|Fig. 1: Flight path of Osirak strike force.  (Source: Wikimedia Commons)
Lunchtime at a nuclear reactor is not a peaceful period, especially for a country at war. Anti-aircraft operators at the Osirak nuclear reactor, under construction in central Iraq, learned this the hard way when they returned from an afternoon meal to find their post a smoking ruin.
Deeming the construction of a breeder reactor that could be used to produce weapons-grade nuclear material for an overtly hostile power an existential threat, the Israeli Security Cabinet had opted to preform a preemptive strike on the Osirak facility. Saddam Hussein acquired an Osiris-type reactor from the French in 1976; building commenced in 1979, and the reactor was slated to become operational sometime in 1981.  Although it would require extensive modification, which would be difficult to perform with hundreds of French personnel still onsite, the facility could theoretically be used to produce nuclear weapons. In a Middle East fraught with conflict, with two Arab-Israeli wars in recent memory and an Iran-Iraq war that was only just beginning, Osirak represented a very real danger.
The story of the Israeli mission is compelling and interesting, but it will not be recounted here in detail; it has been described thoroughly by Sutter.  Fourteen aircraft hugged the ground to avoid radar, duped Jordanian and Saudi ground controllers, and, despite happening to fly directly over the Jordanian king's pleasure yacht, delivered at least eight unguided 2,000 lb bombs on the target. Fig. 1 shows their flight path.  The reactor was heavily damaged and was not repaired partly due to extenuating geopolitical factors. American pilots later finished what their ally had started, entirely destroying what remained of the reactor in the Gulf War ten years later.
The Osirak strike was intended as an act of deterrence, and whether it succeeded is a discussion still hotly debated among circles of defense and policy experts. This report is concerned with the physical effects that follow such a strike on a nuclear facility. Although airstrikes against nuclear facliities are by no means common, Osirak is the best-known example of a phenomenon that also includes Operation Scorch Sword, an Iranian strike on the same reactor a year earlier, and Operation Orchard, an 2007 Israeli strike against a suspected Syrian nuclear facility. As regimes will doubtless continue to pursue nuclear weapons, and airstrikes will be proposed as a means of halting this pursuit, it is important to understand the literal fallout of such an event.
A successful airstrike will necessarily penetrate the containment dome and destroy the sensitive portions of the reactor; otherwise the damage would not impair the reactor's function.  The containment dome exists for a reason. Besides the fuel rods themselves and the highly radioactive water in which they are commonly immersed, many reactors also store waste onsite. Critically, we note that the Osirak reactor had not yet become operational, and the only radioactive materials present were 25 kg of uranium provided by its French backers. As nuclear materials degrade, they produce a cocktail of radioactive isotopes that emit all types of radiation; α and β particles, as well as γ rays. All are harmful in their own ways and can be extremely dangerous to nearby biology, and their half-lives are as varied as their atomic weights.  Explosives have a way of dispersing whatever they hit; we have only to look to Chernobyl for the effects of an explosive release of radioactive material.
We will indeed look to Chernobyl for an example of the primary short-term effect of a strike, a radioactive plume. Notably, a nuclear power plant contains much more fissile material than a reactor of the sort used for research or a weapons program; however, the nature of the effect is the same. The principal hazardous fission products that rose over the western USSR and much of Europe were Cs-137, a highly water-soluble fission product with a thirty-year half-life, and I-131, a material with an eight-day half life that would later reappear in Fukushima.  Around 200,000 square kilometers of Europe and Russia were contaminated with measurable levels of Cs-137; however, cases of radiation sickness were only reported relatively close to the plant. The Soviet Air Force went so far as to seed rain over modern Belarus to prevent the plume from reaching heavily populated areas. Ultimately, fewer than 50 deaths were directly attributed to the disaster, which seems oddly low for so large an event.  The Pripyat area, evacuated immediately after the disaster, remains vacant, though its radioactivity levels are beginning to settle. Heightened radioactivity also persisted in some European plants and wildlife well after the event, but it too returned to normal without compromising public health.
Chernobyl is our best historical example, but we must recognize three critical caveats before applying our analysis of Chernobyl to a strike like Osirak. First, the amount of fissile available material is far lower. Full-fledged nuclear power plants operate with a much higher volume of fuel than rogue nations' reactors, both because of the economies of scale inherent in nuclear power and the fact that nations building nuclear power plants ostensibly do not have to hide their pursuit of uranium or plutonium in the way rogue nations do.  This is simple arithmetic; the less radioactive material present, the lower the potential for a massive fallout zone. Mercifully, sufficient data does not exist to place a mathematical correlation between fuel volume and area contaminated.
Second, a reactor produces more nuclear waste the longer it runs. Hazardous zones for a power plant that has been in operation fewer than 3 months are vastly smaller than those for a "mature core."  Nuclear reactors do not spring up overnight, and in the modern world, anyone with access to the tools of an airstrike also has acccess to satellite imagery. We can reasonably assume that any such strike on a nuclear reactor will occur before it has begun operating, as in the case of Osirak, or as soon as possible thereafter. The attacker's goal of minimizing the amount of fissile material produced has the happy by-product of limiting the volume of material subsequently available for environmental contamination. In the case of Osirak, as noted above, this quantity was confined to the 25 kg of fuel originally provided to the project by the French.
Finally, radioactive steam issued from Chernobyl over a period of days, driven by high pressures within the reactor vessel and the continued release of decay heat from uranium rods deprived of a cooling system.  Though a strike has the potential to merely rupture a containment vessel and cause a similar meltdown scenario, a successful strike, involving weapon detonation within the reaction chamber, will disperse the decaying material such that it is less concentrated and destroy the reactor dome such that release occurs all at once. This carries its own hazards, but the slow release of heat and steam from a large concentration of fuel is not among them.
The question of whether a strike on a nuclear facility holds significant environmental risks, then, seems to hinge on our second question: for how long has the reactor produced nuclear material? In the case of Osirak, the reactor was not operational, and therefore the only material available for dispersion was the 25 kg of French fuel. 25 kg may not sound like much, but each kilogram can produce 24M Wh of energy in a power plant. In the case of Chernobyl, most hazardous radioactivity came not from the fuel itself, which has a relatively long half-life, but from secondary products.  Documentation is not available to the general public concerning levels of radioactivity at Osirak in the aftermath of the strike; such is the nature of dictatorial regimes.
However, our number describing the amount of fuel in the reactor is all we need to engage in some extrapolation. The IAEA helpfully provides a maximum safe concentration of uranium dust, which we may assume would be the logical consequence of bombing a stock of uranium. This limit is 14,000 Bq, or 14,000 nucleus decays per second.  The half-life of U-238 is 4.5 billion years; as the computational tools available to me could not handle these numbers, I linearized to the half-life point to find that 25 kg of uranium, or 105 moles, experiences 224 million decays per second. This is an underestimate, as the rate of decay is initially somewhat higher. However, it is in the correct order of magnitude. As these decay products can go in any direction, predicting the precise exposure a human body would receive within this piece would require making a travesty of probability and solid angles. Instead we will simply point out that even if we allowed 14,000 decays per square meter, and all those decays would hit a person within that square meter, a vast overestimate of the severity, only 16,000 square meters would be prohibitively contaminated. This admittedly flawed but nonetheless instructive figure indicates that an Osirak-type strike could never come close to the disastrous radioactivity release from Chernobyl.
The above analysis would suggest that striking a nuclear reactor used for the production of fissile material, whether prior to or during its operation, poses surprisingly small risks for the environment and for public health. Naturally, the Israeli strike force was not particularly concerned with the health of the Iraqi populace, but they accepted the risk that fallout might spread back to their country. Israel again accepted that risk when it bombed a Syrian facility in 2007; our 200,000 square kilometer cone would place the entire nation beneath a Chernobyl plume. History, then, is consistent with the conclusion of this report: the principal risk of a preemptive strike is not the spread of fallout, but the retaliation that may follow.
© Evan Long. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 B. Sutter, "Operation Opera," Physics 241, Stanford University, Winter 2016.
 P. Seale, Asad: The Struggle for the Middle East (University of California Press, 1989).
 B. Ramberg, "Attacks on Nuclear Reactors: The Implications of Israel's Strike on Osiraq," Political Science Quarterly 97, 653 (1983).
 N. A. Beresford et al., "Thirty Years After the Chernobyl Accident: What Lessons Have We Learnt?," J. Environ. Radioactiv. 157, 77 (2016).
 Uranium Mining in Virginia (National Academies Press, 2012).