The Danger of Nuclear Reactors in War

Originally published at Project-Syndicate | August 5th, 2022

Russia’s scorched-earth war of aggression in Ukraine poses a threat to nuclear reactors unlike anything the world has ever seen. After decades of inaction, the international community can no longer afford to rely on loosely defined norms and warring parties’ own self-restraint.

LOS ANGELES – The Russian Army’s march into Ukraine on February 24 was horrific for many reasons, not least because it introduced the prospect of a military strike on any one of Ukraine’s 15 nuclear power reactors and spent-fuel pools, risking an immense release of radioactive elements. Fortunately, that has not happened – yet.

Whatever does happen next, the Ukraine crisis has shown that wars and reactors make for a dangerous mix, raising troubling questions for other simmering conflicts around the world: Israel and Iran, China and Taiwan, and North and South Korea, among others. All have large nuclear power or weapons plants, and all are at risk of war.

Complicating matters further, there is no international convention that bans attacks on nuclear plants in wartime. The only global accord that even purports to address the matter is the Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol I), 8 June 1977 (the “Protocol Additional”), and it is too ambiguous to be effective.

Because the International Atomic Energy Agency (IAEA) has made little headway in persuading the Conference on Disarmament (the world’s Geneva-based multilateral arms-control forum) to adopt a new protocol, the only remaining safeguard is self-restraint. But while countries largely have been reluctant to attack operational nuclear reactors in the past, the Ukraine war has shown that taboo to be fragile.

Ultimately, the scale of the threat demands a renewed effort by the international community to fashion a strict legal prohibition. But international jurists need not invent a new standard, because a perfectly serviceable template already exists. In 1988, India and Pakistan, two of the world’s most ferocious adversaries, cobbled together an accord that included everything the Protocol Additional lacks. This model should be adopted as a universal norm.


The history of military conflicts involving nuclear reactors underscores the urgency of the issue, because it shows that some disasters have been avoided only by chance. It has been four decades since Iran initiated the first-ever military strike on a nuclear reactor, at the outset of the Iran-Iraq War. The target was Osirak, a French-engineered plant in Iraq that some feared was a cover for a nuclear weapons program. On September 30, 1980, an attack by two Iranian Phantom bombers damaged the site’s water-cooling building and a waste treatment facility, but failed to damage the reactor itself.

Nine months later, Israel got involved. Having failed to stop construction of the Osirak facility by assassinating Iraqi scientists, pressuring France to withdraw its contract, and sabotaging reactor elements awaiting shipment to Baghdad, Israeli officials feared that time was running out. To bomb the reactor after it had begun operations was to risk the release of radioactive elements. So, in a raid on June 7, 1981, eight Israeli F-16 aircraft, escorted by six F-15s, dropped 16 iron “dumb” bombs on Osirak.

That attack both destroyed the plant and implicitly established a taboo against bombing live reactors. In the years that followed, countries contemplating military action generally adhered to this informal norm. In the Iran-Iraq War, Iraq hit two Iranian nuclear power reactors while they were still under construction. In the 1990s Yugoslav civil war, and in the 2020 Azerbaijan-Armenia conflict, combatants considered but did not undertake strikes on operating nuclear power reactors. And in 2007, Israel struck Syria’s suspected nuclear weapons reactor, Al-Kibar, before it went “critical.”

But there were also hiccups. In the 1991 Persian Gulf War, both Iraq and the United States targeted active nuclear sites. Saddam Hussein’s regime lobbed Scud missiles at Israel’s Dimona weapons reactor, but failed to hit it. And the US deployed F-16s, F-117s, and F-111Fs to strike Iraq’s square-mile Al Tuwaitha nuclear research complex, which included two small reactors.

To justify the attack, US authorities pointed to the plants’ highly enriched uranium fuel as a nuclear weapons risk. In the event, the strike had very little radiological consequence – but only because Iraqi authorities had already buried much of the irradiated material. The political effect was far more consequential, because the bombing set a dangerous precedent.


That brings us back to Russia’s war in Ukraine, which involves a threat to nuclear reactors unlike anything we have seen before. Russia’s military doctrine spares no quarter. Following the approach that it applied in Chechnya and Syria, its forces have targeted and struck Ukraine’s civil infrastructure with abandon.

Moreover, the Russian military’s indiscriminate approach has already set off nuclear alarm bells. In late February, it attacked a radioactive disposal site near Kyiv, luckily failing to hit the storage facility. Then, in early March, it bombed the Kharkiv Institute of Physics and Technology, which was initially thought to be operating a small reactor (fortunately, it housed only an onsite accelerator subcritical assembly that did not pose a nuclear hazard).

Notwithstanding these scares, Russia so far has flinched when it has come to Ukraine’s main nuclear power reactors. But, again, luck has played a large part in averting a disaster. Although Russian forces have not bombed any nuclear power plants, they have occupied two installations, including the defunct Chernobyl plant and its thousands of spent fuel rods.

Even more disturbing was the occupation of the Zaporizhzhia nuclear power plant, which has six large active reactors. A two-hour firefight not only damaged one reactor building and another reactor’s power transformer but also downed two of the four high-voltage transmission lines that are needed to pump coolant and keep the reactors and spent-fuel pools in a steady state.

Ukraine’s other nine nuclear power reactors – at Rivne and Khmelnytsky, and in South Ukraine – remain untouched; but multiple risks remain. Combat in the vicinity of the reactors could envelop the plants. By locating missiles and artillery in Zaporizhzhia, Russia has turned the installation into a military base from which it has launched munitions, leading Ukraine to hit the site with several small combat drones. The reactors suffered no damage, but the attack reportedly wounded multiple employees.

Adding to dangers, Russia has flown cruise missiles over Zaporizhzhia to reach other targets, highlighting the risk of accidental bombardment. In turn, the IAEA and others have called attention to the acute stress of Ukrainian plant operators working under duress, including arrest. Along with the interruption of supply chains for spare parts to Zaporizhzhia and normal maintenance, the deficits augur additional safety concerns. This week, IAEA Director General Rafael Grossi warned that the site is “completely out of control,” noting that “every principle of nuclear safety has been violated.”

Then there is the possibility that the Kremlin might intentionally destroy Ukraine’s nuclear plants to create a radiological legacy for a war gone bad. Obviously, this would be a risky move, because Russia would be betting that the radioactive debris would remain concentrated in Ukraine. But a dangerous nuclear escalation cannot be ruled out. There have already been reports of Russian occupiers mining the coolant intake at Zaporizhzhia; and, of course, a cyberattack could always still disrupt the plant’s management.

However farfetched such acts might sound, they would be consistent with the Kremlin’s scorched-earth policy. Moreover, in the event that Ukraine’s defenses fail, we also cannot rule out the possibility that a Ukrainian reactor operator would dismantle a reactor’s safeguards to leave Russia with a pyrrhic victory.


Whatever the outcome in Ukraine, the danger associated with nuclear reactors in combat should be clear, offering a warning to others. In the Middle East, Iran, Syria, and Hezbollah all envision strikes on Israel’s Dimona reactor as part of their war plans. In a recent military exercise, Iran reportedly simulated an assault on the plant with 16 ballistic missiles and five suicide drones. Any such attack could provoke either a “proportionate” Israeli response – such as a strike on Iran’s Bushehr nuclear power plant – or worse: a nuclear weapons response.

East Asia also has zones of concern. China’s rumblings about taking Taiwan by force, for example, put a spotlight on the island’s two active nuclear power plants. Moreover, China has numerous reactors on its southern coast that could become Taiwanese military targets. Arguably, this mutual vulnerability could serve as a deterrent against reactor attacks or even war itself. But that is far from guaranteed.

The Korean Peninsula presents a different profile. North Korea has its Yongbyon nuclear weapons reactor and a larger reactor under construction, not to mention significant radioactive waste deposits on site. And South Korea operates nearly two dozen nuclear power reactors. Borrowing from the Kremlin’s playbook of using nuclear brinkmanship to deter a NATO military intervention, North Korea could threaten South Korea’s reactors to temper a US-South Korean response to its aggression.

Finally, South Asia was once a theater of heightened nuclear tensions. But it now shows what could be achieved with a new agreement to manage these risks more comprehensively. By the early 1980s, India and Pakistan had fought three wars and risked a fourth, this time over each side’s budding nuclear weapons programs. The wars initially led India to develop nuclear weapons. But with the maturation of Pakistan’s own bomb program, intense anxiety took hold, prompting a search for safeguards.


Two options emerged. India could attempt a military strike to destroy Pakistan’s nuclear enrichment site at Kahuta, or it could practice forbearance. Concerned that an attack would result in a counterstrike on its Bhabha Atomic Research Center (which included a weapons reactor and related facilities), India settled on forbearance, and Pakistan agreed to come to the negotiating table.

The result was the 1988 India-Pakistan Non-Attack Agreement, which went much further than the Protocol Additional. Where the South Asia agreement definitively immunized reactors against military bombardment, the Protocol Additional equivocates, by lumping reactors together with dams, dikes, and other “installations that contain dangerous forces.” These, it stipulates, “shall not be made the object of attack, even where these objects are military objectives, if such attack may cause the release of dangerous forces and consequent severe losses among the civilian population.”

Leaving aside the obvious definitional questions (what counts as “dangerous” or “severe”?), the Protocol Additional’s prohibition breaks down in the next paragraph. “The special protection against attack shall cease,” it states, “for a nuclear electrical generating station only if it provides electric power in regular, significant, and direct support of military operations and if such attack is the only feasible way to terminate such support.”

This language is problematic because determining what is “regular, significant, and direct support” leaves far too much to the attacker’s discretion. Moreover, the prohibition doesn’t even apply to other sources of irradiated nuclear material, such as spent-fuel storage, reprocessing plants, and research reactors.

By comparison, the South Asian agreement removes all caveats in a clear comprehensive prohibition:

“i. Each party shall refrain from undertaking, encouraging, or participating in, directly or indirectly, any action aimed at causing the destruction of, or damage to, any nuclear installation or facility in the other country.

ii. The term ‘nuclear installation or facility’ includes nuclear power and research reactors, fuel fabrication, uranium enrichment, isotopes separation and reprocessing facilities as well as any other installations with fresh or irradiated nuclear fuel and materials in any form and establishments storing significant quantities of radio-active materials.”

Equally important, an additional clause requires the parties to provide an annual site update, with each side informing the other, once a year, on January 1, “of the latitude and longitude of its nuclear installations and facilities and whenever there is any change.”


The contrast with the Protocol Additional is glaring, underscoring the need for a new international standard. But getting there will require overcoming longstanding resistance within the Conference on Disarmament. Prohibiting attacks on reactors has never been a Conference priority, and when the group did take up the issue in 1980 it ended up spinning its wheels.

Then, in 1984, a frustrated IAEA General Conference issued this admonition:

“[The Committee] urges once again all Member States to make, individually and through competent international organs, further and continuous efforts aimed at the prompt adoption of binding international rules prohibiting armed attacks against all nuclear installations devoted to peaceful purposes.”

Beaten down by the failure to make progress, the IAEA’s General Conference hasn’t addressed the matter since 2009, when muted joint language was all it could muster: “Any armed attack on and threat against nuclear facilities devoted to peaceful purposes constitutes a violation of the principles of the United Nations Charter, international law, and the Statute of the Agency.”

Now that the war in Ukraine has laid bare the risks we face, the international community has no excuse to continue avoiding prophylactic action. The world needs a formal convention that bans attacks on nuclear reactors and related facilities. Having beaten the drum to discourage attacks on Ukraine’s reactors, the IAEA can lead the process by convening an expert review committee to assess the Protocol Additional, the political resistance to bolder safeguards, and legal alternatives, in addition to methods to improve physical defenses in and around nuclear power plants.

Ukraine is hardly the only reactor bubble that could burst. A formal agreement would provide sorely needed guidance that countries can then adopt in their military planning. The India-Pakistan agreement should serve as a model. Backstopped by recent nuclear scares in Ukraine and an IAEA endorsement of its conclusions and recommendations, an expert review committee’s work should be enough to spur a new round of serious negotiations.

Bennett Ramberg: A former foreign affairs officer in the US State Department’s Bureau of Political-Military Affairs, is the author of Nuclear Power Plants as Weapons for the Enemy (University of California Press, 1985).

Related Posts

Pin It on Pinterest

Share This