Skip to main content

Could the US stop a nuclear missile attack?

Could the US stop a nuclear missile attack?


Our defense system hasn’t aced all of its tests

Share this story

If a nuclear-tipped missile were hurtling toward the United States, would we be able to stop it? Maybe, if we were very lucky. But some experts warn that the United States’ missile defense system isn’t as reliable as people might think.

Right now, a constellation of sensors and 36 interceptor missiles make up the ground-based midcourse defense system, or GMD. It’s intended to act as insurance against a small-scale nuclear attack from North Korea, or possibly Iran, according to the Department of Defense. (Neither country has missiles capable of reaching the US, although US officials say North Korea is getting closer.) It’s not meant to ward off an unlikely attack from the much larger and more sophisticated arsenals of Russia or China — nor would it be able to.

Still, it’s the only defense we have against an intercontinental ballistic missile or ICBM once it’s in the air. On May 30th, 2017, the US tested these defenses against an ICBM-like target for the first time. To stop it, a ground-based interceptor missile fired from Vandenberg Air Force Base collided with the incoming warhead and smashed it to smithereens.

The test appears to have been a success — but that doesn’t necessarily mean the GMD could stop an enemy weapon under real-world conditions. In fact, the Government Accountability Office — a nonpartisan government agency also known as the congressional watchdog — reported in 2016 that the GMD “has not demonstrated through flight testing that it can defend the U.S. homeland against the current missile defense threat.”

Here’s what you need to know:

How is our missile defense system supposed to work?

For a second, let’s imagine a frightening future where North Korea actually does have working ICBMs — and decides to launch one. Satellites with infrared sensors and radar systems deployed in Japan and on US Navy ships would spot the missile launch, and alert control centers in the US. Sensors, including a sea-based, high-resolution radar, would track the hostile missile as it flies.  

When the missile leaves the atmosphere, it enters longest phase of its flight called the midcourse. At this point, the missile breaks up into the warhead, debris, decoys intended to confuse our sensors, and the last stage of the burned-out rocket booster.

On the other side of the Pacific, people in control centers in Alaska and Colorado would work quickly to find the warhead and figure out where to intercept it. Then, they give the order to fire interceptor missiles from Vandenberg Air Force Base in California, or Fort Greely, Alaska. There are 36 interceptors stashed in silos at these two sites, each carrying a “kill vehicle” on a three-stage rocket booster. (Tom Karako, a missile defense expert with the Center for Strategic and International Studies, describes the kill vehicle as a “funny looking telescope with a jetpack attached to it.”)

As the interceptor leaves the atmosphere and enters space, the 120-pound kill vehicle and the rocket separate. Using infrared sensors to find the incoming warhead, the kill vehicle moves into the warhead’s path by firing its own little thrusters. When the two objects collide, the kill vehicle should, theoretically, obliterate the warhead without causing a nuclear detonation.

Infrared image of a successful intercept in 2008, where a ground based interceptor collided with a target launched from Alaska.
Infrared image of a successful intercept in 2008, where a ground based interceptor collided with a target launched from Alaska.
Image: Missile Defense Agency

Okay, but can it actually do that?

We don’t really know. Since 1999, the GMD has been tested 18 times. But the Union of Concerned Scientists argues that these were conducted under artificial conditions where the timing of the incoming missile, for example, was known in advance. (Of course, for safety reasons, you can’t exactly launch a surprise ICBM at an air force base.)

And even so, the system has failed either eight or nine of those tests, depending on whom you ask. The Department of Defense’s Missile Defense Agency says eight failures, counting a “glancing blow” delivered to a target in 2006 as a success. But the Pentagon’s director of Operational Test and Evaluation testified in a congressional hearing that it was a “hit, but not a kill” — actually destroying the target apparently wasn’t one of the test’s objectives.

That’s why Karako says that there is a need for more testing — like the one on May 30th, which tested an upgraded component of the kill vehicle. The target took off from the Marshall Islands in the Pacific, and flew toward the US with the speed and trajectory of an ICBM. An interceptor missile launched from Vandenberg Air Force Base in California collided with the target and destroyed it. Laura Grego, a global security expert for the Union of Concerned Scientists with a background in physics, says the GMD system has been described as like “hitting a bullet with a bullet.” She adds: “It’s amazing that we ever actually do it.”

The difficulty is compounded by the fact that the interceptor missiles are essentially advanced prototypes, Karako says. “I think fundamentally what has challenged the program is the lost opportunities to go back and improve upon the basic design that was fielded back in 2004,” he says. “Everybody knows that that’s what needs to be done, and now finally we’re on a path to getting there — but we have to keep at it.”

The missile defense complex at Fort Greely, Alaska.
The missile defense complex at Fort Greely, Alaska.
Photo: Missile Defense Agency

How’d we get to this point?

For decades the US went back and forth about whether, and how, to defend itself against incoming missiles. Ever since the US developed the atomic bomb in 1945, our best defense has been the prospect of devastating nuclear retaliation.  

But to supplement this game of nuclear chicken, the Army asked Bell Labs in 1955 to start looking into possibilities for an antiballistic missile system. Since then, the US very briefly deployed nuclear-tipped interceptors designed to stop incoming missiles in their tracks with a nuclear explosion. In the ‘80s, Ronald Reagan announced his famous “Star Wars” plan — which included a proposal for an X-ray space laser powered by a nuclear detonation (it never got off the ground).

An anti-ballistic missile treaty signed by the US and the Soviet Union in 1972 — plus scientific and technological challenges — stopped us from getting serious about defending against an ICBM attack. There was also the concern that a sophisticated defense system could fuel a race for weapons that could overpower it.

The Clinton administration started making moves toward a National Missile Defense in the 1990s. But the real push came after 9/11, when George Bush pulled out of the anti-ballistic missile treaty and fast-tracked what became the GMD. This, Grego and her colleagues at the Union of Concerned Scientists argue, is what set the system up for failure by rushing missiles that were essentially still prototypes into silos. Karako counters that the speed of the rollout was necessary: “We are unwilling, as a nation, to accept vulnerability, complete vulnerability, to be blackmailed by North Korea. So we're going to do what we can to have some defenses, and we're going to improve it over time.”

Launch of a target for a 2010 test from the U.S. Army's Reagan Test Site at Kwajalein Atoll
Launch of a target for a 2010 test from the U.S. Army's Reagan Test Site at Kwajalein Atoll
Photo: Missile Defense Agency

Because interceptors were rushed out before rigorous testing, the Missile Defense Agency had to go back and fix them as flight tests uncovered issues. And those fixes have been both expensive, and incremental — so not all the missiles have the same hardware, according to a 2014 report by the Pentagon’s director of Operational Testing. It went on to say: “The reliability and availability of the operational [ground based interceptors] is low, and the MDA continues discovering new failure modes during testing.” Chris Johnson, a spokesperson with the MDA, says that the agency upgrades systems once new improvements are designed.

One of those upgrades was to address a pretty serious design flaw discovered a few years ago that caused the vibrations from the thrusters to throw off the kill vehicles’ navigation. That made the kill vehicle more likely to miss the warhead it was designed to intercept. A fix for the thrusters was flight-tested in 2016 — and it was described as a success until an LA Times investigation revealed that, in fact, one of the thrusters had failed and thrown the interceptor way off target. (“We disagree with that characterization of the 2016 test,” Johnson says. “The test met its objective, and the thruster performed as designed.”‎)

The next challenge was to see whether the upgraded kill vehicle could actually hit an ICBM-like target as intended — and it did, the Missile Defense Agency reports. But again, it was under pretty ideal conditions: the test took place during the day, there was only one ICBM-like missile, we knew where it was coming from, and where it was going. And, since it’ll likely take several interceptors to destroy an incoming warhead, the next hurdle to clear will be for the Pentagon to fire several missiles that will need to work together to stop one target.

Launch of an interceptor for a 2010 test.
Launch of an interceptor for a 2010 test.
Photo: Missile Defense Agency

What does this mean?

That depends on who you ask. If you ask Grego and her colleagues at the Union of Concerned Scientists, it means that the Missile Defense Agency needs more oversight. It should be required to test its acquisitions before it buys them, like many other defense programs. “They're meant to be tested first and put through their paces before they're bought and put into the field,” she says.

In the meantime, the government should avoid statements about the GMD that give a false sense of security, Grego says. The DoD's 2010 Ballistic Missile Defense Review said that "the United States is currently protected against limited ICBM attacks." A Senate report in 2015 said the GMD protects the entire US against an attack from North Korea or Iran. That’s not true. At least, not yet, according to the GAO, that says, “GMD flight testing, to date, was insufficient to demonstrate that an operationally useful defense capability exists.”

But Karako at the Center for Strategic and International Studies disagrees: “I don’t think anybody takes this lightly. This is a form of insurance policy. You don’t have fire insurance and then go lighting matches around your house recklessly.” Instead, he argues that the Missile Defense Agency needs to go back and methodically fix the interceptors based on what we’ve learned from testing the GMD, as well as from regional missile technologies.

“There’s been a number of decision points along the way, when the opportunity to fix these very fixable things could have been taken, but wasn’t,” he says. “I don’t think anybody really disputes that.” Making the system more reliable will take more frequent testing, a culture that allows those tests to fail, and enough breathing room to actually implement the fixes.

Still, it’s pretty clear that regardless of how the GMD performed on May 30th, the best option we have against an ICBM attack is never getting to that point in the first place.  

Update June 1, 2017 9AM ET: This post, originally published May 30th, 2017, has been updated to include a video and the results of the May 30th GMD intercept test.