In the weeks following the Fukushima accident, Nuclear Regulatory Commission (NRC) and nuclear industry officials have been asserting that US nuclear plants are better prepared to withstand a catastrophic event like the March 11 earthquake and tsunami than Japanese plants because they have additional safety measures in place.
According to internal NRC documents, however, there is no consensus within the NRC that US plants are sufficiently protected. The documents indicate that technical staff members doubt the effectiveness of key safety measures adopted after the September 11, 2001 terrorist attacks.
Therefore, it remains highly uncertain whether the US would be better prepared than the Japanese to manage the aftermath of such severe events. Although the Japanese have engaged in heroic efforts, they have not been able to prevent significant damage to reactor cores, spent fuel and containment structures, resulting in huge radioactive releases into the atmosphere and the ocean.
UCS obtained the NRC documents on March 25 from a Freedom of Information Act (FOIA) request we made a month before the Japanese disaster.
NRC and industry officials recently testified before Congress that U.S. reactors are fully prepared for the worst. For example, at a hearing hosted by the Senate Energy and Water Appropriations Subcommittee on March 30, NRC Chairman Gregory Jaczko testified:
As a result of the events of September 11, 2001, we identified important pieces of equipment that regardless of the cause of a significant fire or explosion at a plant, the NRC requires licensees to have available and staged in advance, as well as new procedures and policies to help deal with a severe situation.
Likewise, testifying on behalf of the Nuclear Energy Institute, William Levis, the president and COO of the Public Service Enterprise Group, which owns two nuclear plants in New Jersey, told the subcommittee:
Since the terrorist events of September 11, 2001, U.S. nuclear plant operators identified other beyond-design-basis vulnerabilities. As a result, U.S. nuclear plant designs and operating practices since 9/11 are designed to mitigate severe accident scenarios such as aircraft impact, which include the complete loss of offsite power and all on-site emergency power sources and loss of large areas of the plant. The industry developed additional methods and procedures to provide cooling to the reactor and the spent fuel pool, and staged additional equipment at all U.S. nuclear power plant sites to ensure that the plants are equipped to deal with extreme events and nuclear plant operations staff are trained to manage them.
NRC calls these post-9/11 procedures “B.5.b measures,” referencing the section of the compensatory measures order the agency issued in 2002 to all reactor licensees. The agency codified them in its regulations in 2009 in a document titled CFR 50.54(hh)(2), but because their details are security-related, they are not publicly available.
At the March 30 hearing, both Jaczko and Levis sounded confident that B.5.b measures would protect U.S. reactors from a situation such as the ongoing crisis at Fukushima Daiichi, which lost off-site and on-site power for an extended period, eventually leading to the loss of all cooling.
Internal NRC documents obtained by UCS tell a different story.
In February 2011, UCS filed a FOIA request for all information associated with a secretive NRC program known as the “State of the Art Reactor Consequence Analyses,” or SOARCA.
SOARCA, according to the NRC, is “a research effort to realistically estimate the outcomes of postulated severe accident scenarios that might cause a nuclear power plant to release radioactive material into the environment. The SOARCA project applies many years of national and international nuclear safety research, and incorporates the improvements in plant design, operation and accident management to achieve a more realistic evaluation of the consequences associated with such accidents.”
The NRC also stated that SOARCA takes into account enhancements required by NRC after 9/11 — the B.5.b measures.
The SOARCA program, which the agency initiated in 2006, focused on two plants: Surry in Virginia and Peach Bottom in Pennsylvania. Coincidentally, Peach Bottom is a Mark I boiling water reactor (BWR) like Fukushima Daiichi reactors 1 through 4.
One of the hypothetical accidents that the SOARCA program analyzed was a station blackout at Peach Bottom where the plant failed to recover power before the backup batteries ran out — the very situation that occurred at Fukushima. That analysis would be extremely useful to understand what happened at Fukushima. However, the NRC has withheld nearly all documents related to SOARCA from the public.
In most Mark I BWRs experiencing a station blackout a cooling system that runs on battery power, known as the Reactor Core Isolation Cooling system, or RCIC, is available. But when the battery runs down — after eight hours or less — the RCIC will stop operating. If plant workers do not restore alternating current power by then, no cooling systems will be available and the fuel in the reactor will overheat and eventually begin to melt. Most experts believe that is what happened at Fukushima Daiichi units 1 through 3.
According to the emails obtained by UCS, NRC’s B.5.b measures contain unspecified strategies to continue operating the RCIC even after battery power is lost. However, the emails make clear that there are disagreements between NRC senior reactor analysts who work in NRC’s regional offices under the Office of Nuclear Reactor Regulation and the staff conducting the SOARCA project, who are in the agency’s Office of Research.
In particular, one NRC staff email, dated July 28, 2010, described senior analysts’ objections to SOARCA as follows:
One concern has been that SOARCA credits certain B5b mitigating strategies (such as RCIC operation w/o DC power) that have really not been reviewed to ensure that they will work to mitigate severe accidents. Generally, we have not even seen licensees credit these strategies in their own [probabilistic risk assessments] but for some reason the NRC decided we should during SOARCA.
My recollection is that [Region I senior reactor analysts] in particular have been vocal with their concerns on SOARCA for several years, probably because Peach Bottom is one of the SOARCA plants.
In other words, senior reactor analysts who work directly with the Peach Bottom Mark I BWR apparently do not have faith in the effectiveness of the very B.5.b measures that the NRC and nuclear industry officials are touting as a reason why the United States is better prepared to deal with a Fukushima-like event than Japan.
Another (undated) document reinforces this concern:
The application of 10 CFR 50.54(hh) [2009 regulations] mitigation measures still concerns a number of staff in [the Office of Nuclear Reactor Regulation]. The concern involves the manner in which credit is given to these measures such that success is assumed. . . . 10 CFR 50.54(hh) mitigation measures are just equipment on-site that can be useful in an emergency when used by knowledgeable operators if post-event conditions allow. If little is known about these post-event conditions, then assuming success is speculative.
If we are going to have any confidence that US plants are safe, the NRC and the industry have to be completely open and honest about what they know and what they don’t know. They are doing Americans a disservice if they are saying publicly that these untested measures are effective when privately they are expressing doubts they will work.
The concerns of NRC senior reactor analysts with regard to the credibility of post-accident mitigative measures need to be taken seriously by the NRC task force established to review regulations and policies in light of the Fukushima crisis.
A third NRC document is an email the briefly discusses the schedule of the SOARCA analysis.
Edwin Lyman is a senior staff scientist in the Global Security program at the Union of Concerned Scientists (UCS) in Washington, DC. This article was first published in the All Things Nuclear blog of the Union of Concerned Scientists on 6 April 2011; it is reproduced here for non-profit educational purposes.