Serendipity: The vagaries of radar

I RECENTLY HAD dinner with Neal Stephenson, the cult science fiction writer. He had just written Cryptonomicon, a thriller based on encryption, and I had just written The Code Book, a history of cryptography. Neal told me a story with a cryptographic angle and a serendipitous twist, concerning the Japanese attack on the Pacific island of Midway in June 1942. The Americans knew that an assault was imminent because they had cracked much of the Japanese code, but they were unsure about the exact location of AF, an encoded grid reference.

The Americans suspected Midway, and to check out this theory they sent out a fake distress message in English stating that the Midway fresh-water distillation plant had broken down. Sure enough, two days later they picked up a Japanese message stating that AF was short of fresh water. According to Admiral Chester Nimitz, the subsequent American victory was largely the result of superior intelligence.

In addition to attacking Midway, the Japanese had also attacked Dutch Harbour in the Aleutians, in an attempt to draw American forces away from Midway. On 4 June American radar detected the invasion force, but later it was learnt that the Japanese were further away than the radar suggested - so far away, in fact, that they should have been out of radar range.

News of this apparently impossible detection reached radar researchers back in America, including Neal's grandfather S T Stephenson. They soon learned that an English physicist, Henry Booker, had done some theoretical calculations and believed that the extended radar observations were the result of peculiar weather conditions. He argued that a layer of dry air above humid air would give rise to certain refractive differences, which in turn would channel radar further than anticipated.

To test Booker's theory, Stephenson travelled to Flathead Lake, a 30km body of water in Montana, where it was not unusual to find just the right sort of weather for so-called "radar ducting". At one end of the lake they set up a transmitter, and at the other end they fired an arrow attached to rope over a high branch. They used the rope to haul up a radar detector, which searched for radar ducting at various altitudes. The experiment was a success. What had accidentally been observed in the Pacific, had now been deliberately detected in Montana.

Fifty years later weather centres all over the world, including the Met Office, provide the military with data that helps them predict the performance of their radar systems. Radar ducting is also of interest to UFO researchers, who believe that it is responsible for many otherwise inexplicable radar observations.