MARICOPA — University of Arizona professor of entomology Peter Ellsworth crawled through the furrow between two rows of tall cotton, wearing a gas mask and rolling up a 25-foot-long sheet littered with stunned insects.
Ellsworth is an integrated pest management specialist. Four days before, Maricopa County Cooperative Extension Agent Ayman Mostafa, Ellsworth and summer technician Jose Partida had used a tractor to spray a 40-by-40-foot plot with milk, a protein not usually found in cotton fields.
The night before the collection, they had spread and staked two 25-foot-long sheets over the furrow, one sheet on top of the other, so the top one would stay clean. The morning of the collection, they spread a 25-foot canopy over the two rows of cotton and fogged under it with a botanical insecticide made from dried chrysanthemum flowers. The collection area was about 600 feet from the plot that had been sprayed with milk.
Ellsworth rolled up the insect-covered sheet so he and his research assistants could take it to the lab and count and identify the insects, carefully lifting each one with its own toothpick and depositing it in a clean test tube. The insects were sent to James Hagler’s lab at the U.S. Department of Agriculture to be tested for milk, a sign that they had moved from the area that had been sprayed. Collections would be repeated at two, four and seven days.
Ellsworth said they are trying to understand how lygus and their predators move within a field, whether they concentrate in an area and stay there or continually move around. The fogging collection is a one-year pilot study funded by the Arizona Cotton Growers Association and is part of a long-term study about how insects move.
Ellsworth said that in the early 1990s, Pinal County cotton farmers used to spray their fields 10 to 12 times a season with broad-spectrum insecticides that were toxic to humans, wildlife, honey bees and the insect predators that help keep pests in check.
Starting in 2006 and lasting until 2011, the use of broad-spectrum insecticides dropped 95 percent, and farmers sprayed their fields an average of 11⁄2 times a season with selective insecticides. Ellsworth said 23 percent of them did not spray at all in 2011.
Bt cotton had been introduced in 1996, he said. It was genetically modified to have one or two genes of a soil bacterium, Bacillus thuringiensis, which creates a protein that kills caterpillars.
Selective insect growth regulators had been introduced in 1996 to help control whiteflies. Selective feeding inhibitors had been introduced in 2006 to help control lygus bugs. Farmers had learned more economical and sustainable ways to manage insects without disrupting the natural processes and saved $388 million over 15 years.
Everything looked great until 2012, when the numbers of brown stink bugs, Euschistus servus, exploded.
The bugs had always been there, Ellsworth said. He used to find one or two every summer, but the last time farmers had seen them in great numbers was 1963.
“We don’t know why that’s happening,” he said.
Brown stink bugs feed on developing cotton bolls, attack the seeds, stain the fiber and introduce organisms that cause boll rot. No safer selective insecticides have been found yet to stop them, so farmers had to return to using broad-spectrum insecticides, which killed the predators, too, and increased the need to spray to more than 31⁄3 times per season.
Ellsworth said farmers used to roll their eyes in the early 1990s when he talked about good and bad bugs, but there were no selective insecticides then. Last year when he suggested they use a broad-spectrum insecticide on the brown stink bug, they didn’t want to for fear of harming their beneficial insects. But the stink bug threatened their yields, and they had to spray.
Ellsworth said understanding how pests and their predators move through a field helps scientists find ways to control them. Maybe it won’t be an insecticide next time. Maybe it will be more about how crops are managed or arranged.
He was involved in a five-year study that tracked the movement of lygus bugs when they were surrounded by different crops and found that some crops were sources of lygus infestation. Planting near them increased the damage. Some crops were sinks and planting near them decreased the damage. Some crops, like forage alfalfa, were sources or sinks depending on how they were managed.
Ellsworth took the information gathered from 50 fields in Pinal County and created a computer simulation that allows farmers to plant a virtual farm with different cropping patterns and watch the changes in lygus infestation and profitability.
The farmers learned the game fast, he said. In four runs, they decreased the virtual infestation from 38 percent to 13 percent without decreasing the cotton acreage. They also increased virtual profits from 79 percent of possible to 92 percent.
“We finally developed enough computer power and collected enough information from real, commercial-scale operations that we can now put it all together,” he said.