Farm Progress is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Serving: Central

Computer modeling predicting resistant weeds

It may not be on the catwalk, but Paul Neve does plenty of modeling. His latest work isn’t aimed at the couture crowd, unless your idea of high fashion is a pair of jeans and work boots.

No, Neve is a professor at Warwick University in Birmingham, England. On his current trip to the United States, Neve brought along his latest computer model — one aimed at glyphosate-resistant weeds in Arkansas.

Invited to the state by the Arkansas Extension Service last summer (see, Neve toured fields full of glyphosate-resistant weeds and researched cropping practices. After taking that information back to England, he began developing a computer model to help predict and manage the burgeoning Mid-South resistance problem.

The formal introduction of that model came during a Syngenta-sponsored press conference Feb. 19.

“We feel there’s a potential for the (resistance) problem to explode this summer,” said the Syngenta moderator. “That’s why we want to talk about this new model and program that’s available.”

The focus of the model is Palmer amaranth, “a very important, very aggressive (pigweed) species that is causing havoc in the Mid-South and Southeast,” said Chuck Foresman, Syngenta’s head of weed resistance strategies.

“Our intent is to provide answers for growers on what they should implement on their property prior to the time actual resistance occurs. That’s what models are about — they allow you to make intelligent decisions, to extrapolate the odds, if you will, in favor of delaying resistance.”

Neve, showing a slide including USDA weather and crop bulletin data, said producers are already well aware of, and receptive to, modeling. “That’s really what we have with the glyphosate-resistance model: we’re taking our knowledge of the current state of play and what’s happened in the past and are using it to try and predict future glyphosate resistance problems. More importantly, we want to find management to prevent those problems.”

There are two main categories where models can help growers, said Neve. “One, they can help research and Extension scientists to understand glyphosate resistance and help us identify where the gaps in knowledge are. Where does future research need to be directed to help solve it?”

More importantly, “models help us identify the major risk factors for resistance. Once those are known, we can start to design management strategies we think will reduce the risk of resistance. Then, those can be run through the model to come up with recommendations for farmers.”

The models allow researchers to do long-term, virtual experiments — “experiments where we’re sitting at the computer. Obviously, the benefit is the model can be run into the future over 10, 20, or even 30 years. That isn’t possible with field trials. Or, at least, it isn’t possible to do (such lengthy) field trials and come up with solutions before” the weed problem outruns the research.

Resistance is really a numbers game. “What the models help us do is load the dice in our favor, to reduce the risks of resistance as much as possible.”

The bad news, if there is any, is if the current level of glyphosate use continues, “we may expect the current weed resistance problems could be the tip of the iceberg.”

The good news?

“We know glyphosate-resistant genes are rare, much rarer than resistance genes to many other herbicides. That provides a relative advantage. It means if glyphosate is used appropriately we can ensure its efficacy far into the future.”

Diversity of crops and diversity of modes of action are key to ensuring such longevity.


In Arkansas, cotton production is where most researchers have focused initial efforts with weed resistance. In the state — and across much of the South — most of the cotton is monoculture with producers growing cotton in the same fields year after year.

“In the late 1990s, we began to adopt Roundup Ready cotton,” said Jason Norsworthy, research weed scientist and assistant professor at the University of Arkansas. “As a result, we had heavy reliance on glyphosate. Most production fields had multiple applications.”

Producers also made a switch to stale seedbed production. “Essentially, the crop ground is bedded in the fall and in-crop cultivation is eliminated. So we have reduced tillage in these (fields).”

In such an environment, Palmer amaranth quickly became one of the state’s most problematic weeds. After survey work in 2006, “I’d say Palmer amaranth was second only to glyphosate-resistant horseweed. By now, I daresay Palmer amaranth far surpasses even horseweed in terms of being a problem in Arkansas.”

Last summer, Arkansas-based researchers began working with Neve and Syngenta on efforts to “use (Neve’s) model framework developed for rigid ryegrass and adapting it for Mid-South Palmer amaranth. Then, we wanted to take the model we adapted and review the risk of glyphosate resistance under our current management regimes in the Mid-South.

“Using the model, we also wanted to look at alternative management options to see if they’d reduce the risk of glyphosate-resistance.”

That model shows in a glyphosate-only system, “we predict the evolution of resistance to occur quite rapidly. In the fourth year, the risk of resistance occurring increases substantially. Over a short period, we see a maximum of about 67 percent likelihood of resistance occurring in a field.”

In a second scenario — adding Valor at layby directed in the last trip across the field — “we slightly reduced the risk of resistance from 67 percent to about 60 percent. That isn’t a substantial change.”

The third scenario shows “a real change” when applying Reflex at burndown or prior to planting. There’s a preplant residual and Reflex is “very effective at giving residual control of Palmer amaranth. The model shows we’ve gone from four years to about six years before the chances of resistance increase.

“We also see that rather than a 67 percent likelihood of resistance, we’re now at maximum of about 30 percent. So Reflex (caused) more than a two-fold reduction in the chance of glyphosate resistance.”

Another run from the model looks a bit different than the others.

In it “over time, the resistance probability is increasing.... The reason is there are periods during the season … when we don’t have herbicides out and weeds are emerging. As a result, we’re not controlling the weeds and a tremendous amount of (weed) seed is being produced. That means, over time, there are more and more plants in the field and the probability of resistance increases.

“The point from this slide is we need to know where to actually place the residual herbicides. We need almost a season-long program of controlling Palmer amaranth. Any (control) gap in the season could increase the likelihood of resistance evolution.”

When researchers first began looking at early-season Palmer amaranth control, Reflex wasn’t labeled in Arkansas for that window of application.

“Although our data shows it’s very effective and the model shows it, we weren’t allowed to use it,” said Ken Smith, Arkansas Extension weed specialist. “Since looking at the data, we went back and got a label for Reflex used at 14 to 21 days preplant. This offers farmers a tool they wouldn’t have had otherwise.”

With the help of Neve and Syngenta, Smith said, “we now have a little better handle on what to expect from (resistant Palmer amaranth) and how to manage it. We can take the information (from Norsworthy’s model runs) to farmers and say, ‘If you have these situations, you need to do this. If you don’t have resistance and everything is working, this is what will happen.’”

Smith, who has attended a “tremendous number” of farmer meetings this year, said farmers are very interested in resistance management. “They realize the magnitude of this problem and want to know what to do. Pigweed that’s been sprayed with 44 ounces of Roundup Weathermax and isn’t even affected gets everyone’s attention.”

Roundup Ready crops have changed the way the Mid-South farms. “We’ve been farming more acres with less equipment and less management,” said Smith. “(Roundup Ready) is the very best tool we’ve ever seen in weed control. It revolutionized the way we do weed control.

“But glyphosate-resistant weeds will force us to change again. We know that and our farmers are accepting that fact.”

Accessibility and cost

How easy will it be for farmers to access Neve’s model?

Not very, said Smith. “It probably won’t be at a level most farmers would want to access. The information coming out of the model will be readily accessible through our educational program. I don’t think this model was ever intended to (be interactive and allow) farmers make inputs and then come up with an answer on the backside. It’ll be utilized more in helping (researchers) make good recommendations to farmers.”

In its present state, the model “isn’t really user-friendly,” said Neve.

How much is glyphosate resistance costing producers?

“That’s rather difficult to get a handle on,” said Foresman. “In part that’s because some (farmers) are battling glyphosate resistance and don’t even know it.

“That aside, we did a survey in April 2006 asking growers … what glyphosate weed resistance was costing them. There was a range of responses but as I recall (the answers) were anywhere from $10 to $20 per acre.”

Arkansas consultants were surveyed in 2006, said Norsworthy. “I believe consultants estimated it would cost $15 to $20 per acre. But the range is very wide depending on how (each operation) is producing a crop.”

“We know that for resistant horseweed … it’s costing between $5 and $8 in additional costs,” said Smith. “And that’s on about 3 million acres in the state.”

e-mail: [email protected]

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.