Late in July the National Research Council released a report on a hitherto little-known, or perhaps unknown, U.S. capability: nuclear forensics. Nominally the report was a plea for more resources; too many skilled people trained during the Cold War were retiring, and too much equipment was becoming obsolete. However, the report was probably issued far more to disclose an important national capability. Nuclear forensics makes it possible to identify the source of a nuclear weapon based on evidence the weapon generates when it explodes. In effect it is a much bigger brother of the science that makes it possible to say where the explosive in a terrorist's bomb originated. The new disclosure may be associated with the Obama administration's view that nuclear proliferation must be curbed because of the threat of terrorists armed with nuclear or semi-nuclear weapons ("dirty bombs"). It could also be a response to the apparent failure of that initiative. In either case, the important news is that a nuclear explosion cannot be anonymous.
The primary means of defending the United States against nuclear devastation is deterrence. We try to assure our potential enemies that, however satisfying an attack may be, we can and will strike back. To do so we need to be sure of the source of any nuclear attack. That is relatively easy in the case of a land-based nuclear missile, because existing U.S. systems monitor launches, and the system being built for strategic defense will track missiles once they have been launched. The situation becomes much more difficult if the missile does not originate in the territory of the attacking country. At least from the point of view of retaliation, that was no great problem as long as only the Soviet Union had submarine-launched ballistic missiles. Now the situation may be changing.
For years, strategists have wondered how the United States would or could respond to a weapon launched anonymously. A classic example is a freighter offshore armed with medium-range ballistic missiles. We try to track merchant ships, but their ownership is often obscured by layers of corporate veils. Missiles can, at least in theory, be loaded on board far from the point of origin, in containers (the Russians even advertise a containerized cruise missile). The anonymous freighter then launches its deadly cargo a few hundred miles offshore. Much the same might be said of a bomb carried on board an airliner. Its point of origin might be concealed by swapping it with a similar airliner at an overseas airport.
The new report says that any nuclear weapon carries a signature that can be traced back to its manufacturer (at least, to the point of origin of its components). We may not be able to kill the terrorists who plan an attack, but we can hold responsible the government that supplied them with a bomb. No small band of terrorists has the industrial power to build anything as complex as a nuclear weapon (including procuring the critical material). It is therefore reasonable to assume that some government (against which retaliation is certainly possible) is always responsible for a nuclear explosion. The administration can argue that governments are safest if they abandon nuclear ambitions, hence the risk that some irresponsible elements within them will chance U.S. retaliation by allowing terrorists to use their bomb.
New Intelligence
This is important news. A nuclear explosion should be anonymous, because the bomb is vaporized in the explosion. It is not like a terrorist bombing, after which the FBI can comb through the rubble to find a detonator or tell-tale chemical traces. However, the U.S. government has more than 60 years of experience monitoring nuclear explosions here and abroad to learn about the bombs that produced them. A great deal can be deduced, for example, from the elements the bomb disperses (and in what ratios), the details of the way the bomb produces its shock waves, and from the radiation it emits. Bomb-making material is never entirely pure, and remarkably small amounts of trace elements can be detected and measured.
The explosion products also include the non-nuclear parts of the bomb, such as the non-nuclear explosive that triggers it and even the glue that holds its components together. An active nuclear intelligence program may help identify particular kinds of components with specific makers. For example, knowing that A. Q. Khan, the father of the Pakistani bomb, sold his designs to other countries makes it easier to guess how they are trying to make their own bombs, and the components they use will not quite match Dr. Khan's.
Nuclear forensics is part of a relatively new kind of intelligence called Measurement Intelligence. It was adopted as a kind of catch-all for types of intelligence that rely on specific physical measurements. Many in the U.S. Navy will recognize it as a version of the successful Cold War attempt to associate Soviet submarines with acoustic signatures. First it was discovered that submarines had low-frequency signatures that could be detected at great distances. The key discovery of postwar antisubmarine warfare was that a submarine's signature was consistent over time and hence could be distinguished from considerable random background noise. That is, if what was received was compared over time, the random background canceled out (a concept widely used in experimental physics).
This idea was the basis of, for example, long-range submarine detection by the sound surveillance system (SOSUS) network of sea-bottom arrays, by low-frequency acquisition and ranging (LOFAR) sonobuoys, and by towed arrays. Conversely, as the Soviets learned which signatures the U.S. Navy and NATO were using, it was able to fight back by silencing at least some of them. This game of move and countermove explains why the details of signature detection are highly classified. No one can blank off every possible signature, but directing attention to particular ones is not very helpful to us.
It also turned out that individual submarines had recognizable distinct signatures: Whiskey No. 35 sounded almost like Whiskey No. 53, but not quite. This distinction made tracking, as opposed to detection, possible, because a submarine detected at one time in one place could be seen to be the same submarine detected later at another place. Tracking mattered because, if it was widespread enough, it could reveal the larger pattern of Soviet submarine activities.
For example, in the late 1950s, as SOSUS was going into place, a possible surprise Soviet attack was a major concern of U.S. policy-makers. At the time the U.S. Air Force's Strategic Air Command (SAC) was the core of U.S. deterrence. The fear was that SAC could be neutralized by having its aircraft and missiles caught on the ground. Much of the enormous investment in the North American Air Defense Command was intended specifically to give SAC sufficient warning of a bomber attack. In 1958 a report on U.S. vulnerability to such an attack argued that SOSUS would provide warning, because the Soviets would surge their large submarine force (which stayed mainly in port) to sea before attacking. The problem was exacerbated when the Soviets introduced ballistic missiles, and it helps explain why Polaris was so important.
More recently, much the same idea has been applied to radars as "specific emitter identification." Initially it was thought that only handmade radars like the major Cold War-era Soviet naval radars were different enough to distinguish, but later more precise measurements made it possible to identify a specific radar with its waveform. This type of measurement may turn out to be the only way to identify radars using software to control their waveforms. Anonymous freighters are not so anonymous after all, and the more powerful the computers, the more corporate veils they can identify.
In each case, precision measurement makes it possible to draw up a list of signatures that become specific identifiers. The greater the precision, the better the identification. Perhaps this type of intelligence should really be called signature intelligence, but the designation SIGINT has long been used for something different, signals intelligence. Signals intelligence does include the attempt to identify a specific signaler. For example, even before World War II there were attempts to identify specific radio signalers by their characteristic "fists." That provided useful information even when what they were sending could not be read. It also turned out that individual radio sets had characteristic—identifiable—quirks. This particular kind of signature intelligence was quite useful when the German battleship Bismarck was being run down in 1941.
In a wider sense, the ability to reconstruct the cause of destruction supports deterrence, because it is the basis for retaliation. Someone who feels he cannot be tied to what he has done cannot be deterred. We are seeing that logic unfolding now in the case of North Korea and the March sinking of the South Korean corvette Cheonan. It's now clear that at first the North Koreans felt confident that, whatever the interpretation of the sinking, it could not be tied unambiguously to them. It was, for example, entirely plausible that the South Korean ship had somehow triggered an unswept mine from the past. Unfortunately for the North Koreans, pieces of a torpedo were recovered. Worse, it was a torpedo that could easily be tied directly to them. The North Koreans can try to claim that they have been framed (the South Koreans, after all, had previously recovered such a torpedo), but it is unlikely that they will be believed. From a wider point of view, the South Koreans' feat of recovering the torpedo may well make it less likely that anyone will try a similar stunt in the future. Anonymous is not what it once was.