US Nuclear Agency Hid Concerns, Hailed Safety Record as Fukushima Melted
March 11, 2014
Bill Dedman / NBC News
In the tense days after a powerful earthquake and tsunami crippled the Fukushima Daiichi power plant in Japan on March 11, 2011, staff at the US Nuclear Regulatory Commission made a concerted effort to play down the risk of earthquakes and tsunamis to America's aging nuclear plants, according to thousands of internal emails reviewed by NBC News.
(March 10, 2014) -- In the tense days after a powerful earthquake and tsunami crippled the Fukushima Daiichi power plant in Japan on March 11, 2011, staff at the US Nuclear Regulatory Commission made a concerted effort to play down the risk of earthquakes and tsunamis to America's aging nuclear plants, according to thousands of internal emails reviewed by NBC News.
The emails, obtained via the Freedom of Information Act, show that the campaign to reassure the public about America's nuclear industry came as the agency's own experts were questioning US safety standards and scrambling to determine whether new rules were needed to ensure that the meltdown occurring at the Japanese plant could not occur here.
At the end of that long first weekend of the crisis three years ago, NRC Public Affairs Director Eliot Brenner thanked his staff for sticking to the talking points that the team had been distributing to senior officials and the public.
"While we know more than these say," Brenner wrote, "we're sticking to this story for now."
There are numerous examples in the emails of apparent misdirection or concealment in the initial weeks after the Japanese plant was devastated by a 9.0 earthquake and 50-foot tsunami that knocked out power and cooling systems at the six-reactor plant, eventually causing releases of radioactive material:
* Trying to distance the US agency from the Japanese crisis, an NRC manager told staff to hide from reporters the presence of Japanese engineers in the NRC's operations center in Maryland.
* If asked whether the Diablo Canyon Power Plant on the California coast could withstand the same size tsunami that had hit Japan, spokespeople were told not to reveal that NRC scientists were still studying that question. As for whether Diablo could survive an earthquake of the same magnitude, "We're not so sure about, but again we are not talking about that," said one email.
* When skeptical news articles appeared, the NRC dissuaded news organizations from using the NRC's own data on earthquake risks at US nuclear plants, including the Indian Point Energy Center near New York City.
* And when asked to help reporters explain what would happen during the worst-case scenario -- a nuclear meltdown -- the agency declined to address the questions.
As the third anniversary of Fukushima on Tuesday approaches, the emails pull back the curtain on the agency’s efforts to protect the industry it is supposed to regulate. The NRC officials didn't lie, but they didn't always tell the whole truth either. When someone asked about a topic that might reflect negatively on the industry, they changed the subject.
NBC News requested in late March 2011 all of the emails sent and received by certain NRC staffers during the first week of the crisis. Other news organizations and watchdogs filed similar requests. The NRC has now been posting thousands of emails in its public reading room over the past two years.
[See details of the 62 US nuclear power plants, along with their age and safety records.]
The NRC declined to discuss specific emails or communications. But Brenner provided an emailed statement: "The NRC Office of Public Affairs strives to be as open and transparent as possible, providing the public accurate information in the proper context. We take our communication mission seriously. We did then and we do now.
The frustration displayed in the chosen e-mails reflects more on the extreme stress our team was under at the time to assure accuracy in a context in which information from Japan was scarce to nonexistent. These e-mails fall well short of an accurate picture of our communications with the American public immediately after the event and during the past three years."
Dating back to the Three Mile Island nuclear crisis in 1979, many nuclear watchdogs and critics have said that the NRC acts first to protect the industry, and its own reputation. One critic said these emails solidify that perception.
"The NRC knew a lot more about what was going on than it wanted to tell the American people," said Edwin Lyman, senior scientist at the nuclear watchdog group Union of Concerned Scientists and co-author of the new book Fukushima: The Story of a Nuclear Disaster, which relied on some of the same emails. "They immediately put out information that implied that US reactors were in a better position to withstand Fukushima type events than Fukushima reactors were, but it was clear that the what the NRC knew internally was not nearly as positive."
'We All Need to Say a Prayer'
From the earliest hours of the crisis, the emails among NRC staff show deep concern about the developing crisis in Japan, particularly among the technical experts.
The first word that the powerful earthquake and tsunami waves had devastated the Fukushima plant came early morning (Eastern time) on March 11, 2011. Throughout the day, staff at NRC headquarters in Rockville, Md., struggled to learn what was going on in Japan. The chief of the NRC Component Integrity Branch, senior engineer David Rudland, was asked by a colleague if he had any new information. [The emails excerpted in this article are shown in full in a PDF file.]
From: Rudland, David
Date: Friday, March 11, 2011, 10:54 AM
No, at this point all we know is that they are struggling to shut down the plant.
We all need to say a prayer. . . .
By that afternoon, the news was worse. An officer in NRC research passed on to his colleagues a status update from the Japanese electrical company.
From: Nosek, Andrew
Date: Friday, March 11, 2011, 4:46 PM
There was a triple SBO.
SBO is nuclear jargon for a station blackout. The earthquake had cut electrical power to the plant, and the tsunami had damaged the backup diesel generators.
NRC operations officer Daniel Mills had an emotional reaction:
From: Mills, Daniel (NRC operations officer)
Date: Friday, March 11, 2011, 4:49 PM
BBC is reporting radiation levels at reactor are 1000x normal. I feel like crying.
The NRC staff recognized immediately the public-relations nightmare that Fukushima presented for nuclear power in the United States. More than 30 of America's 100 nuclear power reactors have the same brand of General Electric reactors or containment system used in Fukushima.
American nuclear reactors are well into middle age. The median age of an operating reactor in the US is 34 years, placing start-up in midst of the Carter administration. The oldest -- the Ginna plant near Rochester, N.Y. -- was licensed in 1969, the year Neil Armstrong walked on the moon. Only four of the 100 reactors have begun generating power since 1990. The newest, at Watts Bar in Tennessee, was licensed in 1996, when many of this year's high school seniors were born.
The unfolding disaster in Japan triggered immediate alarm inside the NRC about plans to announce regulatory actions. Seeing the video from Japan, NRC engineer Richard Barkley pointed out that the NRC staff that week to recommend extending for 20 years the license for reactors a nuclear power plant in New England called Vermont Yankee.
He warned colleagues, "That was a very scary picture to myself, much less the public, especially since the machine is a GE designed BWR (boiled-water reactor) not radically different in size, age and design than some high visibility plants in my region. I can see the cards and letters coming to my in-box by Monday." (Ultimately, the NRC delayed the Vermont Yankee re-licensing only briefly, approving it on March 21. This year the plant's owner plans to close it, a victim of the competition from falling prices for natural gas.)
Three decades after the partial meltdown at the Three Mile Island nuclear plant in Pennsylvania, nuclear power companies saw hope for a renaissance, with the first new reactors in years being planned. But public opinion was fragile: If the Fukushima reactors, built by American companies, could be overwhelmed by natural disasters, could the public trust that American power plants were safe?
'We Are Not Talking about That'
In the NRC's Office of Public Affairs, the first talking points had been written and distributed by 10:25 a.m. on Friday, less than 10 hours after the quake. NRC technical experts were cautioned repeatedly not to make any public statements. All information had to come from Public Affairs.
In an email sent at 2:56 p.m., the updated talking points were unequivocally reassuring: "The NRC has regulations in place that require licensees to design their plants to withstand environmental hazards, including earthquakes ... based on historical data from the area's maximum credible earthquake, with an additional margin added."
But privately, the NRC was aware of uncertainties.
An hour before that email was sent, Brenner, the public affairs director, sent a "great work so far" memo to his staff at HQ and around the US His third bullet point highlighted he NRC's role in helping Japanese engineers deal with the problems at Fukushima -- a fact not mentioned in the NRC's press releases that day. The emails indicate that the Obama administration and the NRC were keen to keep up the appearance that they were merely observing the Japanese nuclear crisis and had no responsibility for helping resolve it.
From: Brenner, Eliot
Date: Friday, March 11, 2011, 1:54:57 PM
While one reporter knows or has guessed that there are Japanese here in our Ops center in communication with their home authorities, we will NOT make the[m] available and we will NOT volunteer their presence. If anyone knows they are here and wants to talk with them, they will have to make the request through the embassy to have it relayed to these folks.
The memo also instructed staff to evade any questions about efforts by the NRC's Office of Nuclear Reactor Regulation (NRR) to model the effects of similar earthquakes and tsunamis on California plants:
“NRR is getting tasked with making an overlay of the Japanese conditions . . . to see how west coast plants stack up against it,” it said. “We think preliminarily Diablo would have had no trouble with a wave that size. [For an earthquake of about] 8.9 we're not so sure about, but again we are not talking about that.”
[Find the distance from any US location to the nearest nuclear power plants with this map from Esri.]
In congressional testimony and interviews in that first week, NRC Chairman Gregory Jaczko was quick to say that the NRC could learn lessons from Fukushima.
"We're going to take a good solid look at everything that comes out of Japan, and if we need to make modifications to our facilities in this country, then we'll do that," he told NBC News on March 16. He did not disclose that the NRC technical staff had already been reassessing, before Fukushima, increased risks from earthquakes, tsunamis, dam failures and power blackouts.
Jaczko did push for release of a report on Fukushima and its lessons just 90 days after Fukushima. Some of those recommendations have been implemented. Jaczko, who resigned in 2012, declined a request last week to be interviewed.
The talking points written during the emergency for NRC commissioners and other officials were divided into two sections: "public answer" and "additional technical, non-public information." Often the two parts didn't quite match.
One topic the NRC avoided in the talking points, even when responding to a direct question: meltdown.
"Q. What happens when/if a plant 'melts down'?
"Public Answer: In short, nuclear power plants in the United States are designed to be safe. To prevent the release of radioactive material, there are multiple barriers between the radioactive material and the environment, including the fuel cladding, the heavy steel reactor vessel itself and the containment building, usually a heavily reinforced structure of concrete and steel several feet thick.
"Additional, non-technical, non-public information: The melted core may melt through the bottom of the vessel and flow onto the concrete containment floor. The core may melt through the containment liner and release radioactive material to the environment."
The Japanese public television network, NHK, asked if the NRC could provide a graphic depicting what happens during a meltdown of a nuclear reactor.
From: McIntyre, David
Date: Friday, March 18, 2011, 9:02 AM
NRC would not have such a graphic. I suspect any number of anti-nuclear power organizations might.
When reporters asked if the Japanese emergency could affect licensing of new reactors in the US, the public answer was "It is not appropriate to hypothesize on such a future scenario at this point."
The non-public information was more direct: This event could potentially call into question the NRC's seismic requirements, which could require the staff to re-evaluate the staff's approval of the AP1000 and ESBWR (the newest reactor designs from Westinghouse and General Electric) design and certifications."
On the subject of tsunamis, the public assurances omitted the "non-public " nuances that might have given the public reasons to doubt nuclear power safety:
* Design standards varied significantly from plant to plant in the US
* The experience in Japan had taught the NRC that it needed to study the dangerous effects of “drawdown,” the powerful receding of ocean water near the shore that can precede a tsunami's arrival.
* And although the US was developing new tsunami standards, those wouldn't be in draft form for another year.
'It Was a Hydrogen Explosion'
The NRC spokespeople sometimes had trouble following the public debate, because for days their computers were blocked by security rules from accessing Twitter and YouTube. And they often had incomplete information about events in Japan.
From: McIntyre, David
Date: Saturday, March 12, 2011, 10:02 PM
Just saw an incoherent discussion on cnn by Bill Nye the science guy who apparently knows zilcho about reactors and an idiot weatherman who said Hydrogen explosion? Pfft. I'm not buying it.
His boss sent back the following reply, correcting the staffer and explaining plans to ask the Obama administration to help blunt critical news coverage.
From: Brenner, Eliot
Date: Saturday, March 12, 2011, 10:07 PM
1: There is a good chance it was a hydrogen explosion that took the roof off that building, though we are not saying that publicly.
2: I have just reached out to CNN and asked them to call (former NRC Chairman Nils) Diaz, and reached out to push the white house yet again to start talking on background or getting out in front of some of this crap.
On March 20, when Energy Secretary Steven Chu hesitated on CNN when asked if US plants could withstand a 9.0 earthquake?
McIntyre, one of the agency’s spokesmen, suggested to his bosses what Chu should have said:
From: McIntyre, David
Date: Sunday, March 20, 2011, 10:01:00 AM
He should just say "Yes, it can." Worry about being wrong when it doesn't.
Sorry if I sound cynical.
The public affairs staff showed disdain in the emails for nuclear watchdog groups, including the Union of Concerned Scientists.
After the UCS raised concerns about diesel backup power and batteries being inadequate, as at Fukushima, spokesman McIntyre dismissed it as "bleating" from nuclear power foes.
When Steven Dolley, former research director of the NCI and a reporter for McGraw Hill Financial's newsletter Inside NRC, asked McIntyre for a nuclear containment expert to speak to a reporter, the spokesman asked if the reporter had contacted the industry's lobbying group, the Nuclear Energy Institute.
Dolley asked, "So, should I say NRC is deferring inquiries to NEI?" suggesting that the NRC was deferring to the industry it is supposed to regulate.
McIntyre shared this exchange with his bosses, adding the comment, "F---ing a-hole."
There is NO SUCH NRC REPORT!
The NRC's Public Affairs staff attempted to discredit news reports that raised questions about nuclear plants, even when they were based on NRC data.
A story by this reporter for msnbc.com (now NBCNews.com) reported that the NRC had published a study six months earlier with new estimates of the risk that an earthquake could cause damage to the core of US nuclear power plants. The plants were listed in alphabetical order, along with the NRC's risk estimates.
The msnbc.com story, published on March 16, ranked the US nuclear plants by those NRC estimates. Surprisingly, the highest risk was not on the Pacific Coast, where plants are designed and built with severe earthquakes in mind, but in the Central and Eastern states, where scientists have raised their estimate of the earthquake risk since the plants were designed and built. The story said that the NRC still described the plants as safe, but also said the margin of error had shrunk.
We had checked our understanding of the report with NRC earthquake experts, but NRC spokesman Scott Burnell responded to the story by asking the same staff to find fault with it.
From: Burnell, Scott
Date: Wednesday, March 16, 2011, 6:22 AM
I know you're going to have a cow over this -- somewhat inevitable when a reporter new to the subject tries to summarize things. Apart from "you're totally off-base," what specific technical corrections can we ask for??
OPA (Office of Public Affairs) -- this is likely to spark a lot of follow-up. The immediate response would be "that's a very incomplete look at the overall research and we continue to believe US reactors are capable of withstanding the strongest earthquake their sites could experience." I'll share whatever we get from the experts.
Senior officials at the industry's lobbying arm, the Nuclear Energy Institute, sent emails asking the NRC for help rebutting the story. Burnell urgently asked again for errors in the article.
From: Burnell, Scott
Date: Wednesday, March 16, 2011, 11:11 AM
Folks, the expected calls are coming in -- We need a better response ASAP!
But the NRC experts found nothing to correct.
From: Beasley, Benjamin
Date: Wednesday, March 16, 2011, 12:31 PM
I have received no concerns or corrections regarding the MSNBC article.
Nevertheless, the Public Affairs staff waved other news organizations off the story, particularly after New York Gov. Andrew M. Cuomo reacted to his state's Indian Point nuclear power plant having the worst risk in the NRC data.
From: McIntyre, David
Date: Thursday, March 17, 2011, 2:20 PM
I just filed this request for correction with The Huffington Post, which has a report of Cuomo wanting to shut IP based on the MSNBC report:
There is NO SUCH NRC REPORT! The NRC does not rank nuclear power plants according to their vulnerability to earthquakes. This "ranking" was developed by an MSNBC reporter using partial information and an even more partial understanding of how we evaluate plants for seismic risk. Each plant is evaluated individually according to the geology of its site, not by a "one-size-fits-all" model -- therefore such rankings or comparisons are highly misleading. Please correct this report.
His colleague in Atlanta, spokesman Joey Ledford, replied, "Great talking point, Dave. I wish I had it during my 10 or so calls today trying to debunk this thing."
The New York Times, which was reporting a story about Indian Point, was dissuaded from using the NRC's risk estimates. We asked the New York Times reporter, Peter Applebome, why he ignored the NRC data. He replied in an email, "Burnell said it wasn't accurate and included rankings the NRC never made. I have no idea if that's correct, but I was writing a column on deadline and figured I did not have the ability to figure out who was right in the time I had."
In his piece, Applebome quoted the NRC downplaying the risk: "Officials with the Nuclear Regulatory Commission say the site is safe and that its earthquake threat is on the lower end nationally and in the Northeast." The NRC's recent study with a different picture was ignored.
The NRC followed up with a blog post from Brenner, the public affairs chief, cautioning the public, “Don't Believe Everything You Read.” Brenner called the msnbc.com report "highly misleading."
He didn't mention that its figures came from the NRC.
Note: Emails excerpted in this report can be read in full here in a PDF file. A cache of many emails is included in larger PDF files No. 1, 2, 3, 4, and 5. More are available in the NRC's online public reading room.
Bill Dedman is an investigative reporter.
Posted in accordance with Title 17, Section 107, US Code, for noncommercial, educational purposes.