Nuclear safety and the perception of risk

A nuclear reactor core has to maintain two equilibria: the rate of neutrons produced has to be equal to the rate of neutrons lost, and the rate of heat generated has to be equal to the rate of heat removed. Both equilibria are linked through so-called reactivity coefficients, the values of which make the system stable or not. Perturbations in these equilibria may be intro­duced by internal or external causes and by human error. Consequently, the design of a reactor has to contemplate all the possible inputs and determine how the plant will react against them in such a way that no damage to the core is produced and no release of radioactive products should occur; under these circumstances, the reactor is declared safe.

Nevertheless, accidents may occur when not all the potential inputs (or a combination of them) have been considered, when their magnitude has not been measured properly, when multiple human errors have been com­mitted, or multiple safety equipment is unable to work due to common — cause failures. Combinations of all these are possible, although remote. When such accidents occur, there is a possibility that radioactive products will be released to the exterior, with the likelihood that the health and safety of people will be affected and that environmental contamination by radio­active products will occur.

The accident at TMI-2 was due to a combination of equipment failure, poor maintenance and human error, the accident at Chernobyl-4 was caused by human mismanagement of the reactor’s unstable condition (INSAG, 1992a), whilst the accident at Fukushima-1 had its origin in the multiple common-cause failures produced by an earthquake and a following tsunami, against which the plant was not designed. In TMI-2 the release of radioac­tivity was limited and the health and safety of people was not at risk; in Chernobyl the release was very large and the radiological consequences very serious; in Fukushima-1, the release from the three affected units was about one-tenth that of Chernobyl but the radiological consequences were limited, due to an efficient emergency management.

Relevant lessons have been learned from such events and these lessons have served to improve the safety of present and future nuclear power plants. These accidents demonstrate that absolute safety is not achievable, and that there will always be a residual risk, although that it should be as low as possible, and acceptable. Prevention of accidents and mitigation of their consequences is the main aim of nuclear safety.

The safety level of nuclear installations and activities that give rise to radiation risks can be improved and maintained by following the IAEA Fundamental Safety Principles (IAEA, 2006), which provide the basis for the safety requirements and safety guides and programmes which have been developed by the IAEA as part of its safety standards activities.

These principles apply to a nuclear power plant in all modes of operation and to its entire fuel cycle installations and activities, such as transporta­tion of radioactive waste and nuclear materials, and their final disposal. The principles recommend the creation of a series of administrative enve­lopes and technical barriers that prevent accidents and mitigate their consequences.

These administrative barriers assign the prime responsibility for safety to the licensee, as well as allocating to government the responsibility of enacting a complete and satisfactory legal and licensing system and the creation of a regulatory body with three major activities: the development of a consistent set of safety standards, the verification of compliance with applicable standards, and an enforcement authority to correct any devia­tion. The licensee is also obliged to develop leadership and management for safety, based on the promotion of a safety culture within the installations and all related activities, on the regular assessment of safety performance and on feedback from operating experience. These administrative and pro­cedural barriers are essential to achieve and maintain the required safety levels.

Technical barriers also help to prevent accidents and mitigate their con­sequences by adhering to the concept of defence-in-depth, through a com­bination of consecutive and independent levels of protection which would have to fail to cause the release of radioactivity to the environment. Such levels include conservative designs and use of materials of high quality and reliability; the introduction of control, limiting, protection and monitoring systems; the addition of technological safeguard systems to cope with acci­dental situations and to mitigate their consequences, including so-called passive systems; the application of well-developed and trained emergency procedures; and the availability of emergency measures to protect people outside.

The safety level of a nuclear power plant can also be measured through its complementary function: risk. A quantitative risk assessment method­ology for nuclear power plants was first introduced in 1975 by the Reactor Safety Study (NRC, 1975); the methodology was later repeated in Germany and an English translation produced (EPRI, 1981) and later refined to consider five specific nuclear power plants covering the nuclear technolo­gies used in the USA (NRC, 1990). This new methodology has been used widely (although covering only the first two levels of these studies) across practically all nuclear power plants in the world.

Such ‘Probabilistic Safety Analyses’ (PSAs) are divided into three levels. Level 1 PSA determines the expected frequency of accidents producing core damages, the values obtained ranging from 1 in 10,000 to 1 in 100,000 per year and reactor. Level 2 PSA estimates the conditional probability of an early release of radioactive products by failure of the containment system within a damage core, with values found to vary from 1 in 10 cases to 1 in 100 cases. The accepted recommended values (INSAG, 1992b) are less than 1 in 100,000 per year and reactor for Level 1, and a conditional probability of 1 in 10 cases for Level 2 PSA. Level 3 PSA determines the complementary distribution function of the radiological consequences to the health and safety of the public and also the economic consequences derived from losing the plant and restoring the environment. When these results are compared with other technological risks and those from natural events, it is concluded that the risks of nuclear power plants are several orders of magnitude lower.

Although it may seem sufficient to estimate nuclear risks and put abso­lute, and also relative, values on acceptable risks, risks perceived by the individual and society are also a reality to be considered. In an analysis of perceived risks, it is necessary to consider, among the major aspects, the benefits obtained, familiarity with the type of risk, and the nature and time dependence of the harm produced.

The benefits obtained determine the perception of risk. Individuals accept high risks when the benefits are clear to them, which explains the acceptance of driving a car or smoking. The benefits from nuclear power are not clearly estimated by individuals and society: the need and apprecia­tion of the benefits obtained from electricity are well understood, but elec­tricity can also be provided by other means. Because of this, it is necessary to explain, once again, the worldwide, national and local socio-economic benefits coming from nuclear power. To make the picture complete, it would be necessary to compare the risks and benefits of the other sources generat­ing electricity, but such an analysis is outside the scope of this chapter.

Familiarity with the nature of a risk and its frequency is another major ingredient in the perception of risk. Although the use of radiation is now part of everyday life for many people, mainly through its use in medicine, fear of radiation is very high due to its peculiar nature, which is difficult to understand. An average individual receives doses, for medical purposes, which are much larger than those from natural radiation and orders of magnitude larger that those the most exposed person will receive from the operation of nuclear power plants and related fuel cycle activities. Nevertheless, society grossly exaggerates the risks perceived from nuclear power, despite the efforts made and the evidence presented to explain the real situation.

The timing of the harm produced is another aspect of interest. When damage done shows immediately, the perception of risk is different from when it may or may not come later in the life of the person, or if the risks may still exist for future generations. It is well known that high radiation doses may produce deterministic effects and that damage will show up soon after exposure but, most frequently, even in the case of severe accidents, most exposures produce low doses with the potential to produce stochastic effects, sometimes many years later. These circumstances have produced a considerable increase in perceived risk, with people believing that any exposure, however low, will produce the expected effects with certainty, despite efforts made to show that the probability of the effect is very low and proportional to the dose received.

Despite efforts made to increase safety, accidents cannot be completely discounted and preparation for them should be in place. Two instruments have been created to protect the health and safety of the public, and to protect private and public properties and the environment. The first is a legal instrument based on the concept of third-party liability for the damage caused. The second is the preparation and maintenance of emergency pro­cedures and the corresponding equipment needed to protect the health and safety of the individuals affected.