The authoritative guide to ensuring science and technology make life on Earth better, not worse.
By Henry Sokolski | March 5, 2025
The Kori nuclear power plant in South Korea is the largest nuclear plant currently operational in the world. An attack on one of its reactors or spent fuel storage pools would release large amounts of radioactivity and cause massive evacuations, diverting attention from a possible strategic crisis in East Asia. (Credit: Korea Kori NPP / IAEA, via Flickr)
A wargame sought to test if a major radiological release that would prompt the evacuation of millions of civilians could distract key US allies from assisting and rebuffing an all-out military invasion of Taiwan. The short answer was yes.
The game originally presumed that China, wanting to keep the United States, South Korea, and Japan out of a fight over Taiwan, would ask North Korea to strike the Kori nuclear power plant in South Korea to force massive evacuations. But a brief by Joshua Stanton, an expert on North Korea, changed that scenario significantly. Stanton argued that a North Korean military offensive was not the only threat Seoul faced. In addition, military planners need to worry about what South Koreans sympathetic to the North might do.
As Stanton explained, North Korea would prefer to mobilize South Koreans who favor confederating with the North to act on Pyongyang’s behalf rather than invade the South. This insight prompted an adjustment to the wargame: Instead of North Korea firing its missiles from over the border, the aerial attacks against Kori in the game would come from locally lofted drones, thereby significantly complicating attribution and response.
Shortly after making this change to the game, political events in South Korea seemed to confirm Stanton’s judgment. South Korean president Yoon Suk Yeol’s impeachment proceedings are still ongoing. What animates the controversy is not only Yeol’s declaration of martial law but the national debate over how to secure peace on the Peninsula. South Koreans are split between those who believe North Korea is an arch enemy that must give way eventually to a government more favorable to the South and those who believe that a confederation with North Korea is the only way to produce lasting peace.
Another challenge in designing the game was to convince the players—which included former senior Japanese, South Korean, and US security and nuclear regulatory officials, US legislative staff, and US military service members—that attacking nuclear reactors was realistic. Up until Russia’s attacks on Ukraine’s reactors, the conventional wisdom was that even hostile states would not choose to attack nuclear facilities. A key reason was the fear that such attacks could release massive amounts of radiation that could blow back on one’s own troops and population. Putin’s assaults on Ukraine’s nuclear plants have all but shattered this assumption. Security officials in the United States and allied countries now are worried and refuse to publicly discuss this prospect, saying it is too “sensitive.”
Then, on December 31, leaked Russian military plans to attack Japanese civilian infrastructure exacerbated these fears. The plans included targeting the Tokai nuclear complex. Should they choose to also target the Rokkasho nuclear reprocessing plant in northern Japan, one of the largest spent nuclear fuel storage facilities in the world, it could result in spent fuel fires and the release of massive amounts of radioactivity—several times greater than at Chernobyl.
The game players assumed the United States would come to Taiwan’s defense. In light of recent events, that may be mistaken. Following Ukraine President Volodymyr Zelensky’s disastrous meeting with President Donald Trump and Vice President J.D. Vance in the Oval Office, some analysts now believe the United States will abandon its global defense of liberal self-government. Instead, Washington would recognize regional spheres of influence for Russia, China, and the United States. If this is correct, the United States would not militarily intervene in East Asia at all. Although still viewed as a possibility, it only fortifies the game’s key findings.
The game offered three major takeaways:
Attacks against civilian nuclear plants could strategically sideline Japan and South Korea and impede the United States from effectively dealing with concurrent strategic crises in East Asia. Given the many Japanese and South Korean civilian infrastructure targets North Korea and China could strike, many view strikes against civilian nuclear plants as just another military option among an endless list of targeting possibilities. Nuclear power plants, however, are different. Unlike other civilian infrastructure targets, these facilities can release dangerous radiation and force the evacuation of thousands to millions of people, both inside and outside the targeted country.
During the war game, locally lofted drones hit reactors and spent fuel storage pools at South Korea’s Kori nuclear plant. This forced the evacuation of millions of South Koreans and Japanese from affected areas and created dramatic alliance tensions: South Korea wanted to be equally involved in any strike against the North, whereas Japan was more preoccupied with evacuation and humanitarian issues. The United States chose to ignore these concerns and focused instead on messaging China by unilaterally striking North Korea without consultations with Japan or South Korea.
For the last seven decades, Washington has tried to protect its interests and the security of Japan and South Korea with bilateral mutual security pacts with Tokyo and Seoul. These pacts have helped deter North Korea from attacking South Korea and China and Russia from attacking Japan. In this game, the pacts were not compelling enough to overcome the divergent security interests involving Washington, Seoul, and Tokyo. Both Japanese and South Korean teams were concerned that these bilateral treaties needed to be fortified with more robust trilateral arrangements. The teams wanted to forge a trilateral security pact with the United States to tamp down these divergent interests. Washington, however, feared such an arrangement would limit the freedom of action of the United States and hinder its ability to respond effectively.
Recommendations: The United States’ defense of the Western Pacific likely will turn on what Washington chooses to do. The prospects of success without finding new common interests that would work trilaterally, however, do not look bright. Suppose the United States is unable—or unwilling—to support a massive, Reagan-style military buildup to project significant new forces in the Western Pacific. In that case, it will either have to collaborate with Seoul and Tokyo in compelling new ways or slowly retreat.
Continued silence of US and allied officials in peacetime about the vulnerabilities of civilian nuclear plants to overt and gray warfare increases the odds and security impacts of such attacks. The recently leaked Russian military documents revealed intentions to attack South Korean and Japanese military and civilian infrastructure targets—including nuclear facilities—to prevent NATO from exploiting any Russian weaknesses on its Eastern borders. If Russia struck these facilities and spent fuel fires ensued, much of Japan (conceivably many millions of people) would have to evacuate. Despite the presence of many spent fuel storage facilities at many large nuclear power plant sites across South Korea and Japan, none of these sites currently have any active air defenses. Nor are there yet sufficient nuclear evacuation plans.
In the game, Japan pleaded with its Pacific allies and China to grant immediate humanitarian visas to displaced Japanese. On such short notice, Japan’s allies issued only limited numbers of such visas, and China refused. As for the critical mission of identifying the “local” attackers of South Korea’s nuclear plant, this too was unaddressed. As one game participant (a former senior official) privately explained, public discussion and planning related to the targeting of nuclear plants is necessarily limited as there is no easy way to limit the military vulnerability of these facilities. The continued silence in these matters, however, serves more as an invitation for attacks than a defense. The targets, after all, are public utilities, and their defense requires a substantive public brief. The United States, Japan, and South Korea currently game and consult on the remote possibility of US nuclear weapons use in Asia, as the security implications of such use are so great. Yet, there have been no serious consultations or planning regarding the possible targeting of civilian nuclear facilities.
Recommendations: Washington should assess the threat of Russian, Chinese, North Korean, and locally staged strikes against civilian nuclear plants and facilities and game possible responses as part of routine military nuclear consultations. More generally, the United States should work with key security allies who operate their own large, domestic civilian nuclear plants and coordinate civil defense efforts to secure civilian infrastructure against overt and gray warfare attacks. Washington should conduct this effort both in classified and unclassified forums. This should be seen as a new form of cooperative threat reduction that should be conducted in close collaboration with the Defense and Energy departments.
With gray warfare strikes against civilian infrastructure, including nuclear plants, attribution and deterrence of such attacks can no longer be assumed. What caught most of the players off guard in the game was their inability to attribute with certainty the attack against South Korea’s Kori plant. Japan insisted on such proof but was unconvinced North Korea was responsible and, therefore, refused to support military action against North Korea in response.
The United States team, however, decided to act unilaterally against Pyongyang without such proof of attribution. This, in turn, strained US-Japanese relations to a near-breaking point. During the immediate after-action debrief, several war game participants said clarifying who struck the Kori plant was critical. At the same time, the group understood that such attribution could not be assured.
The absence of definitive attribution puts a premium on providing effective, proven active and passive defenses for nuclear plants that might be hit and challenges the wisdom of constructing new nuclear plants in war zones without such defenses.
Recommendations: The United States and its allies, individually and collectively, need to stress test any active and passive defense systems for their civilian infrastructure, including nuclear facilities, against the probability that they will not be able to attribute exactly who launched any given attack. Undoubtedly, such stress tests will spotlight major failures. It will therefore be necessary to prioritize which failures can be significantly mitigated and which cannot. This analysis ultimately needs to be made public. Unless the public understands how their public utility systems might be attacked and defended, it cannot be an active part of that protection effort, which requires public support to be successful.
Editor’s note: This article is based on a wargame, “Strategic Distraction: Pyongyang’s Proxies Target South Korea’s Reactors,” organized by the Nonproliferation Policy Education Center (NPEC). The full report is available here.
The Bulletin elevates expert voices above the noise. But as an independent nonprofit organization, our operations depend on the support of readers like you. Help us continue to deliver quality journalism that holds leaders accountable. Your support of our work at any level is important. In return, we promise our coverage will be understandable, influential, vigilant, solution-oriented, and fair-minded. Together we can make a difference.
Keywords: China, East Asia, North Korea, South Korea, Taiwan, United States, nuclear power plant attack, nuclear reactors, nuclear safety, wargame
Topics: Nuclear Risk