Nuclear Bombs Crossing Over U.S. Borders
It's a weighty topic that deserves more attention and management
A subscriber to this grand community suggested that a topic of interest might be the question, “how easy would it be to get a nuclear weapon into the United States?” That’s a good question, and my off-the-cuff response is “it wouldn’t be hard at all.” There are people moving illegal drugs and undocumented immigrants across the border all the time. Not all of them are caught, so sure, a nuclear device hidden in an SUV or van could probably get across. And this fact is why you might hear politicians say, “this is the nightmare scenario that keeps me up at night.”
I’m going to tease you with that opening. The probability that a sub-state actor or an adversarial nation-state is moving a nuclear weapon with the intention of blowing up a major U.S. city is very, very small. It’s not zero, but it’s pretty damn small. Almost infinitesimal, but for technical reasons rather than border controls. The ability of a single nuclear device to devastate a city is enough, however, for the U.S. government to put significant funds across several executive agencies to prevent that possibility. I’m going to talk about that infrastructure, although my information may be a little dated. I haven’t seen much about U.S. government efforts to detect and interdict a nuclear device for some time, but I don’t think it’s changed much.
I talked about nuclear terrorism in an earlier post, largely to critique a recently-released National Academies’ report assessing the U.S. government’s strategy to address the threat of nuclear terrorism. That report does have a nice intro describing the science and history of this topic (if you’re interested in technical details), although I was not thrilled with their recommendations. I’ll start with this note in my past newsletter, that the fear of nuclear terrorism has been around for decades.
This general concern has existed ever since J. Robert Oppenheimer told Congress in 1946 that it was entirely possible that people could smuggle units of an atomic bomb into New York City and detonate it. When asked what instrument officials would need to use to detect a hidden atomic bomb, he replied “a screwdriver [to open up each and every crate or suitcase].” There wasn’t too much to do at the time, but when the Soviet Union fell, the concern about “loose nukes” led to the establishment of the Nunn-Lugar Act in 1991 and the Cooperative Threat Reduction Program to dismantle the former Soviet Union’s nuclear weapons and delivery systems. After the 9/11 attack, concern refocused on the possibility that al Qaeda might have a nuclear bomb and could bring it to the United States (spoiler, they didn’t).
There are many aspects to the USG’s approach to nuclear terrorism, which spans the “prevent, protect, mitigate, respond, recover” spectrum of U.S. national preparedness. There are a lot of executive agencies involved in this because U.S. politicians believe that redundancy and depth can be good things when it comes to the (remote) possibility that a city could disappear under a nuclear mushroom. We can map out the general flow of agencies and their responsibilities in this figure.1
There’s a ridiculous saying that “government officials have to be lucky all the time, terrorists just have to be lucky once.” That’s fundamentally untrue, as terrorists have to avoid being monitored in the first place, successfully purchase and assemble the necessary materials, move the device to its destination, and detonate it without being interdicted at each step. An FBI official once told me, in addition, the bigger the planned incident, the more people will be involved, which increases the possibility that the authorities will find out about this group’s plans. People tend to focus on the terrorist moving a device with the intent of starting an explosion, but there are a lot of steps prior to that moment, reducing the antagonists’ odds of success.
I don’t intend on getting into detail on every aspect of the USG preparedness for nuclear terrorism, although it’s of interest that the 2013 and 2024 National Academies’ reports both note that there is no lead government agency who works the interagency aspects of this particular policy focus. Every executive agency has its own authorities, budget, and agenda. It is unclear to me whether the National Security Council staff has any full-time coordination effort or whether there are any working groups in which these agencies look at the gaps and seams between their efforts. I’m sure there are sufficient funds to execute these missions, but no single executive agency is in charge. That said, I will focus on the detection and interdiction aspect of this effort.
The problem is significant if left undefined. One can use a form of “Mad Libs” to create any scenario one would like to use as a hypothetical. We want to evaluate programs that prevent (terrorists, criminal organizations, nation-states) from acquiring/building a (radiological dispersion device, improvised nuclear device, nuclear warhead) from (any country with nuclear reactors that breed plutonium, any nation with nuclear weapons) and moving that device (by air, sea, or land) from (any geographical region of the world) to the United States and blowing up a (nuclear reactor, major city, military installation), causing mass casualties and threatening our very Way of Life. The DHS planning scenario focused on a 10-kiloton improvised nuclear device. Why 10 kT? They made an assumption based on the minimum amount of highly-enriched uranium or plutonium required for a nuclear detonation of an IND. If a sub-state group had access to more “special nuclear material” or a military weapon, the yield could be much higher.2 The yield doesn’t really matter in this discussion, it’s just something they do. Let’s assume a 10-kT IND just to avoid all the debate about how other nations with nuclear weapons operate.3
Back in the post-9/11 days, the Bush administration created a “Global Nuclear Detection Architecture” or GNDA. This was described as a “defense-in-depth” to offer multiple opportunities to detect and interdict a potential nuclear device from entering the United States. The figure below is the usual graphic used to generalize its interests. Strangely enough, I don’t see the term “GNDA” used within the past 10 years or so. According to the latest National Academies report, the programs that make up GNDA are still in place (which we will discuss soon), but maybe because the Domestic Nuclear Detection Office (DNDO) who championed the architecture no longer exists (being consumed into the DHS Countering WMD office).
So how does this work? The general philosophy is to detect and interdict the device as far from U.S. shores as possible. There are three general layers; the interior (domestic United States) that DNDO handled, the intermediate regions through which U.S. cargo transited (global commons), and the exterior regions where foreign nations had the material, addressed by State, Defense, and Energy. Let’s move past the efforts by the State Department and Energy Department to convince nations with nuclear programs to control and account for their special nuclear material.4 Assume that there is an unknown sub-state group that built an IND and is moving across geographic boundaries to a target somewhere in the United States. Lots of unknowns. To start, we would like to assume that U.S. intelligence sources would detect a plot, identify the specific individuals who have obtained or created such a device, allowing U.S. Special Operations Command to track, contain, and render safe the device. Makes for a thrilling movie. But what if that doesn’t happen?
Much of the GNDA relies on the cooperation of U.S. allies and partners. DHS Customs and Border Protection has a “Container Security Initiative” that attempts to regulate all cargo departing from major foreign ports with “best practices” that essentially prescreen and clear cargo containers before they enter U.S. waters. DOE’s National Nuclear Security Agency created a “Second Line of Defense and Megaports Program” (now called Nuclear Smuggling Detection and Deterrence) Program) that offers radiation detection equipment, training, and technical support to foreign customs offices and overseas port authorities. These programs are voluntary and do not exist at all ports where U.S.-bound cargo ships depart.
Let’s say that through national technical means, the U.S. government has a good idea that a specific group is planning to move an IND in one of thousands of cargo containers on a large freighter. In this event, the U.S. Navy would swing into a maritime interdiction operation under the Proliferation Security Initiative, an agreement under which 116 nations have agreed to work together, using existing laws and regulations, to stop and board ships of interest and inspect their cargo. Now if the ship in question is on the open sea, it depends on what flag the freighter flies and where it puts into port along the way. The Navy can’t just fire upon and board civilian ships at will. If, however, the freighter sails within 12 miles of the United States, the Coast Guard can and will board that ship and look for radiation sources. The Air Force does have an air interdiction concept in which it works with other nations’ air forces to work out how it would stop a terrorist group moving a nuclear device by commercial or government air traffic.
In our hypothetical scenario, the terrorists have avoided detection and interdiction at sea, and they are trying to enter the United States at some point in the border. It might be through a cargo plane landing at a major airport, it might be on a train moving over the Canadian border, it might be in a large van moving up through Mexico. In this event, the former DNDO provided radiological detectors to every customs agent at every land border crossing and established radiological portal monitors at every major port and transit station that moved large amounts of cargo. The instrumentation was pretty sensitive, but non-intrusive, and could not identify specific radiological sources (is it uranium or americium or cesium?). If the radiological source was shielded, the signal might not be picked up. DNDO got into some trouble in trying to upgrade its portal devices, and the program foundered somewhat. I think the detectors are still out there, but the problem is, DHS cannot promise that every piece of cargo coming into the United States is being screened. Congress wants DHS to promise that 100% of the cargo containers are inspected for nukes, but the best that DHS can say is that the paperwork was screened for 100% of the containers and the questionable ones went through the rad portals.
The problem is one of scale. On a typical day, more than 90,000 truck, rail, and sea containers are entering the United States. All due credit to the work that CBP does, but they can’t watch everything. The 2024 National Academies report noted:
The UNSC 1540 has no mandatory compliance vehicle, and many countries lack the resources and capabilities to prevent the transport of nuclear materials within and across their borders (or just don’t provide an annual report)
The PSI effort is reliant on consent of the flagged ship’s owner and/or working with the port authorities of a participating nation is willing to remove all of the containers to get to the suspected cargo
The MegaPorts initiative relies on other nations allowing their shipping schedules to be disrupted by laborious screening techniques, and the screening won’t detect shielded radiological materials
The Coast Guard’s IPS program isn’t adequately funded to carry out regular inspections at the individual port level
The CBP’s radiological detectors are at U.S. border control points and major airports and seaports, but not all of the hundreds of small, often short-staffed airfields and harbors along the coasts
Can a sub-state group get a nuclear bomb into the United States? It’s not a sure thing, but yes, given some careful maneuvering and luck, they can get it through. The multiple layers of security complicate this planning and hopefully discourage bad actors from trying to get a nuclear device to the United States. But the more limiting factor is getting ten kilograms of plutonium-239 or 50 kilograms of uranium-235 for a first-generation nuclear device. That’s very difficult to do. Amounts of that nature are not going to come from the black market or “lost and unaccountable materials” from nuclear reactors. There is a much greater focus on ensuring that these special nuclear materials are locked down and secured to prevent any development of an IND. And for the scenario of a nation with nuclear weapons losing or selling a nuke to a sub-state group, no, I don’t think that’s remotely possible, given the possibility that forensics and attribution could lead back to that country. In any event, no nuclear nation wants to lose track of what amounts to its crown jewels of its military arsenal.
I don’t agree with many of the recommendations of the recent National Academies report on assessing the U.S. government’s strategy for countering nuclear terrorism, but I do agree that it’s overdue for a review and public discussion. I don’t know for sure, but it seems like the COVID-19 crisis just mind-wiped everyone on issues not related to pandemics or global health crises. While fixing the U.S. public health system ought to be a top priority, Congress ought to be concerned with and overseeing the execution of federal efforts to counter nuclear terrorism as well. I don’t get the impression that this Congress (or the past few Congresses) are on top of this issue at all, and the recent draconian budget cuts could imperil this effort. All of the programs mentioned above require considerable taxpayer money. Congress needs to determine its expectations of the federal agencies’ efforts and fund them accordingly.
For a more detailed list of government agencies and their capabilities, see https://www.fema.gov/pdf/emergency/nrf/nrf_nuclearradiologicalincidentannex.pdf.
The Bush 43 administration worried about the possibility that the Taliban might infiltrate the Pakistani military and get access to a small nuclear warhead. The Pakistani government worried about the possibility of U.S. special forces operators roping in on their nuclear weapons stockpiles and “securing” them. But what if a loose nuke came from France? or India? Attribution becomes more important than immediate retaliation.
There used to be a hypothetical scenario in which the Soviet Union sent nuclear warheads to its embassies inside of the United States, using diplomatic mailing privileges that would get past any screening. Then the Soviet Union could cripple the United States prior to a major conflict by detonating them all. I don’t think this was ever a real concern other than a hypothetical, but it’s fun to think “what if.”
These include nonproliferation efforts such as convincing nations to modernize their nuclear reactors to use low-enriched uranium instead of HEU and to secure or give up their plutonium waste, and to have nations put up laws that prohibit or prevent individuals from accessing WMD-related material (UNSC 1540).
Really thought-provoking stuff. The tension between what's technically possible and what's realistically detectable at the borders is something we don’t talk about enough. Makes you think about where the real vulnerabilities lie.