Farms are battlefields, pitting growers against rapacious pests and aggressive weeds in never-ending, costly campaigns that often involve chemical weapons. Those weapons, alas, also harm innocent bystanders such as bees, fish, and crustaceans. Now, a large study charts epic shifts that have occurred in recent decades as U.S. farmers have changed their arsenal of pesticides. Birds and mammals have fared much better, whereas pollinators and aquatic invertebrates are suffering. The toxic impact to land plants has also skyrocketed, likely because farmers are using increasing kinds of chemicals to fight weeds that have become resistant to common herbicides.
“These trends show remarkable shifts over time in toxicity,” says John Tooker, an entomologist at Pennsylvania State University, University Park, who was not involved in the new research. “Just the scale of what they did is really, really impressive,” adds ecotoxicologist Helen Poynton of the University of Massachusetts, Boston.
In recent decades, the amount of insecticides used in the United States has gone down by about 40%. But at the same time, active ingredients have become more powerful. For example, pyrethroids, fast-acting insecticides that affect the nervous system, are very toxic at extremely low concentrations. Some require as little as 6 grams per hectare, compared with several kilograms of the older organophosphate and carbamate pesticides. This made Ralf Schulz, an ecotoxicologist at the University of Koblenz and Landau, wonder whether overall toxicity in the ecosystem had changed. A few studies had looked at certain compounds and organisms, but nothing had been done on a national scale.
Schulz and colleagues started with U.S. Geological Survey data on self-reported pesticide use by U.S. farmers from 1992 to 2016. They also gathered acute toxicity data from the U.S. Environmental Protection Agency (EPA) on those same compounds—381 in all. Next, they compared EPA’s regulatory threshold levels—the point at which a substance might harm vegetation or wildlife—with the amount of each pesticide applied to farm fields and determined a “total applied toxicity.”
The good news is that total toxicity plummeted more than 95% for birds and mammals from 1992 to 2016, the team reports today in Science, largely because of the phaseout of older pesticides. Toxicity for fish declined by less—about one-third—because they are more sensitive to pyrethroids. The bad news: Pyrethroids have caused toxicity to double for aquatic invertebrates, such as plankton and insect larvae that are a key part of food webs. And another popular class of pesticides, neonicotinoids, has doubled the risk to pollinators like honey bees and bumble bees. This overall trade-off—vertebrates impacted less and invertebrates hit harder—has also been seen in a smaller study.
For some pesticides and species, however, estimating the real-world impact is tricky. That’s because many factors affect whether a chemical will harm plants or animals, such as the weather or the time of year. To see how directly pesticides affected aquatic crustaceans and insects, the researchers looked at peer-reviewed toxic exposure data from 231 lakes and streams across the United States. When they compared the data with the amount of pesticides applied nearby, they found a “relatively strong” correlation.
Plants have also been impacted. Since 2004, the total applied toxicity from weed killers has doubled in land plants. One of the major herbicides contributing to the rise is glyphosate, which has simplified farming, improved soil conservation, and allowed farmers to switch away from more toxic herbicides after the advent of crops genetically modified to tolerate glyphosate in the 1990s. But since then, some weeds have evolved resistance to glyphosate, and farmers are spraying additional types of herbicides. That threatens flowering plants that grow in field margins, providing food and habitat for other species.
Even one crop species genetically engineered to reduce pesticide use—corn containing an insect-killing chemical called Bacillus thuringiensis (Bt)—has seen its toxic exposure rising fast. Total applied toxicity in Bt corn has been increasing just as quickly—8% per year over the past decade—as in non–genetically modified corn. “It was a bit astonishing,” Schulz says. “I didn’t expect that, I must admit.” The reason, Schulz suspects, is that pests are evolving resistance to chemicals that are overused in both types of corn, requiring more frequent applications. “That is really one of the major problems agriculture is suffering from.”
Schulz hopes the results will help policymakers and others think more broadly about the complexity of pest and weed control, and the trade-offs for wild species, in order to reduce unintentional harm. Tooker notes that the rising toxicity in plants and aquatic invertebrates could lead to less diverse habitat and food resources that eventually ripple through animal populations, potentially causing losses. “The patterns in the U.S. pesticide use and toxicity data should be a cautionary tale for the rest of the world, much of which seems to be leaning more heavily on pesticide use rather than ecological interactions for pest control.”
Ultimately such decisions come down to how society values various groups of species, says Edward Perry, an agricultural economist at Kansas State University, Manhattan. For example, regulators could restrict the use of neonicotinoids, as has happened in the European Union, to benefit pollinators. But farmers would likely switch to other insecticides that could pose different hazards to species—or face lower yields and higher food prices.