What is the purpose of underground nuclear testing?

Illustration by Slug Signorino

Cecil replies:

What, you think nuclear weaponry is a mature (I use the word loosely) technology? Well, it isn’t, and considering the untold billions we’ve spent on nuclear weapons research since 1980, it better not be. The purpose of underground nuclear testing is to: (1) help build better bombs, (2) test the old models to see if they still work, and (3) see what happens to something if you drop an atom bomb on it. The U.S. is thought to have conducted more than 900 nuclear tests since 1945, all but 10 of them at the Nevada Test Site, a vast federal reserve about 100 miles north of Las Vegas. These tests have helped make nuclear weapons the sophisticated, reliable instruments of destruction they are today.

The U.S. hasn’t actually exploded a nuclear device since 1992 and may not ever again, now that the comprehensive test ban treaty has been signed. But that doesn’t mean we’ve lost interest in testing. On the contrary, the Energy Department is working with several universities to figure out ways to test bombs using simulations on supercomputers.

Many people are surprised to learn that serious nuclear weapons research is still underway. What’s the point, if we’ve already got enough bombs to destroy the world several times over? This thought is so naive. Strategic planners assume that most of our nuclear weapons won’t survive a first strike and we’ll have to wreak what havoc we can with what remains. Thus the search for better bombs.

This is not necessarily a bad thing. Early nuclear weapons were big and dumb. The idea basically was to get them within fireball distance of a major city, industrial complex or military facility, on the theory that if you obliterated every work of man for 10 miles around you were bound to take out a few targets of military significance. Strategic planners finally realized this was a waste of good neutrons and ever since have been trying to build bombs and warheads capable of being delivered more efficiently.

For example, one project being considered in the late 80s was the strategic earth penetration warhead, or EPW. These “are designed to penetrate some tens of feet underground before detonating,” says my bomb guide. “The principle underlying EPWs is that a warhead exploded underground couples more of its energy to the earth than does one exploded in the air or on the earth’s surface, creating more ground shock per kiloton of explosive yield.” Perfecting one of these babies will require tests, as will Star Wars fantasies such as nuclear-pumped X-ray lasers, should they ever be built.

Another reason for nuclear testing is to make sure that bombs already in service will work. In 1981, the government became concerned that model W80 warheads, which had been deployed the previous year, would fail at low temperatures. You can imagine the reaction if we nuked our enemies one frosty morning only to have the things drop like rocks into somebody’s borscht. As a consequence of testing, the problem was headed off at the pass.

In another case in the early 60s, scientists switched the chemical explosive on the W52 warhead because it was prone to premature detonation. (Chemical explosions are used in atom bombs to trigger the nuclear ones.) But an underground test revealed that the new chemical explosive resulted in a distressingly low-wattage atomic blast. Designers came up with a fix, did another test, and were pleased to find that the desired lethality could now be achieved.

So there you go. A nasty business, I agree. But the bomb jockeys say, and it’s hard to argue with them, that if you’re going to have nukes, they might as well work.

What’s bad about bombs

Dear Cecil:

You are usually so hip I checked your naive repetition of the excuses for bomb testing five times before I was sure it was not satire. Let’s get two points out of the way first: (1) there is no known case of a nuclear bomb being tested and not going “boom.” The likelihood of a nuclear warhead failing is almost nil. (2) The purpose of nuclear weapons is to threaten to use them. Combine these and the scenario that weapons tests are supposedly preventing runs like this: “Ten percent of the weapons with which the Yankee dogs are threatening us haven’t been tested. Our scientists are convinced there is one chance in a thousand they won’t work. If we attack now, 10 percent of our cities have one chance in a thousand of surviving until U.S. satellites redirect the last sub-based warheads. I suggest we attack now!”

You mention the danger of most of our nuclear weapons not surviving a first strike. Russia could put up a credible first strike against American ground forces. They haven’t up to now, and it is hard to figure why we need to test any weapons against the likelihood of them doing so with decreased armaments, decreased motivation, and increased complexity in command and control. For that matter, we have a sufficient threat in telling them we will suspend all aid to any country attacking us with nuclear forces.

The fact is, we have a bomb-testing bureaucracy because we once needed to test bombs; we now test bombs because the bureaucracy wants to have a purpose.

— Frank Palmer, Chicago

“We will suspend all aid to any country attacking us with nuclear forces?” Tell me, Frank, what do the other Martians think about this?

As for your specific gripes: (1) The fear is not that our nuclear bombs won’t go boom but that they won’t make a big enough boom. We covered this in the original column. (2) Loss of confidence in the nuclear arsenal is cumulative. If we arrived at a point someday that none of the active warheads had been tested, the credibility of our nuclear deterrent would be significantly reduced. (3) I agree Russia does not pose much of a threat and that the nuclear arsenals of both countries should be dismantled. Pending that happy day, however, it’d be foolish to go to all the expense of maintaining nuclear weapons without being sure they’ll work.

Cecil Adams

Send questions to Cecil via cecil@straightdope.com.