‘One in infinity’: failing to learn from accidents and implications for nuclear safety in India

Abstract

Safe operation of nuclear power facilities requires a culture of learning, but Indian nuclear authorities appear to continuously fail to learn the lessons of accidents including at facilities they operate. This paper examines how nuclear authorities in India responded to the Fukushima accidents and a previous accident at one of India’s nuclear power plants, and infers what they seem to have learned from them. By evaluating this experience in light of a wide body of research on factors promoting reliability and safety in organizations managing complex and hazardous systems, it seeks to draw lessons about the prospects for nuclear safety in India.

Keywords:

• India
• nuclear power
• learning
• nuclear safety
• accidents

Acknowledgments

MVR would like to thank Jan Beyea, Zia Mian, Sebastien Philippe, and Frank von Hippel for their comments on an earlier draft. We would like to thank the anonymous reviewer for several useful suggestions.

Notes

1. It could have been pointed out, for example, that Japanese reactors had survived numerous earthquakes before March 2011.

2. In many, if not all, of the countries that have continued supporting nuclear power, government or nuclear officials were quick to dismiss the possibility of a Fukushima-scale nuclear accident in their own countries (Ramana 2013).

3. For example, the Konkan Bachao Samiti, a local environmental organization, had raised concerns about the safety of the EPR design proposed for Jaitapur in a meeting with the Nuclear Power Corporation on 4 August 2010, citing problems identified by regulatory authorities in the United Kingdom and lack of regulatory approval in the United States (Pednekar 2010). Likewise, the Peoples Movement Against Nuclear Energy identified numerous safety concerns with Koodankulam, including volcanic and tsunami hazards, and inadequate fresh water availability onsite (PMANE 2011).

4. An exception to the AERB’s regulatory purview is facilities that have potential nuclear weapons applications, including fuel cycle facilities such as reprocessing plants.

5. The constitution of the AEC states that ‘the Secretary to the Government of India in the Department of Atomic Energy shall be the ex-officio Chairman of the Commission’.

6. In 2012, the national Comptroller and Auditor General assessed the AERB in detail and found legal and practical hurdles in the way of independent functioning of the organization (CAG 2012).

7. Though AERB and NPCIL looked at various nuclear power plants and associated spent fuel pools, there was no examination of reprocessing plants that contain large quantities of radioactive materials, both as spent fuel in pools (while awaiting reprocessing) and as high and intermediate level waste produced during reprocessing.

8. One way to quantify these levels has been on the basis of the amount of radioactivity of Iodine-131 (I-131) released to the atmosphere; Level 4 corresponds to approximately 50 to 500 TBq of I-131 release, Level 5 to approximately 500 to 5000 TBq, Level 6 to approximately 5000 to 50,000 TBq, and Level 7 to above approximately 50,000 TBq (IAEA 2012, 17–18).

9. The use of the term plant malfunction to refer to what happened at Chernobyl is also troubling.

10. The example of thyroid cancers resulting from the 1986 Chernobyl accident might be useful to recall in this context. In 1991, the International Atomic Energy Agency concluded that ‘there is no clear pathologically documented evidence of an increase in thyroid cancer of the types known to be radiation related’ (International Chernobyl Project and International Atomic Energy Agency 1991, 388). Since the latency period of thyroid cancers from Chernobyl is around 5.8 years (Nikiforov and Gnepp 1994), attempting to assess thyroid cancer incidence before this period has elapsed would, naturally, lead to the inference that there is no correlation between thyroid cancer incidence and the accident. As time passed, the increased incidence of thyroid cancers became more apparent. In 2000, the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) recorded that there were an ‘unusually high numbers of thyroid cancers observed in the contaminated areas during the past 14 years’, and went on to observe that ‘the number of thyroid cancers (about 1,800) in individuals exposed in childhood, in particular in the severely contaminated areas of the three affected countries, is considerably greater than expected based on previous knowledge’ (UNSCEAR 2000, 504–514).

11. At Fukushima, about 90% of the radiation is said to have blown over the ocean, which may explain part of the differences between estimates for Chernobyl and Fukushima (personal email from Jan Beyea, 10 February 2013). For an estimate of what might happen if the radioactive materials were to have fallen over populated areas, see (Gering et al. 2013).

12. The DAE’s emergency plans are similarly dubious and are conspicuously inadequate to the circumstances expected during an accident (Ramana 2012, 214–216). The sociologist Lee Clarke has argued that such implausible plans ‘are rhetorical instruments that have political utility … for organizations and experts’ rather than having any actual use in the event of a real emergency (Clarke 1999, 13).

13. This is not a new claim. In 1999, following the Tokaimura criticality accident in Japan, the AEC Chairman categorically asserted that there ‘is no possibility of any nuclear accident in the near or distant future in India’ (ToI 1999).

14. The AERB and NPCIL do not publicly specify what initiating events they postulate, but it is entirely possible that like the safety authorities in Japan, who chose not to postulate the kind of earthquake and Tsunami that occurred on March 11, 2011, real events might fall out of the bounds of postulated initiating events.

15. The Independent Investigation Commission on the Fukushima Daiichi Nuclear Accident found that ‘regulatory authorities … had encouraged the company [TEPCO] to incorporate new findings regarding tsunami risks into its safety plans, but such measures were not made mandatory’ (Funabashi and Kitazawa 2012).

16. The long interregnum that the authors talk about refers to the recurrence of an earthquake on the same fault, and this may not occur for many thousands of years.

17. To the extent possible, we derive this description from documents put out by the DAE and its sister organizations. If these are not available, or as a supplement, we use news and media reports. We assume that these are accurate unless there is reason to believe otherwise.

18. The source of information for these event details has been the annual reports entitled ‘Operating Experience with Nuclear Power Stations in Member States’ that the IAEA puts out. These reports are based entirely on information that the DAE provides the IAEA. See (Ramana and Kumar 2010) and (Ramana 2012) for specific references.

19. The former chairman of the Atomic Energy Regulatory Board, the organization responsible for the safety of DAE installations, has said that a local core-melt and explosive fuel-coolant interaction could not have been ruled out if the crew had not acted as they did (Chanda 1999).

20. According to the IAEA, an inherent safety characteristic is a ‘fundamental property of a design concept that results from the basic choices in the materials used or in other aspects of the design which assures that a particular potential hazard can not become a safety concern in any way’ [our emphasis] (IAEA 1991, 10). The phrase ‘in any way’ implies that operator inaction or even error should not lead to an accident.