HLSS522Wk6

ORIGINAL ARTICLE

Assessment of probable scenarios of radiological emergency and their consequences

Yehoshua Socola , Yuriy Gofmana, Moshe Yanovskiya, and Binyamin Broshb

aJerusalem College of Technology, Jerusalem, Israel; bBuilding Division, The Standards Institution of Israel and Department of Industrial Engineering, Ariel University, Ariel, Israel

ABSTRACT Purpose: Despite the vast amount of literature on radiological emergencies, to the best of our knowledge there is no systematic review of probable scenarios and their consequences. This work aimed for compiling such review. Materials and methods: The authors comprised a Red Team – that is, simulated best efforts to inflict maximal damage to the society by various means of radiological attacks. Nuclear warfare including improvised nuclear devices is beyond the scope of this work. Results: The direct radiogenic health consequences of any conceivable radiological accident, nat- ural or man-made, are much less dangerous than those which are usually perceived. In each scen- ario, direct health effects are only a small part of the damage caused by fear and over-reaction; the damage is somewhat independent of the small health effect predicted for most of the scen- arios. The reason is that nuclear radiation has become perceptually connected with nuclear apoca- lypses. This connection has caused the emotional description of radiological emergencies to frequently substitute quantitative considerations. For example, Chernobyl and Fukushima became major humanitarian disasters not because of the radiation itself but because of the over-reaction of both the authorities and the public, that led to the unjustified relocation of hundreds of thou- sands of people. In Fukushima, the evacuation was not justified at all and in Chernobyl the evac- uated zone should have been re-populated after 1 month. Conclusions: It is vital to educate decision makers, first responders and the public about the fac- tual extent of possible radiological consequences, as well as about the very real danger of over- reaction. Since the extent of the countermeasures deployed is unavoidably connected, in the eye of the public, with the extent of the danger, we suggest launching educational campaigns that explain the factual extent of the radiation risk, followed by easing regulations and narrowing safety margins. Such measures will probably be the most efficient method of countering radio- logical terrorism: by depriving any adversary of the most important ability which is to cause an over-reaction.

ARTICLE HISTORY Received 3 March 2020 Revised 3 July 2020 Accepted 6 July 2020

KEYWORDS Dirty bomb; red team; disaster management; health effects; radiation protection

1. Introduction

For historical reasons people fear ionizing radiation since it has become perceptually connected with nuclear apocalyp- ses. Even before the first atomic bomb was dropped or even tested, an official report warned that ‘civilization would have the means to commit suicide at will’ (Smyth 1945). It seems that both superpowers – the USA and the USSR – were interested in exaggerating the effects of radiation in order to promote their nuclear deterrence capabilities (Jaworowski 1999). After the atomic bombing of Japan and the start of the nuclear arms race, many scientists became concerned about the very survival of civilization. When considering the terrible risks of nuclear warfare, one can emotionally under- stand those scientists who promoted the radiation risks in spite of the lack of supporting scientific evidence, as they hoped to stop the nuclear arms race. It has also been men- tioned, however, that for some scientists, there was probably

an additional kind of reasoning, less altruistic, behind pro- moting the perception of excessive radiation hazards. Such perception would assist them in obtaining funds and would promote their participation in decision-making (Mauer 2009, Ch. 12; Yanovskiy et al. 2019). Anyhow, radiophobia – the irrational fear of radiation hazards – has emerged. Moreover, radiophobia persists after the Cold War also and is exploited by certain circles for promoting their agenda (Mauer 2009, Ch. 12).

There is a vast amount of literature on radiological emer- gencies, especially after the 9/11 attack. For example, a search for ‘dirty bomb’ within ISI Web of Science Core Collection (https://apps.webofknowledge.com/) yielded 257 peer-reviewed papers published in 2001–2019. Despite this, to the best of our knowledge, there is no systematic review of probable scenarios and their consequences. For example, we failed to find such list neither in IAEA publications

CONTACT Yehoshua Socol socol@jct.ac.il Department of Electrical and Electronics Engineering, Jerusalem College of Technology, Havaad Haleumi 21, Jerusalem, Israel Copyright � 2020 Taylor & Francis Group LLC.

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 2020, VOL. 96, NO. 11, 1390–1399 https://doi.org/10.1080/09553002.2020.1798544

(https://www.iaea.org/publications/search/type/nuclear-secur- ity-series) nor in the 624-page overview published by the MIT Press (Maurer 2009). This paper aims to fill the gap. It seems to us that the following list of scenarios is comprehensive:

1. A nuclear power plant (NPP) emergency. 2. A nuclear waste storage site emergency. 3. A nuclear waste transportation emergency. 4. An emergency with a radioactive source for therapeutic

or industrial irradiation.

Several potential scenarios have been intentionally excluded by us from consideration. First and foremost, we did not consider improvised nuclear device (IND) which is a nuclear weapon. Therefore, IND event is principally differ- ent from any radiological event. There are many differences, two of them seem to us most important: first, the antici- pated casualties’ count is many orders of magnitude higher; second, most IND casualties are expected to be due to the nuclear blast, not radioactivity.

We did not consider separately emergency with nuclear- powered vessels. Such emergencies can be treated as a kind of NPP emergency with specific features beyond the scope of this paper.

We determined that an emergency with a nuclear war- head without a nuclear explosion is somewhat similar to a nuclear waste transportation emergency, but with much smaller amounts of radioactivity. The same can be said about accidents with nuclear-powered satellites.

As for scenarios involving fresh nuclear fuel for an NPP, such an incident involves very low radioactivity to be con- sidered as radiological emergency.

Incidents with a radioactive source for medical diagnos- tics or scientific research are also excluded from consider- ation. Diagnostic sources for nuclear medicine (radioisotope imaging) are much weaker than other types of radioactive sources, and have much shorter half-lives, typically several hours to several days. Without going into details, it seems clear that if a source is considered safe enough to be ingested or entered intravenously as a part of a medical pro- cedure, there should be no safety concerns regarding the misuse of such a source or even several dozens of such sour- ces for radioactive contamination. Radioactive sources used in scientific research (primarily in biological research) are even weaker than the medical ones.

Radiological accidents, both technological and those caused by sabotage or terrorist attacks, have much in com- mon and should be considered on the same basis. The main difference between the two is that a terrorist attack is usually perpetrated in a densely populated area, while the probabil- ity of a natural accident occurring in such an area is much lower. The four scenarios above are estimated for the antici- pated radiological contamination and its consequences pri- marily from the point of view of the advisability of population evacuation.

This paper is a kind of ‘Red Team’ report. Red Teams in general simulate best efforts of adversaries to inflict maximal

damage by various available (to the adversaries) means. Publishing Red Team analysis must balance between provid- ing important information to authorities, scientists and pub- lic, and not providing important information to adversaries. We hope that our paper is balanced in this sense. Comprehensive consideration of countermeasures is the task of ‘Blue Teams’. It is, therefore, beyond the scope of the pre- sent paper. However, we consider some countermeasures that are not often discussed, as, for example, involving local volunteers in safeguarding nuclear power plants and easing regulations, rather than making them more stringent.

The rest of this paper is organized as follows. Our results are presented in Sections 2–6. Since the main consequences of radiological event are due to evacuation and decontamin- ation, we begin with discussion of direct radiogenic effects and advisability of evacuation (Section 2). Then, four above- mentioned scenarios are considered in Sections 3–6, fol- lowed by Discussion and Summary.

2. Human cost of evacuation and its advisability

Evacuations lead directly to loss of life due to the temporary loss of medical care, psychosomatic disorders and even sui- cides. In the case of Fukushima, 1% of the evacuees died during the first 2 years due to evacuation-related causes. This, in addition to the natural mortality (WNA 2018a). Moreover, evacuated people lose much of their quality of life; Yanovskiy et al. (2020) estimated the loss of quality of life as 20%. Finally, evacuation also costs money. Associating human life with monetary value is psychologically difficult and may seem ethically challenging. However, there is no escape from making such an association when making a decision regarding public health and safety. Extraneous expenditure leads to a statistical shortening of life. Therefore, a cost-effectiveness analysis is routinely per- formed when formulating health policies (Neumann and Sanders 2017). Managing safety expenditures is similar to health care policies in the sense that both deal with life extension.

In this section, we compare health effects of radiation exposure (radiogenic health effects) with health effects of radiation exposure averting (evacuation-related health effects). General health includes both physical and mental parameters, including psychological and psychosocial. However, the structure of radiogenic and evacuation-related health damage is very different. In case of radiogenic dam- age – predominantly cancer – deaths contribute 97% of the disability-adjusted life years’ loss, while only 3% are due to disability, including mental (Fitzmaurice et al. 2019). In case of evacuation, mental health is the primary factor. Psychological and psychosocial problems lead to psycho- somatic problems like cardiovascular disease, as well as to alcohol and drug abuse, suicides, etc. (UNSCEAR 2000, p. 513).

We proceed now to consider two separate cases of radio- logical emergency: nuclear power plant failure, and contam- ination by spent nuclear fuel or powerful radioactive source. In the former case, the initial contamination includes many

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 1391

short-living isotopes and the radiation decays rather rapidly. In the latter case, the principal contaminator is the radio- nuclide Cs-137 with a half-life of 30 years.

2.1. Nuclear power plant emergency

Very recently, an important analysis of Church and Brooks (2020) has been published which stresses importance of optimizing the actions to limit radiation exposure while minimizing the social, economic, and health effects. This analysis, however, lacks quantitative recommendations. Yanovskiy et al. (2020) performed a cost-effectiveness ana- lysis for the advisability of population evacuation in the case of a nuclear power plant (NPP) accident characterized by vast amounts of radionuclides with relatively short half-lives (up to 2 years). In this case, life-time exposure (70–90 years) is about 3� [first-year exposure] (UNSCEAR 2013). The most widely used model for estimating the risk of radiation- induced cancer is the linear no-threshold (LNT) model (Lee and Elmore 2019). LNT assumes that every radiation dose, no matter how small, constitutes an increased cancer risk. This model is considered conservative and prudent for the purposes of radiation protection (Shore et al. 2018), though it has been the subject of serious scientific controversy (Tubiana 2005) and has been more and more criticized (Calabrese et al. 2018). According to the LNT model, the number of excess cancer deaths for population of mixed ages is usually estimated as 5% per Sievert (Sv); this number follows the recommendation of the BEIR-VII report of the U.S. National Research Council (NRC 2006, Table 12-6, p. 281). Yanovskiy et al. (2020) estimated the worst-case scenario for the number of additional cancer deaths by using the LNT model, despite its questionable validity. It was rec- ommended that radiation exposure to 500 mSv or higher during the first year may justify a short-term evacuation lasting up to 2weeks. Based on the results in Table 4 of Yanovskiy et al. (2020), we can conclude that in the case of an NPP accident, permanent relocation may be justified only if the projected first-year dose is above 3000 mSv. The Chernobyl accident demonstrated that such heavy contamin- ation is hardly probable. It may be worth to mention here that internal thyroid doses to children from radioiodine has been considered an important radiogenic consequence of NPP incidents. However, the recently acknowledged prob- lem of overdiagnosing indolent thyroid papillary adenoma as thyroid cancer (Vaccarella et al. 2016) demands complete

reconsideration of thyroid radiation carcinogenesis (Socol et al. 2019).

2.2. Contamination by spent nuclear fuel or powerful radioactive source

The situation is different in the case of contamination by spent nuclear fuel (SNF) or a radioactive source used for radiation therapy (RT) or industrial measurements. In this case, the principal contaminator is the radionuclide Cs-137 with a half-life of 30 years. However, the actual decline of the radiation rate in the environment is much faster due to the removal of radionuclides by ecological processes, such as rain and wind. In Chernobyl, for example, the radiation rate decreased by a factor of two during the time period from 10 years after the accident (when the role of short-lived radio- nuclides was already minor) to 20 years (Balonov 2013; Yanovskiy et al. 2020). We assume that the radiation decay in the environment follows the exponential law and suggest, therefore, that the effective half-life time T1/2 of Cs-137 in the environment is about T1/2 ¼10 years (rather than 30 years). For exponential decay with T1/2 �1 year, the life- time exposure is approximately T� [first-year exposure], with T ¼ T1/2/ln(2) � 15 years. The latter dose is about five times greater than the lifetime dose from an NPP accident (3� [first-year exposure]) with the same first-year exposure.

Let us compare now potentially averted radiation-related loss of life with loss of life caused by evacuation. As dis- cussed just above, we take the LNT-estimated number of excess cancer deaths as 5%/Sv. Each cancer death on average shortens life by about 11 years (NRC 2006, Table 12-4, p. 278). Therefore, life years’ loss of an exposed (non-evac- uated) person can be estimated as the dose in Sv multiplied by (11 years/death) � (0.05 death/Sv) ¼ 0.55 year/Sv.

The results are presented in Table 1 for different evacu- ation times. One QALY (quality-adjusted life year) loss appearing in Table 1 is equal to one life year loss (for can- cer) as discussed by Yanovskiy et al. (2020). One can see that relocation may be justified if the first-year effective total-body dose is about 600 mSv or higher. Temporary evacuation may be justified only for the time that the site of the accident is being decontaminated. Unlike in the case of an NPP accident, just waiting until the radiation rate falls due to natural decay factors is not effective: The human cost of temporary evacuation is higher than the projected life extension.

Table 1. Summary of evacuation cost and benefit for Cs-137 contamination.

Duration of evacuation t

Averted dose D(t) for

D12¼600 mSv, mSv

LNT-projected life loss

averteda QALY

Evacuation human costb

QALY

30 days 50 0.03 0.35 1 year 600 0.32 1.16 Permanent relocation 9000c 4.95 4.5

D12 – effective total-body dose during the first 12months. QALY – quality-adjusted life year. aLoss of 0.55 QALY per Sv, according to NRC (2006). bAccording to Yanovskiy et al. (2020), Table 4. cAssuming exponential decay with half-life of 10 years based on the Chernobyl data (Balonov 2013; Yanovskiy et al. 2020), the life-time exposure is about 15 first-year doses.

1392 Y. SOCOL ET AL.

The cost of the clean-up strongly depends on the target value of residual radioactivity. Actually, the dependence is nonlinear since it is much more expensive to perform the advanced stages of the decontamination. Since such depend- ence of the cost on the target is nonlinear and case-specific, we do not provide quantitative recommendations for setting targets for decontamination efforts. Nevertheless, we would like to stress the following important fact for setting the tar- get value: The buildings provide considerable radiation pro- tection. Different building types provide different values of radiation dose reduction (protection factor PF), varying from PF ¼ 2 for 1-story wood frames to PF ¼ 50 and higher for middle-level apartments of multi-story buildings (HSC 2010). For urban environments (multi-story build- ings), the effective dose received by people is only 10–20% of the dose measured in the open air (UNSCEAR 2000, Table 30, p. 581). For example, an open-air dose of 500 mSv per year (about 60 mSv/h, or 20-fold the world-average nat- ural background radiation) corresponds to an effective dose of up to 100 mSv/year.

As shown in Table 1, the LNT-projected statistical life shortening for an effective first-year dose of 600 mSv is 4.95 years. For 100 mSv/year, it is, therefore, about 0.8 years. The effective dose 100 mSv/air corresponds, as mentioned above, to open-air dose of at least 500 mSv/year. Thus, if the 500 mSv/year (open-air dose) area is decontaminated to the background level (2–6 mSv/year), the corresponding LNT-projected life extension is about 0.8 years. In this con- text, it should be mentioned that in every country, the life expectancy varies considerably for different locations due to various circumstances, such as socio-economic and environ- mental conditions. This disparity in life expectancy across countries is typically of several years; in the extreme case of Calton in the UK, it is 25 years below the country average due to poor socio-economic conditions, crime and suicides (Reid 2011).

3. A beyond-design-basis nuclear power plant emergency

We consider here four scenarios: (1) ‘natural’ beyond- design-basis accident as in Chernobyl and Fukushima, and three sabotage scenarios: (2) direct physical damage, (3) tak- ing over the NPP control room, and (4) cyber-terrorism.

1. Two major beyond-design-basis (BDB) nuclear acci- dents – at the Chernobyl NPP in 1986 and at the Fukushima NPP in 2011 – provide important data regarding any conceivable future accident.

After the Fukushima accident, about 160,000 people were ordered by authorities to evacuate; more than 1600 people – about 1% of the displaced – died during the first two years because of the evacuation (WNA 2018a). The evacuation was ordered to avoid exposure to radiation doses below 50 mSv during the first year (UNSCEAR 2013, p. 190). The IAEA (2015) safety guidelines recommend considering evacuation only if the first-year dose exceeds 100 mSv. The evacuation ordered by the Japanese authorities was therefore not justified in the light of the IAEA recommendations.

Waddington et al. (2017) also concluded that the human cost of the evacuation was higher than of projected radiation exposure. We shall add that, according to the conservative (that is, intentionally overestimating) LNT model of radi- ation carcinogenesis, 50 mSv exposure may lead to a life expectancy shortening of 10 days (Lee and Elmore 2019).

The Chernobyl accident has been characterized by Jaworovsky (2010) as probably the worst theoretically pos- sible nuclear accident. The dose rates were higher by about an order of magnitude than the Fukushima accident. According to the UNSCEAR (2000) report (Table 24, p. 528), about 25% of the residents who lived within the 30-km zone around the damaged reactor would have received above 400 mSv had they not been evacuated. Still, the radiation rate decreased rather rapidly. Based on the data of Balonov (2013), Yanovskiy et al. (2020) calculated that the second-year dose (months 13–24) was only about 20% of that of the first-year dose (the first 12months). Moreover, temporary evacuation for 1month only would have prevented 60% of the 12-months dose.

It should be mentioned that the radiation decay may be even faster. In soils like were present following fallout of the U.S. nuclear tests, the environmental half-life of Cs-137 was about 50 days. Also, the internal irradiation by Cs-137 is not of big concern since its biological half-life is about 100 days (Brooks et al. 2016).

A cost-effectiveness analysis performed by Yanovskiy et al. (2020) showed that a temporary two-week evacuation is marginally justified if the first-year dose exceeds 500 mSv. Therefore, temporary evacuation for a maximum of 1 month (as was initially intended) may have been justified in the Chernobyl case – but definitely not the permanent reloca- tion that actually took place. Thus, according to Jaworowski (2010), no future nuclear accident will probably demand long-term evacuation. For the first days after an accident, the population should be recommended to stay indoors (‘shelter in place’).

Let us elaborate on the possibility of a BDB event caused by an act of sabotage.

2. There are containment vessels around modern reactors, protecting them from outside intrusion and to containing potential radioactive materials if released (WNA 2020b). It is typically a concrete and steel structure of about 1-m thickness. It seems that even a commando unit which was trained and deployed by a rogue state would not be able to penetrate the containment vessel and destroy the reactor.

However, another possibility seems to exist: to destroy water tubes at the entrance of the containment vessel. In such way loss-of-cooling accident (like in Fukushima) may be inflicted. Later in this section, we discuss involvement of local volunteers for both the countering and prevention of any possible attack.

3. Now, suppose that a commando unit succeeded in tak- ing over an NPP control room. Experience shows that oper- ators’ actions by themselves are not enough: The Chernobyl accident would not have occurred without the flawed reactor design. Therefore, there is probably only a small chance that the reactor would be successfully blown up.

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 1393

4. Regarding cyber-terrorism, the same can be probably said. In either case, the issue is whether deliberate com- mands are able to blow an NPP reactor.

After considering in short the three scenarios of NPP sabotage (2–4), we should mention that in some states the very attempt, even if operationally unsuccessful, may be suc- cessful in shutting down the national nuclear industry due to political over-reaction. The reactions of the Japanese and German governments to the Fukushima accident are exam- ples of such over-reaction. In Japan, all NPPs were shut down immediately after the accident instead of reasonable safety re-assessment. Germany decided to close 8 of 17 reac- tors immediately and to phase-out all the rest by the end of 2022. Regarding Germany, it should be noted that the reac- tion to Fukushima was not completely surprising. While the ruling Cristian Democratic Union (CDU) was not alien to nuclear power, CDU tended to form wide coalitions includ- ing Alliance 90/The Greens, which is a principal NPP antag- onist (Kolarz 2016). It should also be mentioned that the accident at the Three Mile Island NPP was a major blow to the nuclear power industry, at least in the US. This was the case even though this accident caused no concerns to pub- lic health.

For the purpose of both the countering and prevention of any possible attack, authorities should be advised to make use of the local law enforcement personnel reinforced by local volunteers also for collecting counter-intelligence infor- mation regarding the potential commando unit. This is important since the latter cannot act without considerable pre-attack intelligence. Educating the counterforces regard- ing the scientifically based extent of radiation effects should be an inherent part of the preparedness effort. This training will also contribute to the prevention of panic and over- reaction (like large-scale evacuation instead of ‘shelter in place’ for several days) in case of an actual event. Once local volunteers join the guard, the local community will place more trust in information received directly from trusted friends and neighbors. Thus, panic based on an emotional misperception will be substituted by rational actions based on high-quality information from highly profes- sional sources.

Another recommendation is of a purely technological nature: to rely more on passive safety. Passive safety features take advantage of natural forces (gravity, heat conduction or natural convection, etc.) which require neither an active power source nor control signals, manual, or automatic. If a reactor is designed to be passively safe, no command sequence – either from the control room or by cyber-terror- ism – is capable of blowing it up.

4. A nuclear storage site emergency

Spent nuclear fuel (SNF) is by far the most radioactive type of material. Interim storage facilities may contain a vast amount of SNF. For example, the SKB Clab facility in Sweden already contains 7000 tons of SNF (SKB 2020) as compared to up to 140 tons of a single reactor (WNA 2020b). SNF is usually stored in ponds which are 7–12 m

deep (WNA 2020a). The fuel assemblies each containing, e.g. 460 kg of uranium (WNA 2020b) are usually about 4m long, positioned vertically. The circulating water both shields the radiation and cools the fuel.

The principal difference between an NPP accident and an SNF storage accident is that active or just-shut-down reac- tors yield high power (heat) that can feed an explosion by various mechanisms. For example, the explosive energy of the Chernobyl accident has been estimated as equivalent to 10 tons of TNT (Pakhomov and Dubasov 2010). No similar internal power source is present in SNF. We consider here in short two emergency scenarios: earthquake and sabotage.

1. Though a high-magnitude earthquake may destroy a facility (assuming the area is seismic), the Fukushima experi- ence shows that fuel storage ponds are likely to survive even the strongest earthquake.

2. Let us consider, as above, the low-probability scenario that a commando unit (trained and deployed by a rogue state) penetrates a storage facility and deploys an explosive charge at the bottom of a storage pond. Let us suppose that the charge contains 50 kg of high explosive. Accurate calcu- lations of potential damage depend on many parameters including material properties and professional ability of the unit to attach the charge. Based on existing experience, it seems that the closest fuel assembly only can be shattered, though much more may be damaged. We estimate, there- fore, that one fuel assembly – up to about half-ton of SNF (WNA 2020b) – may be dispersed by such sabotage. Very important, even this dispersed material is anticipated to remain contained by the storage building. Emergency of similar or even larger scale is considered in the next section.

5. Nuclear waste transportation emergency

Spent nuclear fuel (SNF) is transported by highway or rail only after being stored for at least a year after extraction from a reactor. At this stage, the main radioisotope of con- cern is c-radiating Cs-137. The fuel is transported in so called ‘Type B’ casks typically manufactured with layered forged steel walls 25–30 cm thick; the empty cask weight may reach about 100 tons, and when fully loaded, up to 120 tons (WNA 2017). The amount of transported radioactive material may be extremely high. For example, several types of Type B casks are designed to hold 12 spent fuel assem- blies (WNA 2017) from pressurized-water reactors (PWR), each containing about 460 kg of uranium (WNA 2020b). The amount of radioactive material in SNF can be calculated as follows. Let us assume a burn-up of 50GW� day per ton of uranium, the typical value for German nuclear power plants according to IAEA (2011). If we then take into account 200MeV per fission (WNA 2018b) with a Cs-137 yield of 6.4% per fission (Phillips 1991), we get 2 kg of Cs- 137 with a corresponding activity of 6.5 PBq (petabecquerel) per ton of uranium – about 36 PBq per cask. For compari- son, the amount of Cs-137 released due to the Fukushima accident was 75 PBq (WNA 2018a). Therefore, considerable dispersion of the transported SNF may comprise a major radiological emergency.

1394 Y. SOCOL ET AL.

Fortunately, such dispersion can hardly take place. Type B casks are designed and tested to withstand a fall of 9 m when fully loaded, so they are unlikely to fail in the case of any accident. Moreover, as we will show below, even a com- mando unit (trained and deployed by a rogue state) will have severe problems in releasing considerable amounts of radioactivity into the environment.

Let us suppose that the commando unit succeeds in stop- ping a train. One theoretical option is to open the cask by unscrewing the holding bolts. It is not clear from the unclas- sified sources whether this action can even be theoretically performed outside of a specialized facility.

Another option is to penetrate the cask by means of an anti-tank weapon shot from a stand-off distance. Then, one can insert a high explosive and a fuze into the cask through the hole and detonate the charge. Most probably, only a liquid explosive would be suitable for such a purpose. A rather large hole is needed to insert the explosive. Such a hole can probably be made by means of an explosively formed penetrator (EFP) similar to the Polish mine MPB- ZN. Note, that the Polish mine is designed to penetrate only 10 cm of homogenous steel (BELMA 2019), as opposed to Type B casks that are typically manufactured with layered forged steel walls with a total thickness 25–30 cm (WNA 2017). Detonation of 50 kg of high explosives release about 200MJ of energy, while a 110-ton cask falling from a 9-m height acquires a kinetic energy of 10MJ. So, it is theoretic- ally possible, though far from obvious, that such an explo- sive charge will open the cask.

If the cask is not opened by the burst, the amount of released radioactive material can be overestimated as the amount of material along the EFP path. Let us assume that on its way out the EFP hits 2 out of 12 fuel assemblies. If each assembly has a length of 2.5 m (it is usually longer), a 5 cm-wide EFP ‘channel’ disperses about 2% of an assem- bly’s SNF, thereby releasing at most 2% � 2/12 � (36 PBq) which is about 0.1 PBq. This is a typical value for a fresh radiotherapy Cs-137 source. Dispersion of a Cs-137 radio- therapy source is considered in the next section.

However, as we mentioned already, in some states, the very attempt, even if operationally unsuccessful, may be suc- cessful in shutting down the national nuclear industry due to a political over-reaction. The same goal, however, might be achieved by a lone terrorist shooting a single high-explo- sive anti-tank (HEAT) rocket at a SNF-loaded cask. Though the release of radioactivity in this scenario will pose no health concerns whatsoever, it may still be enough to trigger the over-reaction.

6. An emergency with radioactive source for therapeutic or industrial irradiation

According to IAEA (1987), the typical activity of a fresh Cs- 137 radiation source for radiation therapy (RT) is 0.1 PBq (1014 Bq) or 3000 curie (Ci). Such sources are potentially dangerous with their activity significantly above D-values. D-value is that quantity of radioactive material, which, if uncontrolled, could result in the death or a permanent

injury (IAEA 2006, p. 1). The corresponding D-values for Cs-137 are 1011 Bq if the source is not dispersed, and 2.0� 1013 Bq if dispersed (IAEA 2006, p. 7). Sources for industrial applications, such as flow meters, thickness gauges, and well-logging devices, are weaker. Actually, the use of Cs-137 in radiation therapy (RT) is declining, having been substituted by Co-60 sources and later by compact electron accelerators. Co-60 is metal and does not lend itself to easy handling. As for Cs-137 sources, they are typically 30-ml ampules containing rice-sized granules of cesium chloride CsCl which dissolves in water like table salt (IAEA 1987), so dispersing CsCl seems rather straightforward.

Certainly, high-dose-rate radiation from the source poses serious problems for handling. As shown below, an unpro- tected person can handle such a source for less than half an hour before absorbing a lethal dose. However, a 2-cm-thick steel plate (or 20-cm-thick brick pile) attenuates the radi- ation by about an order of magnitude, enabling about a 1-h exposure before absorbing 1Gy. A dose of 1Gy is what is necessary to produce the onset of the acute radiation syn- drome (ARS) i.e. sickness but with very little risk of death. A steel plate with a narrow viewing slot may partly solve the problem of shielding, while only the hands of the person are directly exposed to the source. The most effective way a potential adversary can disperse Cs-137 is probably by dis- solving the source in water, pouring this solution into an air humidifier, placing this humidifier in the cargo area of a pickup truck and then just driving. The driver can be also protected by a barrier such as a steel plate or brick pile.

There are radioactive sources much stronger than those of RT sources. For example, a 100,000Ci source (approximately equal to 30 RT sources) was considered by Rosoff and von Winterfeldt (2007). Such sources should be shielded by about 6 cm of steel to be able to handle it for an hour; as a result, the viewing slot becomes a problem in such a thick shield and the manipulators’ hands will absorb 6Gy (leading to acute radiation burn) in seconds. The slightest error in handling leads to the absorption of a lethal dose after several seconds (several tens of seconds at most). In summary, the use of a radioactive source that is significantly more powerful than that found in RT for sabotage purposes seems to us purely theoretical.

Let us estimate the radiation dose rate from 0.1 PBq ¼ 1014 Bq ¼ 105 GBq source. For c-radiation, the effective dose measured in sieverts (Sv) is equal to the absorbed dose in grays (Gy). We will use two formulas (IAEA 1987):

1. The c-radiation dose rate r at distance l from a point source of activity A is given by

r ¼ j � A=l2

j ¼ 8:9� 10–5 Gy=h � ��m2=GBq

2. The dose rate at 1m from uniform ground contamin- ation A/S is

r ¼ g � A=S

g ¼ 1:6� 10–12 Gy=h � �

= Bq=m2 � �

Let us now consider three scenarios.

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 1395

1. An adversary conceals a 105 GBq Cs-137 source. Any person situated at 1m from the source is subject to the dose rate of 8.9� 10–5 � 105 � 9Gy/h and receives a lethal dose of 4Gy in less than half an hour. Therefore, concealing such a source in a cafe, library etc. may be a rather effective attack with numerous casualties – though arguably more efficient than the more conventional use of knives, firearms, explosives, or trucks.

2. An adversary disperses 105 GBq (i.e. 1014 Bq) Cs-137 source over a small area, as for example as a ‘dirty bomb’. Let us assume, just for the sake of reference, that the conta- minated area is 100m2. The corresponding contamination is therefore 1012 Bq/m2 with associated c dose of 1.6� 10–12 � 1012 ¼ 1.6Gy/h. One hour of such exposure leads to an accu- mulated dose of 1.6Gy which is above the onset ARS at 1Gy (Glasstone and Dolan 1977). While the contaminated area (100m2) should immediately be evacuated and certainly demands serious decontamination, it is very small. No radio- genic casualties or extensive damage are anticipated provided the first responders are equipped with radiation monitors and do not spend more than 15min in the ‘hot’ zone.

3. Finally, an adversary may choose to contaminate a large area by means of a moving vehicle. Let us suppose that a moving vehicle with an air humidifier contaminates about a 10-m-wide zone behind it. Let us estimate that 100 km of streets of a large city are contaminated during several hours of drive. The corresponding area is 106 m2

and the corresponding dose rate is 1.6� 10–12 � (1014/106 ) ¼ 1.6� 10–4Gy/h¼ 1.40Gy/year ¼ 1400 mGy/year. Such contamination causes no risk of acute effects as the dosage required for ARS onset is about 1000 mGy, if absorbed within 4 days (Glasstone and Dolan 1977). However, such a radiation rate may cause cancer, so the area would have to be decontaminated.

If an adversary chooses to contaminate a smaller area with the same amount of Cs-137 – say, 10 km of streets – the local dose-rate will be higher. The necessary clean-up will have to be more intensive but considerably less extensive. We suggest that it will generally be cheaper and easier to decontaminate a smaller area with a higher level of contamination than a much larger area with a lower level of contamination.

7. Discussion

Speaking about evacuation to avert radiation exposure, we did not consider voluntary self-evacuation. This topic

deserves an independent study. We shall mention only that, unlike popular beliefs, people in general tend to act respon- sibly and generously in spite of isolated cases of panic and antisocial behavior (Jacob et al. 2008). We suggest that eas- ing over-cautious evacuation guidelines will lead to increased public trust in the authorities and therefore decrease the extent of self-evacuation.

A few words should be said about country-to-country dif- ferences in level of trust in the advice from the authorities. While historical, cultural and other factors are important, Letki (2017) maintains that the main factors of trust are freedom, society openness and transparency of the author- ities. This aspect is dealt further later in this section.

As shown above, we cannot rule out the theoretical possi- bility of a large-scale sabotage attempt against a nuclear power plant or against spent nuclear fuel that is being trans- ported. Nonetheless, the probability of such attacks being successful is very low. We would like to stress that a special- forces team capable of performing such an attack is probably capable of performing a conventional population-aimed attack which would easily claim hundreds of lives. For the sake of comparison, we note that in the past, several lone bombers with body-held charges inflicted losses of about 20 lives each (Brym and Araj 2006). Therefore, an enemy must have good reasons to attack a nuclear facility rather than a crowded area. Such reasons may be primarily connected with radiophobia, the irrational fear of radiation, afflicting both society and the government. In this case, an enemy may hope that the results of a targeted nuclear object will inflict a more long-lasting effect than mega-terrorism against people. It should be mentioned that in the present situation such hopes are not unfounded. For example, one may com- pare the public effects of the T�ohoku earthquake and tsu- nami that claimed at least 20,000 lives and was hardly discussed outside Japan, with the effects of the Fukushima nuclear accident that claimed no lives whatsoever yet is still considered worldwide as a major disaster.

Similar considerations are valid for any conceivable sabo- tage with a radioactive source. The outcome may be lethal, but hardly more effective than using knives, firearms or trucks. An important issue is providing every first-response team with a radiation monitor, as well as appropriate train- ing. The main issue, however, is to prevent over-reaction of the population and authorities. Long-term evacuation after Fukushima and permanent relocation after Chernobyl are

Table 2. Tentative program of introductory training on nuclear and radiological emergency for decision makers, teachers, and first responders.

Topic Hours

1 Fundamentals of nuclear physics. Ionizing radiation 1 2 Nuclear reactor

2.1 Reactor basics. Ecological advantages of nuclear energy 1 2.2 Plutonium production. Means to mitigate proliferation of nuclear weapons 1

3 Nuclear and thermonuclear weapons. Factors of damage 1 4 Hiroshima and Nagasaki. Nuclear tests 1 5 Immediate damage and protection: blast, thermal radiation and initial nuclear radiation 1 6 Estimated extent of damage – myths and facts. The factor of panic 2 7 Radioactive fallout. Health effects of ionizing radiation. Chernobyl and Fukushima – myths and facts 2 8 Civil defense 3

Total 13

1396 Y. SOCOL ET AL.

examples of such over-reaction. The prevention of over- reaction is achieved by proper education.

It seems, therefore, that the most effective way to combat radiological terrorism is to deprive it of its radiophobia-fed incentives. In other words, to eliminate radiophobia. Therefore, it is extremely important to educate decision makers, first responders and the public about the scientific- ally based estimations of direct radiological consequences, as well as about the very real danger of over-reaction like mass evacuation and setting unreasonable goals for clean-up. Such education may start with a short introductory training for decision makers, teachers, and first responders. This training must also review nuclear warfare because, as mentioned in the Introduction, ionizing radiation has become perceptually connected with nuclear apocalypses. A tentative program of a 12-h training is given in Table 2.

As for the physical security of nuclear objects, the psy- chological effect of further investment is anticipated to yield just the opposite effect – by increasing radiophobia. Namely, a rational person connects the level of security with the level of threat and reacts accordingly if the threat is realized. It is widely discussed, for example, that for safety reasons spent nuclear fuel is buried in antitank-weapon-proof casks 500 m underground. One should not then be surprised by people’s reaction (or rather their over-reaction, such as mass fleeing an extended area) if such a cask is penetrated on a railway.

In light of the above, we suggest that easing regulations and narrowing safety margins, preceded by educational cam- paigns that explain these measures, will increase nuclear security. For example, we can recommend easing regulations for medical and scientific radioactive sources that comprise no measurable danger, as well as elevating legal limits for irradiation of the public.

Deregulation may be viewed by some as compromising safety and security (for brevity, we shall speak about safety), but the truth is often just the opposite. For every function- ing system, technological or not, there is some optimal range of extent of safety measures. Deviating from the opti- mum in either direction compromises safety. This common wisdom principle can be traced probably to the Pentateuch: ‘do not add … and do not subtract’ (Deut, 4:2, 13:1); the medieval Jewish scholar Rabeinu Bahya (1050–1120) formu- lated: ‘It is a matter of caution not to be over-cautious’. Let us take as an example vehicle speed on a highway. While excessively high speed is obviously dangerous, excessively slow speed is also dangerous: a slowly moving vehicle effect- ively creates a hazardous obstacle that other motorists must avoid.

There are at least three reasons that over-caution and over-regulation compromise safety:

� Every additional feature, including safety feature, adds complexity to the system. Over-regulation causes systems to be too complex – and complex systems are inherently unsafe (Perrow 1999).

� Regulating bodies have incentives to concentrate on easy- to-control items like demanding extensive documenta- tion. These incentives are further strengthened by the

very limited de facto responsibility of the government officials (Tullock 1965). Interestingly enough, terms like ‘lazy bureaucrat’ (Arkin et al. 2003) and ‘lazy cop’ (Offner and Ojakian 2014) emerged in industrial engin- eering and game theory.

� Over-regulation imposes on the industry a heavy burden not necessarily contributing to safety – for example, extensive documentation. The industry has strong incen- tives to invest in the formal frequently controlled items rather than in subtle aspects of safety.

Just to illustrate the above considerations, we give an example of successful deregulation that improved safety con- siderably. Deregulation of the US railroad industry started in 1980. Between 1978 and 2005, the yearly number of rail- related fatalities has declined by nearly half from 1646 to 888, while the total number of rail-related accidents has fallen more than six-fold from 90,653 to 13,969 (Haley 2007).

There is a legitimate concern that drastically changing present regulation and guidance will result in more harm than good by compromising the trust in the authorities. However, we are confident that if the present guidance is over-cautious yielding loss of life more than preserving life, this guidance should be corrected as soon as possible. To err is human, and regulators are humans – so in our opinion admitting errors will only increase the public trust.

Reducing the regulatory burden will give a chance to attract new players and new investors to nuclear energy and other industries related to the use of radioactive materials. Increased investment, increased competition in the field of safety and security, and the mentioned above cooperation with local communities to ensure the safety of nuclear facili- ties are anticipated to provide increased security, also by increasing public trust in the authorities’ guidance.

In addition, firms and nonprofit organizations of growing industries will be much more interested in conducting out- reach campaigns than existing players. The latter are pres- ently survival-oriented and often do not see a long-term perspective. In addition, better understanding by the popula- tion (that is, voters) of the factual extent of the risks will also increase the susceptibility of the elected politicians to rational means of increasing safety and security of nuclear facilities.

Such measures both diminish an enemy’s incentives for nuclear terrorism and serve to ease the post-accident countermeasures.

8. Summary

Direct radiogenic health consequences of any conceivable radiological accident, either ‘natural’ or a result of a terrorist attack, are much safer than usually perceived. Population evacuation may be justified in extremely rare cases. For example, Chernobyl and Fukushima became major humani- tarian disasters not because of the radiation itself but because of the over-reaction of both the authorities and the public. This over-reaction led to unjustified permanent

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 1397

relocation (Chernobyl) or years-long evacuation (Fukushima) of more than 100,000 people in each case. Instead, as mentioned above, ‘shelter in place’ policy should have been chosen in Fukushima, and Chernobyl’s evacuated areas should have been re-populated after 1 month.

Our recommendations fall into two categories. The first category aims to counter radiological terrorism by depriving any adversary of the most important incentive – to cause an over-reaction (like ordering mass evacuation, setting unrea- sonable goals for clean-up, shutting down all the nuclear power plants). We propose:

� Launching educational campaigns for decision makers, first responders, and the public about the factual extent of the radiogenic health consequences, as well as about the very real dangers of over-reaction.

� Easing regulations and narrowing safety margins, since the extent of countermeasures is unavoidably connected in the public perception with the extent of the danger.

The second category aims at improving nuclear safety and mitigating consequences of an accident or a terror- ist attack.

� After appropriate training, providing all first responders with radiation monitors.

� Involving local volunteers in safeguarding nuclear facilities. � In the field of technology, preferring passive safety solu-

tions to active safety features.

Acknowledgments

The authors wish to thank Prof. Avi Caspi (Jerusalem College of Technology – JCT) for his encouragement of this work. The authors would like to thank Prof. Shlomo Engelberg (JCT) for thorough read- ing the manuscript and suggesting many important improvements. The authors are grateful to Dr Moti Brill (Nuclear Research Center Negev, ret.), Dr Col. (res.) Efraim Laor (Holon Institute of Technology), and Dr Col. (res.) Ori Nissim Levi (World Nuclear Forum 193) for numer- ous fruitful discussions and constructive criticism. The authors also acknowledge fruitful discussions with Dr Jerry Cuttler (Cuttler & Associates Inc.), Dr Zeev Dashevsky (Bar-Ilan Universiy, ret.), Prof. Yakov Krasik (Technion – Israel Institute of Technology), Mr Eliezer Tsafrir (Mossad, ret.), and Dr Avi Zigdon (Ariel University). Last but not least, the authors would like to thank the anonymous referees for their constructive criticism that led to considerable improvement of the manuscript.

Disclosure statement

The authors report no conflicts of interest.

Funding

This research was supported in part by the Jerusalem College of Technology Grant no. 5969.

Notes on contributors

Yehoshua Socol, PhD, received the PhD degree in physics in 2000 from Weizmann Institute of Science, Rehovot, Israel. Senior Lecturer at the

Department of Electrical and Electronics Engineering, Jerusalem College of Technology, Jerusalem, Israel. His present research interests include radiation carcinogenesis, epidemiology, risk assessment, public choice and ethics.

Yuriy Gofman, PhD, is a specialist in radiation damage. He is a Researcher at the Department of Electrical and Electronics Engineering, Jerusalem College of Technology, Jerusalem, Israel.

Moshe Yanovskiy, PhD, graduated from Higher School of Economics (Moscow). He is a senior research fellow at Institute of Applied Economic Research, Russian Presidential Academy of National Economy and Public Administration, Moscow, and a Researcher at the Department of Industrial Engineering and Management, Jerusalem College of Technology, Jerusalem, Israel.

Binyamin Brosh, PhD, Lt-Col (res.) is the Director of the Building Division of The Standards Institution of Israel, and Senior Lecturer at the Department of Civil Engineering, Ariel University, Israel. In the past, he served as Chief Engineer, the Israeli Home Front Command (HFC), and Head, R&D section, Department for Protective Structures, HFC. His present research interests include the survivability of con- structions under nuclear attack.

ORCID

Yehoshua Socol http://orcid.org/0000-0003-4167-248X

References

Arkin EM, Bender MA, Mitchell JSB, Skiena SS. 2003. The lazy bureau- crat scheduling problem. Inf Comput. 184(1):129–146.

Balonov M. 2013. The Chernobyl accident as a source of new radio- logical knowledge: implications for Fukushima rehabilitation and research programmes. J Radiol Prot. 33(1):27–40.

BELMA. 2019. Off-route antitank mine with influence fuse MPB-ZN. [accessed 2020 Jul 1]. http://www.belma.pl/en/military-products/p/ 387/off-route-antitank-mine-with-influence-fuse-mpb-zn.

Brooks AL, Church BW, Smith JM, Tolmachev SY. 2016. Annual 137Cs environmental half-life without remediation: impact on radiation dose. Jpn Hoken Butsuri. 51(1):51–59.

Brym RJ, Araj B. 2006. Suicide bombing as strategy and interaction: the case of the second intifada. Social Forces. 84(4):1969–1986.

Calabrese EJ, Hanekamp JC, Shamoun DY. 2018. The EPA cancer risk assessment default model proposal: moving away from the LNT. Dose–Response. 16(3): 1559325818789840

Church BW, Brooks AL. 2020. Cost of fear and radiation protection actions: Washington County, Utah and Fukushima, Japan {Comparing case histories}. Int J Radiat Biol. 96(4):520–531.

Fitzmaurice C, Abate D, Abbasi N, Abbastabar H, Abd-Allah F, Abdel- Rahman O, Abdelalim A, Abdoli A, Abdollahpour I, Abdulle ASM, et al.; Global Burden of Disease Cancer Collaboration. 2019. Global, regional, and national cancer incidence, mortality, years of life lost, years lived with disability, and disability-adjusted life-years for 29 cancer groups, 1990 to 2017. JAMA Oncol. 5(12):1749–1768.

Glasstone S, Dolan PJ. 1977. The effects of nuclear weapons. Washington (DC): US Department of Defense, and the Energy Research and Development Administration.

Haley MT. 2007. The Security of Our Nation’s Passenger and Freight Railroad Network. Testimony before the Subcommittee on Transportation Security and Infrastructure Protection, Committee on Homeland Security, U.S. House of Representatives, February 6, 2007. [accessed 2020 Jul 1]. https://www.transportation.gov/testi- mony/security-our-nations-passenger-and-freight-railroad-network-0.

[HSC] Homeland Security Council Interagency Policy Coordination Subcommittee for Preparedness & Response to Radiological and Nuclear Threats. 2010. Planning Guidance for Response to a Nuclear Detonation (2nd Edition), Fig. 3.1, p.70. [accessed 2020 Jul

1398 Y. SOCOL ET AL.

1]. https://www.remm.nlm.gov/PlanningGuidanceNuclearDetonation. pdf.

[IAEA] International Atomic Energy Agency. 1987. The radiological accident in Goiania. Vienna: IAEA. [accessed 2020 Jul 1]. https:// www-pub.iaea.org/MTCD/publications/PDF/Pub815_web.pdf.

[IAEA] International Atomic Energy Agency. 2006. Dangerous quanti- ties of radioactive material (D-Values). Vienna: IAEA.

[IAEA] International Atomic Energy Agency. 2011. Impact of high burnup uranium oxide and mixed uranium–plutonium oxide water reactor fuel on spent fuel management; NF-T-3.8. Vienna: IAEA. [accessed 2020 Jul 1]. https://www-pub.iaea.org/MTCD/Publications/ PDF/Pub1490_web.pdf.

[IAEA] International Atomic Energy Agency. 2015. Preparedness and response for a nuclear or radiological emergency: general safety requirements. Vienna: IAEA; p. 2015. [accessed 2020 Jul 1]. https:// www-pub.iaea.org/MTCD/Publications/PDF/P_1708_web.pdf.

Jacob B, Mawson AR, Payton M, Guignard JC. 2008. Disaster myth- ology and fact: Hurricane Katrina and social attachment. Public Health Rep. 123(5):555–566.

Jaworowski Z. 1999. Radiation risk and ethics. Phys Today. 52(9): 24–29.

Jaworowski Z. 2010. Observations on the Chernobyl disaster and LNT. Dose Response. 8(2):148–171.

Kolarz P. 2016. Giddens and politics beyond the third way: Utopian Realism in the Late Modern Age. London: Palgrave Macmillan; p. vi.

Lee CI, Elmore JG. 2019. Radiation-related risks of imaging, UpToDateV

R

. Alphen aan den Rijn: Wolters Kluwer. [updated 2019 Oct 24; accessed 2020 Jul 1] https://www.uptodate.com/contents/ radiation-related-risks-of-imaging.

Letki N. 2017. Trust in newly democratic regimes. In: Uslaner E, edi- tor. The Oxford handbook of social and political trust. Oxford: Oxford University Press.

Mauer SM (Ed.). 2009. WMD Terrorism: science and policy choices. Boston (MA): MIT Press: p. 624.

[NRC] National Research Council. 2006. Health risks from exposure to low levels of ionizing radiation: BEIR VII Phase 2. Washington (DC): The National Academies Press.

Neumann PJ, Sanders GD. 2017. Cost-effectiveness analysis 2.0. N Engl J Med. 376(3):203–205.

Offner D, Ojakian D. 2014. Variations of cops and robber on the hypercube. Australas J Combin. 59(2):229–250.

Pakhomov SA, Dubasov YV. 2010. Estimation of explosion energy yield at Chernobyl NPP accident. Pure Appl Geophys. 167(4–5):575–580.

Perrow C. 1999. Normal accidents: living with high risk technologies. Princeton: Princeton University Press (revised edition); p. 464.

Phillips JR. 1991. Irradiated fuel measurements. In: D. Riely, N. Ensslin, H. Smith, Jr., S. Kreiner, editors. Passive nondestructive assay of nuclear materials. NUREG/CR-5 550, LA-UR-90-72, Washington (DC): U.S. Government Printing Office; p. 529–562.

Reid M. 2011. Behind the Glasgow effect. Bull World Health Organ. 89:701–776. [accessed 2020 Jul 1]. http://www.who.int/bulletin/vol- umes/89/10/11-021011.

Rosoff H, von Winterfeldt D. 2007. A risk and economic analysis of dirty bomb attacks on the ports of Los Angeles and Long Beach. Risk Anal. 27(3):533–546.

Shore RE, Beck HL, Boice JD, Caffrey EA, Davis S, Grogan HA, Mettler FA, Preston RJ, Till JE, Wakeford R, et al. 2018. Implications of recent epidemiologic studies for the linear nonthres- hold model and radiation protection. J Radiol Prot. 38(3): 1217–1233.

SKB. 2020. Clab – Central Interim Storage Facility for Spent Nuclear Fuel. [accessed 2020 Jul 1]. https://www.skb.com/our-operations/ clab/.

Smyth HD. 1945. Atomic energy for military purposes; the official report on the development of the atomic bomb under the auspices of the United States Government, 1940–1945. York, PN: Maple Press; p. 224. [accessed 2020 Jul 1]. http://archive.org/details/ atomicenergyform00smytrich.

Socol Y, Shaki YY, Vaiserman A. 2019. Thyroid cancer overdiagnosis: implications for understanding radiation carcinogenesis and for medical imaging. Chem Biol Interact. 305:1–2.

Tubiana M. 2005. Dose-effect relationship and estimation of the car- cinogenic effects of low doses of ionizing radiation: the joint report of the Acad�emie des Sciences (Paris) and of the Acad�emie Nationale de M�edecine. Int J Radiat Oncol Biol Phys. 63(2):317–319.

Tullock G. 1965. The politics of bureaucracy. Washington (DC): Public Affairs Press; p. 383.

[UNSCEAR] United Nations Scientific Committee on the Effects of Atomic Radiation. 2000. UNSCEAR 2000 Report to the General Assembly, with Scientific Annexes. vol. II: Effects. Annex J: Exposures and effects of the Chernobyl accident. New York: United Nations. [accessed 2020 Jul 1]. http://www.unscear.org/docs/publications/ 2000/UNSCEAR_2000_Annex-J.pdf.

[UNSCEAR] United Nations Scientific Committee on the Effects of Atomic Radiation. 2013. UNSCEAR 2013 Report to the General Assembly, with scientific annexes. Volume I: Report to the General Assembly, Scientific Annex A. New York: United Nations. http:// www.unscear.org/docs/reports/2013/13-85418_Report_2013_Annex_ A.pdf.

Vaccarella S, Franceschi S, Bray F, Wild CP, Plummer M, Dal Maso L. 2016. Worldwide thyroid-cancer epidemic? The increasing impact of overdiagnosis. N Engl J Med. 375(7):614–617.

Waddington I, Thomas PJ, Taylor RH, Vaughan GJ. 2017. J-value assessment of relocation measures following the nuclear power plant accidents at Chernobyl and Fukushima Daiichi. Process Saf Environ Prot. 112/A:1–198.

[WNA] World Nuclear Association. 2017. Transport of Radioactive Materials. [updated 2017 Jul; accessed 2020 Jul 1]. https://www. world-nuclear.org/information-library/nuclear-fuel-cycle/transport- of-nuclear-materials/transport-of-radioactive-materials.aspx.

[WNA] World Nuclear Association. 2018a. Fukushima Daiichi Accident. [updated 2018 Oct; accessed 2020 Jul 1]. http://www. world-nuclear.org/information-library/safety-and-security/safety-of- plants/fukushima-accident.aspx.

[WNA] World Nuclear Association. 2018b. Physics of Uranium and Nuclear Energy. [updated 2018 Feb; accessed 2020 Jul 1]. https:// www.world-nuclear.org/information-library/nuclear-fuel-cycle/intro- duction/physics-of-nuclear-energy.aspx.

[WNA] World Nuclear Association. 2020a. Storage and Disposal of Radioactive Waste. [updated 2020 Mar; accessed 2020 Jul 1]. https:// world-nuclear.org/information-library/nuclear-fuel-cycle/nuclear- waste/storage-and-disposal-of-radioactive-waste.aspx.

[WNA] World Nuclear Association. 2020b. Nuclear Power Reactors. [updated 2020 Apr; accessed 2020 Jul 1]. https://www.world-nuclear. org/information-library/nuclear-fuel-cycle/nuclear-power-reactors/ nuclear-power-reactors.aspx.

Yanovskiy M, Levi ON, Shaki YY, Socol Y. 2020. Consequences of a large-scale nuclear-power-plant accident and guidelines for evacu- ation: a cost-effectiveness analysis. Int J Rad Biol. DOI:10.1080/ 09553002.2020.1779962

Yanovskiy M, Shaki YY, Socol Y. 2019. Ethics of adoption and use of the lin- ear no-threshold model. Dose Response. 17(1). DOI:10.1177/ 1559325818822602

INTERNATIONAL JOURNAL OF RADIATION BIOLOGY 1399

Copyright of International Journal of Radiation Biology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.