Genetically engineered crops are becoming victims of their own success.
Over the past decade, cotton modified to produce a toxin lethal to its most pesky insect predator has revolutionized the industry. From the U.S. cotton belt to northern China, farmers have found that with their primary foe neutered, they have been able to dramatically cut their insecticide use -- at least, for a time.
While some U.S. growers have cannily integrated the crop, known as Bt cotton, into pest strategies that have dropped insecticide use by 75 percent, not all farmers have applied the same forethought in adopting the cotton. And these farmers are reaping the consequences, as an echo boom of bugs that had previously been controlled by indiscriminately applied insecticide are beginning to rise again.
Nowhere has this rebound been more problematic than northern China, where a minor pest, the mirid bug, has become a major ordeal. Farmers there widely adopted Bt cotton, which has devastated its target, the bollworm. Pesticide use fell. Few expected the mirid would take advantage of the situation.
"Historically, mirids were not an important insect pest in the Chinese cotton system," said Kongming Wu, an entomologist at the Chinese Academy of Agricultural Sciences. "In most years, farmers haven't needed to control it."
But soon after Bt cotton's introduction, the mirid population began to rise. Immune to the Bt toxin, the mirids began to feed, spreading to other crops like grapes, dates and peaches. The infestations demanded insecticide sprayings, and the environmental gains won from Bt cotton began to be lost.
Today, farmers in northern China, who nearly all grow Bt cotton, are spraying 20 percent less insecticide than before the crop arrived. The mirid has displaced the bollworm and become a significant pest, according to a paper by Wu and several others published last week in Science.
Bt cotton is not the only bioengineered crop that is beginning to have unintended, if not unanticipated, effects. Thanks to the wide use of weedkiller-resistant crops, some nine weeds have evolved tolerance to Roundup herbicide in various regions of the country. And several species of moth larvae in India and the United States have shown an increased ability to tolerate Bt toxins.
Many of these problems, both direct and indirect, stem as much from the success of modified crops and farmers' enthusiastic adoption. Facing unending rows of engineered crops, insects and weeds face intense selective pressure to evolve resistance. And the rise of secondary pests ties directly to the drop in broadly poisonous pesticides used before Bt crops.
Wu's long-term cotton study -- conducted over 10 years at 38 cotton-growing plots across six provinces in northern China -- is the first to confirm that pest-resistant crops, and the drop in insecticide use that accompanies them, can drastically alter the farming landscape, carving out vacant niches for previously suppressed species.
But despite these seemingly damning results, the Chinese study is not surprising, said Peter Ellsworth, an entomologist at the University of Arizona.
"I've got to tell you, the underlying data and the base conclusion come as no surprise to any Arizona grower or any Arizona scientist," said Ellsworth, who has worked with the state's cotton growers to develop a pest strategy based on Bt cotton that has cut insecticide use by 75 percent -- a far larger gain than seen in China.
Rather than hoping Bt cotton would solve their insect problems, Arizona farmers -- who now nearly all grow the cotton, developed by the biotech giant Monsanto Co. -- have adopted a wealth of strategies to control pests that could rise after the demise of pink bollworm, their primary pest.
One of these insects, the lygus bug, is a close relative of China's mirid. And while the lygus has become more of a problem, it is manageable, Ellsworth said.
"While Bt cotton is a wonderful innovation for us, its focus for us is on the pink bollworm," he said. "We're very happy to have it, but to deploy it in a vacuum -- that simply does not acknowledge the complex ecological system in which agriculture is deployed."
Northern China, in fact, is far from representing a simple cautionary tale that should warn society away from biotech crops. Instead, it shows the inadequacy of relying on one technological trick to maintain an ecosystem, said Bruce Tabashnik, the head of entomology at the University of Arizona.
"The lesson here is an old one, but we can always relearn it," Tabashnik said. "There is no silver bullet for insecticide control. ... Bt crops are not the solution. They are a new tool. And we can use that new tool wisely."
Spray early, often
Bt cotton was a godsend for northern China when it arrived in 1997. The region's 10 million farmers, who mostly work on small family plots, had a terrible time warding off the bollworm, a moth larvae that in its many varieties maintains a consistent lust for chewing into cotton's green, football-shaped seed pod.
Facing such an intractable foe, the Chinese farmers used the only tool they had available: They smothered their crops in insecticide. By the 1990s, the growers were averaging more than 15 applications each season, a poisonously high amount for the broader environment, killing pest and beneficial insects alike.
One of the casualties of the sprayings were mirids, spritely green bugs distantly related to the aphid. The mirids, attracted to fruit, would swarm to cotton in the early summer, when it is often the only flowering plant. Just then, farmers would coat the cotton in chemicals, leaving it a deadly, unintentional mirid trap.
That all changed after Bt cotton was introduced, Wu said.
"After Bt cotton adoption, the farmer didn't need to use insecticide to control the cotton bollworm," he said. "This gave mirid populations a space to multiply in cotton fields, and then spread to other host crops."
Wu saw that Bt cotton, despite its "highly beneficial" nature, would alter the ecological landscape of Chinese farms. And so he and his colleagues began monitoring the region's 72 million acres of farmland for more than 10 years, the type of landscape-level study that has rarely been deployed for bioengineered crops, he said.
"Bt crops only have a short history," Wu said. "In previous studies, people mainly used small sizes and short-term experiments. Given the complicated change in the ecological system, we needed to do an ecological assessment and monitoring on the landscape level."
Few studies have had the ambition of the Chinese survey, said Michelle Marvier, an environmental biologist at Santa Clara University who, several years ago, compiled and analyzed 42 small-scale experiments studying the effects of Bt cotton and corn.
Most of these studies have focused on the possible harm Bt cotton could do to innocent insects like butterflies -- after a deep probing, they showed an "absence of effect," she said -- not the indirect impact of falling insecticide use.
There has been some resistance to implementing landscape-level studies, but Wu's work really flags how important it is to start capturing indirect impacts, Marvier said.
"This does point to how complex it is to really assess the risks and benefits" of bioengineered crops, she said.
Spiders in the fields
While agronomists and entomologists in the United States may not have conducted studies as large as those in China, they were not surprised when secondary insect populations began to rise in the wake of less pesticide use. In the Southeast, stink bugs quickly became a problem, said George Kennedy, the head of entomology at North Carolina State University.
"They came to fore pretty quickly," Kennedy said. "We learned how to deal with them."
This has been especially true in Arizona, where the primary cotton predator, the pink bollworm, was fantastically susceptible to Bt cotton (its Southern cousin, however, has shown hints of resistance). Scientists saw a chance to dial back on pesticides and instead use selective targeting technology and better growing practices to fight the remaining insects.
Whiteflies are now controlled with specialized growth hormones. Lygus bugs, the mirid relative, are attacked with feeding inhibitors, chemicals that cause their straw-like mouths to bend flaccid, unable to pierce fruit. Farmers received growing patterns and pest thresholds that would discourage unnecessary spraying. And sterile moths have been introduced to eliminate any straggling bollworms.
As a result, broad toxic pesticide use has been almost completely eliminated in Arizona's cotton-growing regions, areas that since the 1960s had been sprayed 10 times a year or more. Since 2008, in fields more than 150,000 acres, there has not been a single application, which has allowed beneficial insects to return.
"It allows the natural enemies" -- ladybugs, spiders -- "to do their thing and restores a balance where a lot of insect control is being taken care of by natural control," Tabashnik said.
When Ellsworth first began working in Arizona 20 years ago, he found an ecosystem out of balance, cotton fields laid barren through heavy spraying. Nearly all the growers' questions were pest-related. But in the past five years, he said, it is not uncommon for them to ask different questions, like "What are these spiders?" or "Why are so many ladybugs here?"
"The average grower now acknowledges biological control," he said. "It wasn't even conceivable 20 years ago."
Arizona, it seems, can provide a model for China.
Wu hopes that growers will begin adopting integrated pest management strategies, rather than simply reverting to pesticides. Then, perhaps, the insecticide sprayings can go down again, and the ladybugs can return.