Agriculture is an industry of ingenuity.
There are many factors that can impact crops, like frost, high winds, the amount of rainfall, etc. Weed control is one of the main ways farmers can give themselves the best chance for a successful season.
Herbicides are one option to limit the spread of unwanted plants in the field, but researchers are looking to add more tools to the box – and one in particular is looking promising.
Robert E. Nurse is a research scientist in the Science and Technology branch of the Government of Canada’s Agriculture and Agri-Food department. He’s part of a team that’s in their fourth year of studying the efficacy of high projectile weed control.
“Essentially, the goal here is we want to reduce the reliance on herbicides,” Nurse said.
“By using high velocity projectiles, we’re hoping to either get the plants at very young age – if they’re really, really small, we’re hoping that it’ll kill them completely – or if they’re larger, like say, more than five to 10 centimeters, we’re hoping that we’ll hit and affect the growing point, and then we can come in with an herbicide and take out the rest of the plant.”
There’s also the possibility that they could damage the leaves of the unwanted plants, “and with that damage, it could introduce disease into the weed that could take them out as well. So, we’re looking at a lot of different factors.”
For the experiment – which is taking place in different areas and with different crops across Canada – they chose to work with three mediums for the proof of concept stage: corn grit, walnut shells, and baking soda. These were chosen because they’re all commercially available to some extent.
In their testing, the researchers are using a compressor and sandblasting equipment.
“Ultimately, if it’s successful, it would be commercialized into something like a modified sprayer, (where) you could do it using a tractor,” Nurse said.
Initially, they started their testing by using combinations of the different mediums.
However, “we found that if we tried to combine them, the spray gun would clog up. So, we’ve found over the course of the trial that instead of trying to do them as mixes, it’s better to do them sequentially … either on the same day, or a few days apart.”
That wasn’t the only discovery when it came to testing the different mediums.
All things being relative, the researchers chose the different mediums because it would offer a variety of sizes to test.
Nurse said the researchers were expecting walnut shells to be the most effective as a standalone treatment, believing that “because they’re large they would rip the leaves and everything. But actually, it’s the baking soda that tends to be the best treatment that we have found.”
He explained, “The smaller the particle size, the better, because it gets through the nozzles better; it doesn’t clog up as much, and it actually injures the weeds more, because it’s coming out of the gun at a higher velocity.”
A 2025 report from the team on the effects of the treatment specifically in potato fields found baking soda to have “consistent and high levels of control of annual species including common lambsquarters (Chenopodium album), corn spurrey (Spergula arvensis), and common dandelion” without impacting the marketable yield.
While it hasn’t been part of the trials so far, Nurse said he would be interested in looking at water as a medium for projectile weed control as well.
“Especially using hotter – like boiling – water, or something that would have a double effect,” he said. “It’s just the concern of water being a limited available asset to some growers; they don’t all necessarily have access to the amount of water that you would have to use to do a large acreage field.”
In the meantime, there’s still plenty to investigate with the current mediums being trialed.
With the baking soda specifically, there’s a question of whether it’s changing the pH levels of the soil.
“We don’t have that data yet, but that is a concern in some cases,” Nurse said. “I think at the rates we’re using it, there’s unlikely to be a huge effect on the soil chemistry, but we are going to verify that.”
There are other factors the researchers are looking into with their testing as well.
“We’re not just looking at the abrasives alone,” Nurse said, noting the angle used to apply the abrasives can impact the efficacy.
“We’re finding that the 45-degree angle is the best,” he added.
With the testing, the researchers tried to pick crops that would be more tolerant to the projectile weed control if they happened to get hit with ‘friendly fire.’
For instance, in Ontario, they’re looking at using this process with white beans; in Prince Edward Island, it’s being used with potatoes; and in Quebec, they’re testing it in vineyards.
“We do get a little bit of crop injury through it, but when we’re going for the weeds, we’re trying to aim below the canopy – if there is a canopy that has already formed – so the injury is minimal to the crop,” Nurse said.
Ideally, the projectile weed control should be used when the weeds are just emerging from the soil. Of course, that’s not always going to happen on a busy farm, and projectile weed control is an imperfect practice.
“When we’re going through, (we) don’t necessarily capture everything,” Nurse said. But it doesn’t have to be an either/or choice for farmers – and the data is showing that it shouldn’t be.
“I don’t think the abrasives alone are going to be good enough in most cases, because you have a mixture of large weeds and small weeds,” Nurse said.
“If (the weeds are) larger, we found that it’s best to pair the abrasives with an herbicide. For this project specifically, we’ve switched to looking at a bioherbicide, just to target more of the reduced risk – maybe the organic sector – with the project.”
Nurse noted that when projectile weed control practices have been paired with a bioherbicide – or, as in the past, when they’ve done it with an herbicide – weed control is scoring better than 80%.
“You’re getting good control that way,” Nurse said. “We’ve been able to reduce the amount of herbicide that we’re using, or the amount of treatments or applications that would be required, while still maintaining the weed control. And we’ve had less than 10% crop injury.”
Even with promising results so far, the work is far from done.
“If the project continues … the next logical step is to either start trying to get an engineer on the project and (design a) commercialized unit or do some grower trials, but we aren’t at that stage yet,” Nurse said.
In the meantime, the research continues.
