When Residents Say ‘No’ to Aerial Mosquito Spraying

On a Friday afternoon in late September, Kalamazoo County Health Officer Jim Rutherford announced that aircraft would mist much of the county with an insecticidal spray. Intended to kill mosquitoes, the emergency plan quickly turned into a public relations battle. Hundreds of calls and emails — and even some threats — streamed into Rutherford’s office in southwest Michigan, many expressing concern about the spray.

In the United States, an average of seven human cases of eastern equine encephalitis (EEE) are reported annually, according to the Centers for Disease Control and Prevention. But 2019 has been an especially bad year for the mosquito-borne virus, with at least 35 cases and 13 deaths reported nationwide. In Kalamazoo County, when Rutherford made his decision, EEE had killed a 64-year-old man and sent a 14-year-old girl into intensive care. Faced with the prospect of several more weeks of mosquito-friendly weather, Michigan state officials had offered 15 counties the option of spraying. All of them accepted.

“This technology is fully recognized as a public health intervention for mosquito-borne diseases,” Rutherford said, citing information from the Environmental Protection Agency and the CDC. But that didn’t stop thousands of residents from flooding the state’s pesticide opt-out system, requesting that their properties be exempted from spraying. Rutherford said he repeatedly heard things like “the government told me Roundup was safe forever, the government told soldiers that Agent Orange was safe forever” — only to later be informed of previously unforeseen risks.

Three days after the initial announcement, Kalamazoo County called off the spray. More than 1,400 residents had exercised their right to opt out, creating a patchwork of no-go zones that simply made aerial spraying unworkable. Spraying did occur in the other 14 counties, skirting the property of around 1,600 additional opt-outs, and over the vigorous objections of many residents.

A female cattail mosquito, Coquillettidia perturbans, taking a blood meal. In Michigan, this species is an important bridge vector for the EEE virus, transmitting the virus between species like birds and humans.

Visual: Sean McCann, Department of Biology, Simon Fraser University

A hard frost will soon kill this year’s remaining adult mosquitoes in Michigan, Massachusetts, Indiana, and other affected states. But as a warming climate promises to increase mosquito-borne disease outbreaks across the northern United States, including EEE, the controversy raises questions that may resonate for years to come. When should elected or appointed officials compel people to accept public health interventions? When should people have the chance to opt-out? And, in the face of new public health threats, how can communities have constructive, inclusive conversations about risk?

Finding satisfying solutions might not be easy. “There’s not a broad level of trust with government telling people that [this pesticide is] safe,” Rutherford acknowledged. “I get that part of it.”


In a typical year, the adult mosquitoes that spread EEE would be dying out in Michigan by late August. This year, though, they persisted well into the fall — and by September, facing mounting cases of EEE, public health officials grew alarmed.

“At the point where we first started to do the aerial treatments, we had seen more cases in one year than we had seen in an entire decade combined,” said Lynn Sutfin, a spokesperson for the Michigan Department of Health and Human Services, describing the outbreak as “a public health emergency.”

The state decided to spray Merus 3.0, an insecticide that contains a mixture of plant-derived chemicals, collectively called pyrethrins, that is approved for use on organic farms. Contact with a single, microscopic pyrethrin droplet can kill an adult mosquito. The aircraft often distribute less than an ounce of the chemical per acre — roughly the equivalent of misting a shot glass full of whiskey over a football field, according to Ary Faraji, an entomologist and president elect of the nonprofit American Mosquito Control Association. By contrast, farmers will sometimes use an order of magnitude more per acre when treating crops for pests.

“The overwhelming scientific evidence suggests that these treatments are effective in what they’re designed to do, which is lessen the disease risk, and also that they do not pose significant risks to people,” said Robert Peterson, an entomologist at Montana State University who studies the risks and impacts of pesticide application.

Critics, of course, point to evidence that pyrethrins are highly toxic to honeybees and wild pollinators. But even here, officials and mosquito control experts say that spraying in the evening, when mosquitos are active but bees are not, dramatically reduces the risk to pollinators. Still, that did not reassure some worried beekeepers. The Michigan Pollinators Initiative is currently surveying beekeepers in counties that received aerial spraying, to see if they noticed any difference in their bees.

“No pesticide is inherently safe,” said Drew Toher, community resource and policy director for Beyond Pesticides, a Washington, D.C.-based anti-pesticide advocacy group. “There should be that message out there, that the disease itself is a risk to public health, but so is the use of this pesticide,” Toher said. He cited peer-reviewed research that suggests a correlation between pyrethrin exposure and developmental problems in children. (The studies do not necessarily reflect the exposure levels seen in mosquito spraying). The final decision about how to balance competing risks, said Toher, “should be in the hands of the community.”

Under the microscope, the salivary gland of a mosquito infected with Eastern Equine Encephalitis is visible, shown in red. 2019 has been an especially bad year for the mosquito-borne virus, with at least 35 cases and 13 deaths nationwide. Visual: BSIP/Universal Images Group via Getty Images

That actual decision-making process varies from place to place. During spraying, many places permit opt-outs only for organic farmers, beekeepers, and people with certain health conditions.

In Massachusetts, the site of another significant EEE outbreak this year, once state officials declare a public health emergency, only “certified organic farms, commercial fish hatcheries/aquaculture, priority habitats for endangered species, and surface drinking water supplies,” can be exempted from spraying, according to Katie Gronendyke, a spokesperson for the state’s Executive Office of Energy and Environmental Affairs.

In Elkhart County, Indiana, just across the border from Michigan, elected county commissioners made the decision to spray, only organic farmers could opt out.

The story in Michigan, though, is more complex. Appointed public health officials, at the state and county level, made the decisions to spray, albeit in consultation with elected representatives. And because the state health department, in the interest of time, chose not to seek an emergency declaration from the governor that could override the opt-out, state agriculture regulations permitted individuals to exempt themselves from spraying.

To some people, that opt-out policy may have gone too far. Rutherford, the Kalamazoo County health officer, pointed out that while 1,400 county residents opted out, tens of thousands more did not. “The frustrating part on our end was the fact that 4.5 percent of the population made the decision for the other 95 percent of the population,” Rutherford said.

“The governor could have made this a lot easier had she declared a state emergency, but that wasn’t going happen,” he added.

Elsewhere, supporters of the spraying accused opponents of peddling bad science and obstructing the response to a public emergency. One widely circulated op-ed, published at MLive.com and printed in one of the media group’s papers, The Kalamazoo Gazette, compared the people opting out to individuals who opt-out of vaccines. “Those who have opted out put others at unnecessary risk,” wrote the author, Parker Crutchfield, a medical ethicist at the Western Michigan University medical school.

In an interview with Undark, Crutchfield said he disagreed with the opt-out policy. “I don’t think there should be much choice, frankly,” he said, adding that there should be carve-outs for people with certain medical conditions.

For others in Michigan, the option to avoid spraying did not go far enough. In early October, around 45 people gathered for a special meeting in Webster Township, a community of some 7,000 people about 10 miles northwest of Ann Arbor. Aerial spraying was anticipated in the next few days.

County officials had announced the decision to spray barely 48 hours before the treatments could have begun (they ended up happening later, due to weather), leading to perceptions that the opt-out process was mostly a fiction.

“We value working with our community and hearing from our community and using cooperative decision-making whenever possible,” said Susan Ringler-Cerniglia, a spokesperson for the Washtenaw County Health Department, in an interview with Undark. “But in this situation, this was really a tall order,” she said, emphasizing that officials had been handed a very short timeline because of the public health emergency. She added that effective community outreach had grown more difficult since the scaling-back of the local newspaper.

One speaker at the meeting was Katherine Larson, who lives just outside the spray zone. When we spoke by phone a few days after the meeting, the spraying had finally happened. Larson told me that, prior to the spraying, she had distributed leaflets to residents in the spray zone. The majority of people she talked with, she said, had no idea the spray was coming. When the plane arrived, many residents called her, confused and upset.

“We’re supposed to be free, and we’re still supposed to be able to choose between whether we want to put on pesticides on our skin or repellants on our skin, or whether we want to be sprayed,” Larson said.

“What I care about is people having a choice,” she added. “When your government imposes something on you and says, ‘you have an opportunity to opt out,’ but they don’t give you the time to opt out, then it’s not a choice.”

Where Larson sees an infringement on personal liberties, Crutchfield, the medical ethicist, sees a necessary sacrifice for the sake of community safety. After all, the mosquito flying over my yard today may bite you tomorrow. “We have to allow some sort of intrusion in order to even live as a society,” he said. “And I view the mosquito spraying as one of those minor intrusions.”


One reason for the scale of backlash in Michigan may simply be the novelty of the spraying. Unlike places like Florida, where mosquito-borne sicknesses are recurring threats and aerial spraying is common, if sometimes controversial, Michigan had not sprayed for mosquitos since an EEE outbreak in 1980.

As the climate warms, and northern states potentially deal with more frequent outbreaks of things like EEE, those applications could continue to trigger concern. In an email, risk communication expert Peter Sandman pointed out that spraying for EEE fits two familiar patterns that he sees in other debates. He wrote that, again and again, “natural risks provoke less outrage (concern, fear, anger) than industrial risks.” In other words, something like EEE — a natural risk — may seem less frightening to people than the perceived risks of pesticides. And, Sandman added, “it’s also worth noting that a risky action” — like spraying a chemical — “tends to arouse more outrage than a risky inaction” like standing by while mosquitos spread a deadly virus.

That’s not to say that opponents of the spraying should be dismissed, Sandman said. “I think in many cases public health agencies are overstating the risk of EEE and understating the risk of spraying,” Sandman wrote. “They tell stories about EEE victims without stressing how few such victims they are; they stress the rarity of insecticide poisoning without telling comparable stories about how horrible it can be.” And both risks, Sandman stressed, are relatively small.

All of these conversations will unfold during a time of increasing mistrust in public institutions. Distrust in public health agencies — and in state government more broadly — may be especially acute in Michigan following the Flint water crisis, where government officials wrongly assured the public that Flint’s drinking water was safe. More recently, residents have rebuked state officials for their handling of PFAS contamination in drinking water. (Rutherford noted that Kalamazoo County recently had to shut down a water system because of PFAS contamination.)

For now, the county is left urging people to exercise caution, distributing strong insect repellant to vulnerable populations, and waiting for the first hard frost.


UPDATE: Due to an editing error, a caption in an earlier version of this story misspelled the Latin species name for the cattail mosquito. The story has been updated to correct this error.

Michael Schulson is a contributing editor for Undark. His work has also been published by Aeon, NPR, Pacific Standard, Scientific American, Slate, and Wired, among other publications.