Safe, Secure, Reliable Nuclear Weapons: A Discussion with Former Sandia Labs Director Jill Hruby
The United States maintains a
robust nuclear weapon stockpile to uphold its strategic deterrent and defense
capabilities, but in the words of Spiderman’s Uncle Ben Parker: “With great power
comes great responsibility.”
Headlining the latest in the Nuclear Threat Initiative’s
Seminar Series, ,
former Sandia National Laboratories director and current Sam Nunn Distinguished
Fellow at NTI, explained how national
labs complete required annual assessments of the U.S. nuclear weapons stockpile
and exactly what it means to keep the most powerful weapons on the planet safe,
secure, and reliable.
Why do we assess?
During the Manhattan Project and the
Cold War that followed, nuclear testing was the primary method for assessing
the reliability and performance of nuclear weapons. Safety and security systems
were improved with new systems and military requirements, and these systems
were tested both in the lab and in the nuclear tests. However, with the signing
of the Comprehensive Nuclear Test Ban Treaty (CTBT), President Clinton recognized
the need for a different method to ensure safety, reliability, and performance.
Since then, it has been the responsibility of the national
labs to annually assess the safety,
reliability, and performance of the nuclear weapons stockpile. Each year,
lab directors submit a “letter,” or report, to the Secretary of Energy
detailing the outcomes of the assessment. The Secretary then submits his or her
own letter to the President along with each lab director’s unrevised reports.
Separately, it is the responsibility of US Strategic Command (STRATCOM) to
annually assess the stockpile’s military effectiveness.
Writing the Assessment
The assessment process is
continuous and goes on year-round. As lab director, Hruby received about 100
hours of briefings on assessment activities and then wrote the letter – a
process that could take four to six weeks. The letter is an opportunity for the
director to report the results of the assessment and discuss important or
concerning trends in the data, lab activities, or elements of lab culture.
It’s easy to understand why the President and the scientists
and weapons complex workers should be care about these assessments – but you
should care, too. They don’t always get much publicity, but past have hit, quite literally, very
close to home.
What is “Safe, Secure, & Reliable?”
Safety refers to how likely it is that a nuclear weapon will
produce yield (i.e. create an explosion) when it’s not supposed to. To date, no
nuclear weapon has ever fully detonated when it wasn’t intended, but some
incidents have scattered nuclear materials. These are called “broken arrows” and
they refer to accidents like , where a maintenance worker dropped a heavy ratchet
down a nuclear silo, rupturing the missile’s fuel tank, and ultimately causing an
explosion that launched the missile and nuclear warhead into the surrounding
Arkansas countryside. Luckily, in that instance, a safety mechanism within the weapon
prevented the nine-megaton warhead from producing a nuclear yield. But this was
not an isolated event. Nuclear weapons have been dropped accidentally on the
Carolinas in the United States, as well as Spain and Greenland.
When the national labs address the safety of nuclear weapons,
they work to minimize the risk of a weapon exploding by accident in both
routine situations (i.e. sitting in a silo) or in abnormal situations (i.e.
being accidentally dropped from a plane). They use the Walske Criteria, named for
nuclear expert Carl Walske, which stipulates that each weapon in the stockpile
must have a one-in-a-billion chance
of producing an accidental yield under routine conditions, and a one-in-a-million chance in emergency
situations. This requirement led to the safety theme known as Enhanced Nuclear
Detonation Safety (ENDS), which implements additional, layered components into nuclear
weapons that prevent them from accidentally exploding. Compliance with these
standards is part of the labs’ annual assessment.
Also on the assessment to-do list is examining reliability. Reliability refers to how likely it is
that a nuclear weapon will produce yield when it IS supposed to. It also
refers to the likelihood that a weapon will work as it is designed and that it
will detonate at the intended time, with the intended yield, on the intended
target. The reliability of a weapon is scored on a scale from 0 to 1 and the
assessment requirement is often set at a 0.9 or above. The higher the
reliability, in theory, the fewer nuclear weapons are needed to accomplish a
To determine the safety and reliability of the weapons in
the nuclear stockpile, Sandia will take about 13 weapons from each weapon
family, remove the physics package (i.e. remove the part of the bomb that makes
it capable of a nuclear explosion) from each weapon and then subject it to a
variety of tests, modeling, and extremely environments. They crash weapons into
walls, drop them out of the air, submerge them in water, spin them around in
centrifuges, set them on fire, and even strike them with lightning. Don’t
believe it? Check out for proof.
Though some may call 13 an unlucky number, it’s the magic
number of tests needed each year to meet Sandia’s “90-10-2” rule. The rule goes
like this: executing 13 random system-level tests annually will statistically
provide a 90 percent confidence level that within two years they will find any
defect that has affected 10 percent of the stockpile. Any more would be
wasteful, and any less would not be effective enough to detect possible
Finally, there is the security of the stockpile. Security can refer to both the internal
and external integrity of the weapon at any given point. In even simpler terms,
it means ensuring that the people who have access to the weapons are only the
people who are intended to have access to the weapons. (Hruby noted the
stockpile security breach of and graffitied a secure building containing
weapons-grade uranium.) Security is not part of the assessment completed by the
national labs, but is part of a separate Department of Defense assessment.
Typically, the security of a weapon is the responsibility of whoever is in
custody of the weapon at any given time, so all branches take measures to
ensure their security, whether or not they include it in an annual assessment.
Hruby noted that defects are most
often detected at the beginning and end of a weapon’s life, either due to
errors made during production or wear and tear over time, requiring constant
vigilance related to stockpile quality, but especially as the weapons age. The
guaranteed life of a weapon is between 20 and 25 years, and the average age of
a weapon in the U.S. stockpile today is over 25 years making the assessment
necessary for the safety of those handling the weapons and for the integrity of
US national defense. To remedy defects found during an annual assessment, the
laboratories engage in life-extension programs that involve taking an old
weapon apart, assessing each component for usability and effectiveness, and
replacing only the parts that need replacing before putting it all back
together. This can help extend the life of a weapon for many years without
having to spend the time and resources (not to mention the risk implications)
of creating more nuclear weapons to replace aging ones.
How long will life extensions continue to work?
Hruby posed this question to the
audience at the end of her lecture. She explained that she’s not sure how long
the life-extension programs will continue to serve as the best solution for the
aging stockpile. As more and more weapon components need to be replaced, she
said, the labs are running out of spare parts that are compatible with those
weapons. Additionally, each year, the stockpile loses about 13 weapons from
each family in order to complete the assessment. Eventually, the
infrastructural demands of preserving the stockpile’s integrity will outweigh
the resources available for necessary maintenance.
For more information on the US nuclear stockpile, visit the
created and updated in partnership with the James Martin Center for
Nonproliferation Studies (CNS).
Sign up for our newsletter to get the latest on nuclear and biological threats.
Blundering into a nuclear war in Ukraine: a hypothetical scenario
Vladimir Putin’s invasion of Ukraine is a prime example of a regional conflict that could inadvertently escalate beyond any of the protagonists’ expectations. History is replete with similar instances of humanity stumbling into devastating conflict.
NTI Seminar: Susan Eisenhower on “How Ike Led: The Principles Behind Eisenhower’s Biggest Decisions”
From D-Day to Little Rock, from the Korean War to Cold War crises, from the Red Scare to the Missile Gap controversies, few people have made decisions as momentous as Dwight D. Eisenhower.
NTI at 20: Lynn Rusten on New Opportunities to Reduce Global Nuclear Threats and NTI’s Ability to Foster Progress
The scope of the Global Nuclear Policy Program (GNPP) at NTI is vast: Reducing reliance on nuclear weapons, preventing their use and their spread, and ultimately ending them as a threat to the world. Lynn Rusten isn’t daunted.