In this episode of Stuff You Should Know, Josh Clark and Chuck Bryant examine the 1979 Three Mile Island nuclear accident, America's worst commercial nuclear power disaster. They trace how a simple pump failure escalated into a partial meltdown through a cascade of human errors, inadequate training, poor equipment design, and regulatory failures. The hosts detail how operators misread critical indicators and made decisions that exposed the reactor core for hours, creating conditions for a hydrogen explosion.
Beyond the technical failures, Clark and Bryant explore the broader implications: the challenges of assessing actual radiation exposure and health impacts, the erosion of public trust through misinformation and contradictory official statements, and the accident's lasting impact on nuclear energy development in the United States. The episode also covers the decades-long cleanup effort and the recent decision to restart the facility to power AI data centers, illustrating how public perception of nuclear energy continues to evolve.

Sign up for Shortform to access the whole episode summary along with additional materials like counterarguments and context.
The Three Mile Island nuclear accident began on March 28, 1979, when a water pump failed in the secondary cooling system, preventing adequate heat removal from the reactor core. The control system functioned correctly, automatically inserting control rods to shut down the nuclear reaction. However, a pilot-operated relief valve stuck open to relieve pressure, and inadequate indicators left operators unaware that coolant was escaping.
As emergency systems engaged, operators misread the situation. Seeing sustained pressure readings, they incorrectly assumed the system was overfilling rather than dangerously low on coolant. They shut off the emergency core cooling pumps—a disastrous decision that stopped critical coolant inflow while the open valve continued venting coolant. For nearly two hours, the reactor core remained partially exposed, with superheated zirconium cladding reacting with steam to produce hydrogen gas, creating an explosive threat.
Once the reactor reached cold shutdown, cleanup began. Camera footage in the 1980s revealed that molten uranium had pooled at the bottom of the containment vessel in a partial meltdown. Remarkably, the containment vessel held intact, preventing catastrophic radiation release. The initial decontamination phase ran until 1990, removing 150 tons of radioactive material and processing 2.25 million gallons of contaminated water. Nearly half a century later, about 1% of nuclear fuel remains, with cleanup continuing.
The accident resulted from a dangerous combination of human error, inadequate training, poor design, and regulatory shortcomings. Operators were trained primarily for routine operations and unprepared for emergencies. When violent boiling caused coolant pumps to shake, operators mistook this as overfilling rather than recognizing it as a loss-of-coolant emergency. A misdesigned indicator light showed when a closure command was issued—not whether the valve was actually closed—leading operators to believe the stuck valve was shut.
The control room lacked direct water level measurements, forcing operators to guess coolant amounts from pressure readings. A malfunctioning printer delayed reactor data by 90 minutes during the crisis, and physical obstacles—including one operator's large belly blocking crucial panel indicators—further impeded decision-making.
The Nuclear Regulatory Commission assumed serious accidents were impossible and focused training on normal operations rather than emergencies. Their confusing emergency procedures failed to help operators prioritize safety systems effectively. Metropolitan Edison (Met Ed) maintained they had trained staff to NRC specifications, exposing the inadequacy of regulatory standards.
The nuclear industry culture fostered dangerous overconfidence. The problematic relief valve had malfunctioned for weeks before the accident, but operators falsified maintenance records rather than fixing it. Throughout the crisis, officials repeatedly downplayed severity, only fully acknowledging the disaster's scale years later.
Josh Clark and Chuck Bryant explain that extensive research by the EPA, NRC, state authorities, and NGOs reached consensus that radioactive release was insufficient to significantly affect public health or the environment. Bryant notes that official estimates put public radiation exposure at about the level of a chest X-ray. Clark adds that the highest cleanup worker exposure in 1989 was about a tenth of a chest X-ray dose, and onsite radiation levels were comparable to one hour of air travel.
However, residents reported symptoms, and some studies found correlations between cancer rates and proximity to Three Mile Island. Bryant points out a 1997 study finding increased cancer rates and a 2017 study correlating thyroid cancer with residence near the site. Clark highlights the challenge: researchers couldn't definitively link radiation to illness due to background cancer rates and placebo/nocebo effects. Many reported symptoms like vomiting and hair loss immediately after the accident, but these accounts are often dismissed as psychological rather than direct radiation effects.
To manage the hydrogen bubble, operators deliberately vented radioactive gas throughout the 1980s, with the worst exposures in 1982 and 1989. Clark and Bryant note this presented a moral dilemma: venting was the lesser evil compared to risking catastrophic containment failure.
Metropolitan Edison lost credibility by making false safety claims as conditions worsened. They falsified leak reports and issued contradictory public statements, insisting no radiation had escaped even as evidence showed otherwise. Their pattern of deception caused officials to sideline them from crisis management.
When Met Ed tried to notify the NRC, they found the office closed and had to use an answering service, taking 40 minutes to make contact. This delay hindered crucial emergency decision-making while untrained operators awaited federal guidance. News spread before officials had reliable information, amplifying speculation and fear.
On March 30th, after confirming escaped radioactive gas, Governor Thornburg ordered evacuation of pregnant women and preschoolers—the first public acknowledgment of real danger. About 140,000 people fled, though many refused to leave due to distrust and attachment to their homes. The panicked evacuation showed how deception and delays had undermined expert reassurance, making it impossible for officials to communicate risks credibly.
President Jimmy Carter toured the facility despite conflicting assessments about hydrogen bubble explosion risk. He and First Lady Rosalynn Carter wore yellow rubber boots due to contaminated radioactive water, revealing lingering hazards even as officials claimed control. The President's visit symbolically projected confidence, calming public anxieties in ways official statements could not.
The accident triggered a dramatic shift in U.S. nuclear policy. From 1980 to 1984, 51 planned reactor projects were canceled, with no new orders placed by 1999—a trend continuing decades later. Remarkably, just 12 days before the real accident, the film "The China Syndrome" was released, depicting a fictional nuclear meltdown that intensified public anxiety when reality mirrored fiction.
The accident sparked unprecedented anti-nuclear activism, with protests featuring signs like "You try to kill us, then bill us." High-profile figures like Jane Fonda and Tom Hayden advocated against nuclear energy, and Bruce Springsteen's "No Nukes" concert helped form Musicians United for Safe Energy. Despite minimal direct radiation exposure, the psychological impact halted nuclear development for generations.
Met Ed faced extensive legal fallout. In 1983, they were indicted for falsifying leak tests and fined $45,000 (about $140,000 today). They contributed $1 million (about $3.1 million today) to Pennsylvania's environmental fund. By 1996, a federal judge ruled against plaintiffs in radiation exposure lawsuits, citing inability to directly link cancer cases to the accident.
Cleanup costs fell on ratepayers, prompting resident protests. A billion-dollar cleanup fund was established, much from ratepayers, and continues financing decontamination. The three-decade cleanup revealed societal costs surpassing fines and settlements, with ongoing expenses and monitoring persisting after legal resolution.
In a striking reversal, Microsoft recently signed a 20-year contract to purchase electricity from TMI Unit 1 starting in 2028 for AI data centers—a full-circle moment where America's worst nuclear disaster site will power cutting-edge technology. As climate concerns grow, nuclear power is regaining appeal as a carbon-free energy source, demonstrating that public attitudes can shift even after Three Mile Island's historical impact.
1-Page Summary
The Three Mile Island nuclear accident began on March 28, 1979, with a small malfunction in the secondary cooling system. The precise cause, whether mechanical or electrical, remains uncertain. The water pump failed to send the necessary cooling water to the steam generators, initiating the crisis by preventing adequate heat removal from the reactor core.
As the reactor's temperature rose due to lack of coolant, the control system functioned as intended. It detected the overheating and automatically inserted control rods, shutting down the nuclear chain reaction to prevent further escalation. This initial safety measure operated correctly.
To relieve pressure in the reactor system, a pilot-operated relief valve opened—but failed to close as designed. The valve stuck open, but due to inadequate status indicators, operators were unaware it remained open. As a result, coolant continued to escape from the reactor system unnoticed.
As emergency protocols engaged, the plant's systems tried to compensate by allowing emergency coolant water to flow into the reactor. However, operators misread the situation. Seeing sustained pressure and water flows, they incorrectly assumed the system was at risk of overfilling, not realizing coolant levels were dangerously low.
Trusting these flawed interpretations, the operators turned off the emergency core cooling pumps. This decision proved disastrous, as it stopped the inflow of coolant critical to maintaining safe temperatures, while coolant continued leaking out the open relief valve.
This cascade of errors led to a dramatic drop in water level. The reactor core, supposed to remain submerged at all times, became partially exposed—the top of the core was out of water. For nearly two hours, plant operators did not realize the severity of the situation, and the exposed core continued to overheat.
With the reactor core exposed, the superheated zirconium cladding on the fuel rods reacted with steam, causing water molecules to split and forming hydrogen gas. This raised the dire risk of hydrogen gas accumulating to explosive levels inside the reactor vessel, threatening an even greater catastrophe.
Once the accident was ...
Three Mile Island: Sequence, Mechanical Failures, Technical Causes
The Three Mile Island accident resulted from a hazardous combination of human error, inadequate operator training, poor system design, regulatory shortcomings, and overconfidence within the nuclear industry itself.
Operators at Three Mile Island were trained primarily for standard operations, such as following routine procedures and pushing buttons, but they were not adequately prepared for emergencies or unexpected system behaviors. When the emergency unfolded, they misinterpreted crucial data, making catastrophic decisions. For example, violent boiling in the reactor caused coolant pumps to shake, but the operators did not realize this was due to a loss of coolant. Instead, they mistook it as a sign of overfilling and feared that too much water was causing pump vibrations. The operators did not recognize that such boiling and vibrations indicated something was seriously wrong, demonstrating a critical gap in their emergency training.
Another pivotal failure resulted from a misdesigned indicator light on the automatic release valve. The control room light indicated when a closure command had been issued—not whether the valve was actually closed. Operators took the light’s activation as confirmation that the valve was closed, failing to realize it remained stuck open and continuously vented coolant, worsening the crisis.
The control room at Three Mile Island suffered from confusing instrumentation and missing critical data. Operators did not have any direct measurements of reactor water levels. Instead, they guessed at the amount of coolant based on indirect pressure readings, leading them to believe the system was overfilled when it was desperately low. This lack of direct information contributed to harmful decisions, such as shutting off water when more was urgently needed.
During the emergency, a malfunctioning printer failed to provide real-time data from the reactor for 90 minutes. Operators had to rely on outdated information as the crisis unfolded, leaving them effectively blind to the reactor’s deteriorating condition and unable to respond properly.
Physical obstacles also impeded decision-making. Notably, one operator’s large belly blocked the view of important panel indicators, meaning essential information was literally obscured during critical moments of the accident.
The Nuclear Regulatory Commission (NRC) did not impose adequate safety standards for emergencies, assuming serious accidents were impossible. Their training requirements prepared operators only for normal conditions, leaving both operators and the NRC itself unprepared for emergen ...
Failures: Human Error, Lack of Training, Poor Design, Regulatory Issues
Josh Clark and Chuck Bryant explain that Three Mile Island has been intensely studied, with investigations by the EPA, NRC, state authorities, the Union of Concerned Scientists, and various NGOs. Almost all research and meta-analyses reach the consensus that the radioactive release was insufficient to affect public health or the environment significantly.
Bryant notes that official claims estimated the radiation exposure for the public to be about the level of a chest X-ray. Clark adds further detail, referencing data from cleanup operations: the highest exposure, recorded during the cleanup at its peak in 1989, was 0.98 rems—about a tenth the dose of a chest X-ray.
Onsite readings at the reactor, according to Clark, showed environmental radiation levels comparable to what a person receives from a single hour flying on an airplane. He and Bryant both emphasize that the actual health risk from the accident was far less than everyday exposures such as air travel.
Clark emphasizes that extensive studies, including meta-analyses, did not establish any statistically significant correlation between the exposure and adverse health effects. The consensus across organizations and studies remains that the radiation released was not enough to cause noticeable harm.
Bryant points out a 1997 study finding increased cancer rates in the area and a 2017 study reporting a correlation between thyroid cancer and residence near Three Mile Island. Also cited is a 43% increase in infant deaths around the accident period. Despite such findings, researchers could not conclusively prove a direct causative link to radiation exposure from the accident.
Clark highlights the challenge: most health studies are unable to link the radiation from Three Mile Island to any illness, as prevailing cancer rates and phenomena like the placebo or nocebo effect muddy the data. Many people reported symptoms such as vomiting, nausea, hair loss, and rashes right after the accident. However, over time these accounts are often dismissed as psychological effects rather than direct effects of radiation.
Clark underscores the emotional toll for locals—if they believe lov ...
Radiation Release and Public Health: Actual Exposure vs. Perception and Reported Effects
The Three Mile Island disaster was marked by confusion, a breakdown in trust, and widespread fear fueled by secrecy, misinformation, and delayed communication.
From the outset, Metropolitan Edison (Met Ed), the power company overseeing Three Mile Island, lied about the unfolding crisis. They repeatedly offered false assurances about the safety of the reactors both to the public and even in private to Pennsylvania Governor Dick Thornburg and Lieutenant Governor William Scranton. Met Ed denied the seriousness of the crisis and claimed there were no significant issues, contradicting evidence as conditions deteriorated. Their pattern of deception quickly became evident, causing officials to sideline Met Ed from crisis management due to their lost credibility.
In the midst of chaotic developments, Met Ed issued public statements that minimized the threat, insisting that no radiation had escaped and everything was fully contained. However, on March 30th, new reports revealed that radioactive gas had indeed escaped from the reactor, directly contradicting the company's earlier claims and intensifying public anxiety.
When Met Ed tried to notify the Nuclear Regulatory Commission (NRC) about the emergency, they discovered the NRC’s office was closed. Operators had to rely on an old-fashioned answering service to relay urgent messages—a slow, inefficient process. Attempts to reach the regional duty officer were hampered as he had already left for work and was unreachable.
It took nearly 40 minutes to finally make contact with the NRC. During this critical window, the plant operators, untrained for such a crisis, were left in limbo awaiting essential guidance from federal authorities. This delay obstructed crucial decision-making at a moment when every second mattered.
News of the incident spread quickly—even before complete, reliable information was available. Journalists arrived on the scene, word of mouth amplified uncertainty, and rumors outpaced official communication. By 9 a.m., everyone knew a major problem had occurred, but nobody outside the plant knew how serious it actually was, heightening speculation, confusion, and public fear.
After finally receiving confirmation of escaped radioactive gas, Governor Thornburg issued an evacuation order on March 30th for all pregnant women and preschool children. This was the first public acknowledgment of real danger and signaled that risks may be greater than previously disclosed, causing widespread alarm.
The announcement led to panic and mass exodus, with about 140,000 people fleeing the area. Despite the chaos, many residents refused to leave—citing distrust of authorities and deep attachments to their homes, vowing not to be forced out no matter the warning.
Panic and Evacuation: Misinformation, Secrecy, and Exodus
The Three Mile Island (TMI) accident in March 1979 triggered a dramatic shift in public opinion and policy around nuclear energy in the United States. Previously, nuclear power had been expanding steadily, but after the incident, the industry nose-dived. From 1980 to 1984, 51 planned reactor projects in the U.S. were canceled. By 1999, there had still been no new orders for nuclear plants, a trend that continues decades after the accident. The fear surrounding nuclear power wasn’t isolated to the U.S.—there were global repercussions that stymied its growth.
Remarkably, just 12 days before the Three Mile Island meltdown, the film "The China Syndrome" was released. Starring Michael Douglas and Jane Fonda, the movie depicted a scenario eerily similar to the real accident—a fictional nuclear meltdown. The term "China syndrome" refers to the (unfounded) fear that a reactor could melt through the earth to the other side of the planet. The film’s timing intensified public anxiety when the real accident occurred soon after.
The accident sparked an unprecedented wave of anti-nuclear activism in the U.S. Protests surged, with demonstrators brandishing signs like "You try to kill us, then bill us," referencing not only the health and safety fears, but also the financial burden imposed on citizens for the cleanup. High-profile figures, such as Jane Fonda and Tom Hayden, publicly advocated against nuclear energy. The activism coalesced with major cultural moments—such as Bruce Springsteen’s "No Nukes" concert in 1979, which helped form the Musicians United for Safe Energy (MUSE).
Despite studies showing minimal direct radiation exposure from the accident, the psychological impact of Three Mile Island was profound and enduring. Fear and mistrust effectively halted nuclear development for generations, making the accident the primary reason nuclear energy didn’t become widespread in the U.S.
Metropolitan Edison (Met Ed), the company operating TMI, faced extensive legal and regulatory fallout.
In 1983, Met Ed was indicted by the Department of Justice for falsifying leak test results to federal authorities. The company was fined $45,000—a figure that, adjusted for inflation, would be about $140,000 today.
Additionally, Met Ed contributed $1 million (about $3.1 million today) as part of a fund directed to the Pennsylvania EPA for environmental protection and monitoring in the region.
The company also settled various civil and personal injury lawsuits over alleged radiation exposure. By 1996, with thousands of cases pending, a federal judge selected ten as a test group. The court ultimately ruled against the plaintiffs, citing an inability to directly link cancer cases to radiation from the accident.
The enormous cost of the cleanup fell not only on the company but also on local residents and ratepayers.
Community resentment grew as Metropolitan Edison passed cleanup expenses onto their electricity customers, prompting residents to protest, including signs saying, "you try to kill us, then bill us."
A cleanup fund in excess of one billion dollars was established, much of it sourced from ratepayers, and continues ...
Long-Term Effects: Cleanup, Lawsuits, Settlements, Decline of Nuclear Energy in the U.S
Download the Shortform Chrome extension for your browser
