On March 14, 1977, a woman with the flu climbed aboard a 737 and headed for Kodiak, Alaska, with 53 other passengers and crew. After an engine failed, most of them sat on the runway with the cabin doors shut, and the ventilation system off, for two hours. Within three days, 38 more people were sick.
More than four decades after state and federal epidemiologists showed how easily viruses spread from person to person on airplanes, the novel coronavirus has decimated global aviation. Daily passenger screenings are down 95 percent, according to the Transportation Security Administration.
Though there have been significant advances since the 1970s, and airlines spent weeks touting the safety of flying and their steps against the coronavirus, passenger cabins still pose a danger for the spread of infectious diseases, experts said.
It is a problem of biology, physics and pure proximity, with airflow, dirty surfaces and close contact with other travelers all at play.
Breakthroughs are possible, researchers said. Ultraviolet lights that promise to destroy viruses without hurting humans are being tested by Columbia University scientists, who say the lights would be effective in airplane cabins, airports, hospitals and schools.
"As we speak, there are 100 hairless mice being exposed for 15 months," said David J. Brenner, director of Columbia's Center for Radiological Research. The mice live under the lights eight hours a day and get eye and skin tests every couple of weeks, and after eight months the researchers have found no damage, "which is encouraging."
The lamps could have helped prevent the spread of covid-19, Brenner said, but "it's come a little too soon for us. If it had come at this time next year, we'd be in a good position to fight it."
The Federal Aviation Administration and major manufacturers have long been aware of the risk of diseases spreading on flights and have sponsored research seeking improvements.
Boeing is experimenting with lavatories that can sanitize themselves in less than three seconds. Engineers at the U.S. manufacturer and its top competitor, Airbus, have explored changing the way air moves around passengers to reduce infections.
But fighting illnesses and preparing for disaster were not industry’s top priorities as global air travel soared.
"When you can easily sell your airplanes, you try to defer the problems to the future," said Qingyan Chen, a Purdue University engineering professor who helped lead a major FAA-funded research project on disease transmission aboard aircraft and has received research funding from Boeing. "Today we found that future is actually 2020."
How far airlines and manufacturers might go now, and how far regulators or passengers might push them, remains unclear. Boeing was already under siege for deadly design flaws in its 737 Max aircraft following two crashes, and the company said it is seeking takers for a "voluntary layoff" plan as the aviation sector has slowed "to a trickle."
"When the world emerges from the pandemic, the size of the commercial market and the types of products and services our customers want and need will likely be different. We will need to balance the supply and demand accordingly," chief executive Dave Calhoun told employees in an April 2 letter.
The FAA regulates airplane design and production, down to the requirements for fire-retardant seat cushions, and oversees airline operations. But while the agency sets some air-quality standards, such as acceptable carbon dioxide levels, it has not set rules for preventing the spread of infectious diseases in airplane cabins.
"It's very hard for industry to change. . . . It's not like a flip of a switch," said Chen, pointing to the expense of developing new features and getting them certified by the FAA. But those high costs also underscore the key role for government, he said. "Regulation is very important. If you regulate that, then they will have to do something."
The FAA declined to answer a question about why it has avoided setting requirements to address the spread of viruses on planes. A spokesman instead pointed to comments FAA Administrator Steve Dickson made March 11 on Capitol Hill.
Pressed on what requirements the FAA has affecting the "possible transmission of contagious diseases on commercial flights," Dickson, who took the top regulator job less than a year after leaving Delta Air Lines, instead discussed rules covering "noxious fumes." He also echoed statements from airlines during the pandemic.
"The risk to the public is no higher than it would be in any area where you have folks gathered," Dickson told Rep. David Price, D-N.C., saying the air on planes is "on par" with that in homes.
"The research that's been done over the years is that there's really not any significant difference in what you would have in a building such as we're sitting in today, in terms of air quality," Dickson added, before taking a question from Rep. Mario Diaz-Balart, R-Fla.
A week later, Dickson said he would be self-quarantining after a "brief encounter" before the hearing with Diaz-Balart, who tested positive for the virus and eventually recovered.
As the coronavirus spread globally, the message from the airline industry to would-be customers was that planes were safe and that extra precautions were being taken. But the science behind cabin air quality has been lost in some of the messaging.
As President Donald Trump downplayed the threat of the virus, the White House offered reassurances. Asked March 2 if he would be comfortable traveling with his family to Disney World, Vice President Mike Pence said that "there's been no recommendation about any limitations on travel" in the United States, adding that he travels "all the time."
The TSA screened more than 20 million people over the next two weeks, both passengers and workers, as travelers continued to process the dangers and the virus continued to spread. On March 16, Trump advised against unnecessary travel.
Unions representing aviation workers, who have complained about the lack of masks and other protective equipment, say hundreds of flight attendants and pilots have tested positive for covid-19, and at least seven have died.
Airlines have noted that all, or most, of their planes use high-efficiency particulate air filters. Air flows past millions of particle-grabbing layers in the filters and is blended with air sucked in through the engines, creating roughly a 50-50 mix of fresh and recirculated air.
Delta said that "HEPA air filters extract more than 99.999% of even the tiniest viruses." Alaska Airlines, citing studies from the European Union Aviation Safety Agency, argued that cabin air "can be better than the air found in many office buildings," given how frequently it recirculates. Southwest said the filters help provide onboard air quality "which, in most cases, exceeds the quality of air that can be found outdoors."
Chen, who edits the journal Building and Environment and has studied air quality in offices and cruise ships in addition to planes, said that while planes get high marks on some measures, many of the industry claims are misleading.
While air on a Southwest plane may be cleaner than what is outside in a highly polluted city in China or India, Chen said, that is not the case in United States, where Southwest does most of its flying.
The European research cited by Alaska Airlines focused on certain chemical contaminants, not viruses.
"To be honest, airplanes are not designed to prevent infectious-disease transmission," said Chen, who was co-director of the FAA-funded Airliner Cabin Environment Research center from 2004 to 2010. "They're not designed to do the job."
Experts agree that HEPA filters are highly effective at capturing everything from viruses to skin flakes.
But that is only part of the picture.
The problem is passengers can still breathe in tiny floating droplets from a coughing passenger seated nearby - before the air carrying those droplets can be vented out of the cabin and filtered.
Precisely how far the droplets might float before being pulled out and cleaned has been the subject of intricate scientific observation, airflow modeling and disease tracing, from the onboard flu cases in Alaska to an earlier coronavirus outbreak in 2003, when severe acute respiratory syndrome (SARS) killed 774 people around the world.
On March 15 of that year, a Boeing 737 took off from Hong Kong for a three-hour flight to Beijing, with a feverish 72-year-old man sitting in 14E, a middle seat.
Of the 120 people on board, 22 were later diagnosed with confirmed or probable cases of SARS, according to a reconstruction published in the New England Journal of Medicine that year. Researchers said the "most plausible" explanation was that they were infected on the plane by the man in 14E. He died of atypical pneumonia a few days after the flight.
The World Health Organization had defined "contact" with a SARS patient as sitting in the same row, or in the two rows ahead or behind the infected person.
The researchers found that the risk for those in the three rows in front of the man, or the same row, was much higher than for those sitting elsewhere. But two people seated as far as seven rows in front of him were also infected, as were two flight attendants. Five passengers later died.
People moving around or touching surfaces may have played a role, or the virus may have floated in the air for longer than expected, the researchers concluded.
A decade after the SARS flight, Emory University researchers, with support from Boeing, sought to address those findings.
Public health students carrying iPads on 10 flights from Atlanta built a "network model" of passenger contacts and infection risks. The authors, including a top Boeing physician, concluded that respiratory disease spread by droplets would likely be "limited to one row in front of or in back of an infectious passenger."
In seeking to explain how people sitting so far away on the Beijing-bound flight ended up sick, the Emory researchers said the SARS "transmissions" may have happened at the airport, or come from "other sources before or after the flight."
The authors of the original SARS study had acknowledged those possibilities, but said that many people got sick around four days later and that "clustering" pointed to the day of the flight as a crucial moment. They also said they could have missed others who became ill.
Experts in disease transmission are still wrestling with the implications of that Beijing flight and what insights it may provide for today's much more deadly coronavirus pandemic.
The Centers for Disease Control and Prevention in February reiterated the two rows up, two rows back guidance for defining "contact" with an infected patient, which can shape who gets monitored.
Some local public health officials have cast a wider net. Officials in Monroe County, New York, warned all passengers and crew members on a 50-seat jet bound for Rochester on March 14, along with everyone at the airport around the time it landed, to monitor themselves because a patient diagnosed with covid-19 had been on board.
The death of American Airlines flight attendant Paul Frishkorn last month highlights the need to "avoid further contagion" of airline crew members, the Air Line Pilots Association said, adding that at least three of its pilots have died. The group called on Transportation Secretary Elaine Chao to require that airlines be more exacting about cleaning and inform employees when they have been exposed, or aviation could "grind to a complete halt."
- - -
Chen was born to a farming family in Fujian province, in southeast China, where his university studies were delayed by Mao Zedong's brutal Cultural Revolution. He spent painful years lugging massive rocks for construction projects before coming to the United States and delving into the behavior of tiny particles on airplanes.
Working with two Boeing engineers and a team of researchers from Purdue, Chen wanted to know how changing an airplane's ventilation system would affect the risk ofcontracting SARS, as a stand-in for other dangerous viruses that might emerge.
Their results, published last year, were startling. They found that passengers sitting with a SARS patient in a seven-row section of a Boeing 767 would have a 1-in-3 chance of getting sick from a five-hour flight. On a shorter 737 flight, the risk was 1 in 5.
But they found that changing the existing ventilation system - essentially by having air flow into the cabin from near the floor rather than from above - would make a big difference, cutting the risk by half or more.
Chen said droplets were swept away from passengers more efficiently using alternative systems, one tested by Airbus engineers and another developed by Chen and his team.
One reason: The warmth of the passengers' bodies helped the flow of air coming up from underneath, since warm air rises, he said. Going with that flow, rather than fighting against it, lessened the turbulence that could keep germs on top of passengers.
"We try to not mix your air with your neighbors'," Chen said.
The study assumed the virus that caused SARS could stay airborne for long periods, which is different from what the CDC says is typically occurring with the novel coronavirus.
Scientists and public health officials said covid-19 is largely spread from one person to another through droplets, such as when people cough or sneeze.
Such relatively large droplets, brimming with viruses, generally travel only a limited distance before being pulled down by gravity, sort of like ping-pong balls falling back to the table after being whacked.
That is why social distancing guidelines have emphasized the six-foot distance and call for incessant hand-washing, to avoid picking up fallen droplets and bringing them to your face.
Fortunately, experts said, the coronavirus does not behave like the fearsomely infectious measles virus, which can stay aloft for up to two hours, more like the balls bobbing and floating around an old-school lottery machine.
But Chen - whose background is in mechanical engineering and fluid dynamics, not medicine - said the uncertainties around covid-19 remain a cause for caution.
The World Health Organization said droplets are the main path for infection, but noted research showing that the coronavirus has become airborne in some hospital settings, including during intubation to help breathing.
U.S. researchers reported suspending coronaviruses in the air and said such infections are "plausible."
The National Academies of Sciences, Engineering, and Medicine on April 1 pointed to evidence that the coronavirus can spread by air through "normal breathing," not just via droplets from coughing, and cited research from the University of Nebraska that it was detected in the air beyond six feet from sick people. The White House subsequently recommended that people wear masks in public.
Chen also noted research showing that the coronavirus can be found in feces and wonders if the violent flush of airplane toilets could cause the virus to become airborne. The outbreak on the Diamond Princess cruise ship provides added concern. Though researchers found evidence the virus spread via contacts and droplets, Chen wonders if air systems may also have contributed.
What he has established through years of research is that droplets less than a certain size - about 20 microns, or millionths of a meter - can easily be swept through the air. Following a cough, they disperse widely, and it takes about four minutes on average for such particles to be filtered out of the cabin, his research shows.
Slightly larger droplets can still float for 10 seconds in the cabin's blowing air, or be carried in the wake of someone walking, he said.
Larger, heavier droplets that settle quickly remain the most serious risk with covid-19, he said.
But any person deciding whether to fly, now or in the future, should be vigilant, and wear a mask, he said.
"I personally believe we just provide all possible routes of transmission and make people aware of the risks," Chen said.
On Monday, JetBlue said it would require all passengers to wear masks starting May 4, and American Airlines said it would start handing out face coverings on some flights. "This is the new flying etiquette," said JetBlue president Joanna Geraghty.
For a manufacturer such as Boeing, the choices were difficult, Chen said, including deciding how much to upend current designs in search of progress.
"I don't mean Boeing does not want to change. . . . They definitely think this is a problem they need to solve," Chen said, noting the years the company spent sponsoring his research. But changing an aircraft is a major undertaking.
"The technology we developed is not bulletproof. It's much better than the current one - but not bulletproof," Chen said. "Is that worth spending money to change all the systems?"
Boeing would not comment on the SARS transmission risks outlined by Chen and its engineers and said its product strategies are proprietary. It said research on disease transmission "is and continues to be a high priority."
The company said it tested its "Fresh Lav," which uses automated ultraviolet lights to kill 99.9 percent of germs in lavatories "after every passenger use," on a specially outfitted Boeing 777 last year.
Airbus, facing the same challenges, would not say if alternative ventilation systems may be introduced. Current cabin configurations reflect what is considered "most effective and achievable," the company said in a statement.
- - -
Brenner and his team at Columbia have begun testing their special 3-by-3-inch lights against the novel coronavirus, which has killed more than 210,000 people worldwide.
One colleague, physicist David Welch, is calibrating exposure from the lamps. Another, Manuela Buonanno, is testing how many viruses survive.
They killed two batches of less-threatening coronaviruses in recent weeks with a very low level of exposure, according to research released Monday, and Brenner said things are looking good so far with the one that causes covid-19.
Brenner believes the technology could address the problem of a virus-spewing airplane passenger.
"You're sitting there and the guy right behind you sneezes. . . . All the filters in the world aren't going to help you," Brenner said. "I think the lamp would potentially deal with that."
What takes longer is conclusively proving the long-term safety for people exposed to the light, a type of radiation known technically as far-UVC light. Traditional ultraviolet lights are used to clean water supplies and sanitize operating rooms, but only when no people are under them, because they can cause cancer and eye damage.
But Brenner said far-UVC light is different. It is generated at the far end of the ultraviolet spectrum and is "not in any way penetrating." It barrels right through air or droplets containing viruses, but is quickly absorbed - and stopped - by the protein in dead skin cells or in the layer of tears on the surface of the eyes, Brenner said.
"In order to produce any damage, it has to get to the live cells," he said. "Not once have we yet seen any hazardous effects."
Could the light somehow sneak through and cause damage?
"I think the answer's no. That's why we have all those mice sitting upstairs in this long-term study," Brenner said.
Brenner has studied the effects of CT scans, nuclear plants and dirty bombs. He and his colleagues developed a commonly used radiation protocol for fighting prostate cancer. He initially thought of the lamps as a weapon against infections during surgeries, after a good friend died of one.
There are challenges, including the need for a huge supply of the lightbulbs, though he said industry is already ramping up for that. The lights are mainly for targeting viruses in the air, since they might hide in the tiny shadows of upholstery, he said. A closed tray table would also be a problem.
Boeing said its lavatory prototype uses far-UVC lights, which are "not harmful to humans." The company said it could not speak directly to the Columbia research but is pursuing creative "cabin upgrades."
Airbus declined to say if it would consider the lights, and the FAA would not say if it supports the idea.
The Columbia researchers have been palpating the skin of the mice and looking for lesions on their corneas. Some of the males are getting a bit antsy from being cooped up, but otherwise are looking good, Brenner said. At the end of the study, researchers will check them for any subtle DNA damage.
As scientists scurry to develop vaccines, viruses traverse the globe along with people, moving “from city to city. They go by train or plane or buses,” Brenner said. “If you can target transportation modes in some way, you could do a lot of good.”