Three Mile Island and lessons in crisis communication

At 10:00am on 29 March 1979, John “Jack” Herbein, vice president of Metropolitan Edison (Met-Ed), prepared to address a room packed wall-to-wall with reporters and cameras at the Hershey Convention Center in south-central Pennsylvania. It would be the first time Met-Ed officials faced the public directly since what the US Nuclear Regulatory Commission (NRC) has called the “most serious accident in U.S. commercial nuclear power plant operating history” began unfolding the day before at Three Mile Island (TMI), the now-infamous nuclear power plant owned and operated by the utility company.

Although he had no prior experience in public relations, Herbein took on, at least initially, sole responsibility for translating the complexity of nuclear industry jargon to an unnerved public largely unaware of the inner workings of TMI. Herbein nervously scanned the crowd, avoiding direct eye contact as much as possible. He appeared uncomfortable in the spotlight as he opened with a brief statement . Then reporters hurled one question after another. Were there safety problems? What threats did the radiation pose? How long had the core been uncovered? Herbein stammered through variations of the same answer: We don’t know.

The 1979 partial meltdown at TMI’s Unit 2 reactor has since been largely attributed to human rather than technical errors. There were no lives lost, no reported injuries, and no significant structural damage except to the plant itself. Estimated levels of public exposure to radiation were about that of a chest x-ray exam , according to the NRC. However, the accident was life changing for many who lived in the immediate vicinity, especially for the estimated 76 000 people within 10 miles of the plant who chose to evacuate on their own terms—my mother, aunt, and grandparents among them. A sense of unease still looms over the Harrisburg region more than 40 years later, weighing on a community that received often misleading and contradictory information from the very officials, scientists, and leaders who claimed to have everything under control.

Today, with public trust in scientists and government officials during a crisis again an issue of concern, the story of the mismanagement of public relations and information flow during the TMI partial meltdown is as relevant as ever.

Reactor failure, communications breakdown

The trouble at TMI began at 4:00am on 28 March when several water pumps malfunctioned in the Unit 2 reactor, stopping the flow of water from the steam generators to the turbine (see Physics Today, June 1979, page 77 ). With the feedwater flow stopped, the temperature of the reactor coolant increased and the rapidly heating water expanded. Pressure quickly rose beyond normal operating levels, so a relief valve opened to relieve the excess pressure. However, pressure continued to rise, so TMI-2’s reactor “scrammed"—control rods were dropped into the reactor core to halt the nuclear fission process. With the reaction stopped, pressure decreased to normal operating levels, at which point the valve should have closed. Although the light on the control room panel indicated that the electric power to open the valve had gone off—which led the operators to assume the valve had shut—it had not. The valve was stuck open, and it remained so for the next two hours, which allowed precious coolant water to drain from the reactor.

Unaware that the valve was open, operators took manual control of the water system because they feared the core had too much water and they would lose control of pressure levels. Instruments were sending inaccurate readings to the control room. Craig Faust, one of the control room operators at the time of the accident, later told the President’s Commission on the Accident at Three Mile Island that alarms were going off but “not giving us any useful information.” By around 6:30am, technicians were monitoring increasing radiation levels. A site emergency was declared around 7:00am. Crucially, that morning and multiple times over the next several days, TMI technicians deliberately released radioactive noble gases into the air to relieve pressure within the containment building.

News of the malfunction broke on WKBO, a local top 40 music station, during its 8:25am newscast. The Need for Change: The Legacy of TMI, the official report of the president’s commission, reconstructed the events as follows: A half-hour earlier, the station’s traffic reporter had heard over his CB radio that police and firefighters were mobilizing in Middletown, where TMI is located. He had noticed too that no steam was escaping from the cooling towers. So he contacted Mike Pintek, WKBO’s news director, who then called TMI directly. For some reason, Pintek was connected directly to a control room operator, who said, “I can’t talk now; we’ve got a problem.” Pintek was then put in touch with a Met-Ed communications manager based about 70 kilometers away in Reading, Pennsylvania, who told him, “There was a problem with a feedwater pump. The plant is shut down. We’re working on it. There’s no danger off-site. No danger to the general public.”

In the days that followed, communication from Met-Ed was often unreliable and generally incomprehensible to the media and the public. “There was a lot of questions and a lot of vague, nebulous kind of evaluations of what had happened, and what was going to continue to happen,” said David Solleneberger, a local reporter at the time, on a 2019 radio program . “It was a lot of having to rely on people who either could not explain it in plain language or just didn’t know what was going on.” In the March 2019 podcast series “Three Mile Island: As It Happened” from the PA Post, a local resident shared a passage from the audio diary he kept during the accident: “They get into discussions of millirems, megarems, and rems, and rems per hour, rems per incident, rems per week, rems per year, millirems per year. . . . How the hell they expect anybody to understand that is beyond me.”

Met-Ed generally downplayed the severity of the situation and often appeared dismissive of the media’s concerns. According to The Need for Change, at one point Herbein snapped at reporters, saying, “I don’t know why we need to tell you each and every thing that we do specifically.” The report deemed that remark the one that eliminated any credibility that Herbein and Met-Ed had left with the press and public at large. At another press conference two days later, Herbein addressed the media directly and said, “One of the things that the people that live around the plant have to recognize is that we have to get on with our jobs. Keeping each and every person informed is a very difficult thing to do.”

Radiation release

When the public learned that detectable amounts of radiation were intentionally leaked into the air, the information came not from Met-Ed itself, but in a press release from then lieutenant governor William Scranton. On 28 March at 10:55am, about seven hours after the initial malfunction inside the plant, a press release from Scranton’s office read: “The Metropolitan Edison company has informed us that there has been an incident at Three Mile Island, Unit #2. Everything is under control. There is and was no danger to public health and safety.”

However, in a 2:30 meeting that afternoon, Scranton learned that Herbein hadn’t told reporters in earlier interviews about the radiation releases because, as The Need for Change later chronicled, “it didn’t come up.” At 4:30pm, following the meeting, Scranton’s office released a new statement : “This situation is more complex than the company first led us to believe. . . . Metropolitan Edison has given you and us conflicting information.”

Additionally, as Scranton’s new statement put it, Met-Ed did not notify the Pennsylvania Department of Environmental Resources (DER) of the radiation release “until about the time that it was halted.” The company then “promised” to notify government agencies in the event of further discharges. In fact, DER’s field-monitoring team subsequently measured low levels of radioactivity offsite. When asked by a reporter at the 29 March press conference why radiation was allowed to escape into the atmosphere, Herbein matter-of-factly responded, “Well, radiation is allowed to be released from the reactor plant in the course of normal operation. That’s done on a daily basis at low-level gas release or low-level liquid release into the Susquehanna [River].”

As the crisis unfolded in the coming days, there was extensive speculation about how serious the accident was or would become, causing disagreements among local and federal officials about evacuation necessity and scope. As The Need for Change outlined, it wasn’t until Friday, 30 March—more than 48 hours after the accident began—that then governor Richard Thornburgh finally issued a recommendation for pregnant women and preschool-age children within five miles of TMI to leave the area. Several NRC officials also recommended immediate evacuation that day, but they did so on the basis of a “mistaken interpretation of the release of a burst of radiation,” according to the president’s commission. The miscommunication surrounding evacuation orders led the commission to conclude that the most serious health effect of the accident was “severe mental stress.”

Despite some ensuing clarity from government officials such as the NRC’s Harold Denton , many residents today still cite the lack of clear communication as one of the most painful aspects of the accident. In several of the personal stories from the 2019 “I Remember TMI” project—a collaboration between public radio broadcaster WITF and the PA Post—locals mention mistrusting officials because of spotty and inconsistent communication. “The thing that remains with me most to this day,” one resident wrote , “is how little we all knew, how unfamiliar we all were with nuclear power, and how in a time of crisis, the value of trust in our elected and appointed leaders matters as much as anything else.” Another local recalled , “No one within my earshot, at least, was talking about anything other than TMI, and whether or not we could or should trust anybody to tell us the truth.”

Lessons for today

It is difficult for me to reconcile nuclear energy’s potential for destruction, as nearly demonstrated by the partial meltdown at TMI, with the fact that I am a direct beneficiary of the nuclear power industry’s success. My father started as a security guard at the plant several months after the accident; he went on to earn radiation safety and instrumentation controls certifications and became an instrumentation and controls technician over the course of his 30-plus years there. Until its closure in September 2019, the plant provided my family and many others with clean, reliable electricity and—perhaps more important—livelihoods, for decades. Londonderry Township and Dauphin County rely on the significant contributions to local property taxes made by Exelon, which bought TMI-1 in 2000, as well as on income taxes paid by TMI employees to help support schools and public services. Now that the decommissioning process for TMI has begun, nearly 675 full-time jobs tied to about $60 million in wages per year will be gradually eliminated . The local economic impact, I believe, should not be overlooked in broader debates about nuclear energy.

What can we learn from TMI about effective communication during a public health crisis? First, amplify one consistent message with many voices. Margaret Reilly, a health physicist at the Pennsylvania Bureau of Radiation Protection at the time of the accident, offered these recommendations : plan your initial response, “protect your information flow pattern described in the plan, . . . get about twice as much communications capability as you think you’ll need, and use it regularly.” That strategy is difficult to accomplish when there are multiple competing streams of information, as was the case in the days after the TMI accident, during which private, municipal, state, and federal agencies were simultaneously offering conflicting updates and recommendations. Today, given our instant access to an enormous network of information sources, amplifying a consistent message is even more critical to reducing the spread of misinformation.

Second, scientific experts and other officials should offer full transparency regarding both knowns and unknowns. Although it may seem as if communicating uncertainty would diminish perceived scientific authority, a recent study in Proceedings of the National Academy of Sciences finds that doing so has little effect, in the eye of the public, on the trustworthiness of the information source. The study notes that more research is needed on the topic. But the findings suggest that communicating uncertainties around the evidence can provide important context without hurting public confidence.

Finally, it’s important to acknowledge people’s justified fears and anxieties during a public health threat. Jessica Wieder, director of the Center for Radiation Information and Outreach at the Environmental Protection Agency, has pointed out that “the best tools for reducing high stress, doubt, and fear are compassion, validation, and commitment: compassion for the audience as individuals, validation of their feelings, and commitment to their cause.” Met-Ed’s frequent dismissal of public concerns significantly damaged its credibility and, consequently, caused feelings of doubt and a lowering of public confidence in the nuclear industry that have lingered for more than four decades.

The events at TMI remain an important case study for anyone in crisis communications and management, and they demonstrate the need to take public perception and understanding of scientific endeavors very seriously. As we navigate the COVID-19 pandemic—another uncertain, invisible threat to public health—we are reminded that accurate and consistent communication of risk during a crisis is intimately tied to maintaining the confidence of the general public in the institutions designed to protect them. As we see scientific institutions doubted and their credibility challenged today, building trust with the broader public through effective communication and outreach is of heightened importance.