at application timings of 60, 50, 40, 30 and 20 per cent seed moisture content. Visual ratings were conducted at seven, 14 and 21 days after treatment.

• All three herbicide treatments were found to have similar dry-down rates.
• Application of harvest aid herbicides at 20 per cent moisture level had a 17 per cent greater dry-down rate compared with the application at the 60 per cent moisture level.
• There was no herbicide effect on green and black seed percentage.
• If perennial weeds are present, an application with glyphosate is recommended.

Additional research was conducted to see if there were variety interactions with desiccants between low-tannin and tannin cultivars. Three low-tannin varieties (CDC Snowdrop, Snowbird and Tabasco) and three tannin varieties were compared (CDC SSNS-1, Malik and Taboar). Malik, CDC Snowdrop and Snowbird are considered early maturing (104 days), and CDC SSNS-1, Tabasco and Taboar as late maturing (105–107 days).

Five different herbicide treatments compared were untreated, glyphosate, Heat, Reglone (diquat) and Liberty (glufosinate), and were applied when bottom pods were black at approximately 30 per cent seed moisture. Liberty is not registered for preharvest application.

• Dry-down herbicides produced no significant differences in crop yield, thousand seed weight, plant counts, black pod percentage or straw moisture content compared to the untreated check.
• Reglone and Liberty consistently provided better crop dry-down than the untreated check in the late maturing cultivars.