On January 16, 1966, a B-52G bomber, returning to its North Carolina base following a routine airborne alert mission, collided with the fueling boom of a KC-135 tanker 30,000 feet (9,144 meters) above the coast of Spain while attempting to refuel.1 Both aircraft broke up and the 40,000 gallons (151,000 liters) of jet fuel in the KC-135 exploded, killing its four man crew. Four members of the B-52s seven man crew were able to parachute to safety. Of the four unarmed B28 hydrogen bombs carried by the B-52 (a weapon with yields between 70 kilotons to 1.45 megatons), three crashed on the ground in the vicinity of Palomares, a poor farming community 1 mile (1.6 kilometers) off the coastal highway. The fourth sank off the coast and was missing for nearly three months, before being located by the submersible Alvin 5 miles (8 kilometers) offshore in 2,850 feet (869 meters) of water.2 The high explosives in two of the bombs which fell on Palomares detonated, digging craters 6 to 10 feet (1.8 to 3 meters) deep and scattering plutonium and other debris from 100 to 500 yards (91 to 457 meters) away from the impact area (the third bomb was recovered relatively intact from a dry riverbed).

For three months, 1,700 U.S. personnel and Spanish Civil Guards worked to decontaminate the area. An estimated 1,400 tons of radioactive soil and vegetation was excavated and sent to the United States for disposal (at the Savannah River Plant) and crops of tomatoes were buried or burned. Through all this, U.S. personnel wore protective clothing and underwent regular radiation checks; such measures were not taken for the Spanish workers. (The Air Force commander in charge later stated, "the U.S. Air Force was unprepared to provide adequate detection and monitoring for its personnel when an aircraft accident occurred involving plutonium weapons in a remote area of a foreign country.")

A radiation survey conducted jointly by the Defense Nuclear Agency (DNA) and the Junta de Energia Nuclear (JEN) found that no less than 650 acres (more than 1 square mile [2.59 square kilometers]) of village, crops, and farms were contaminated; however, during the survey winds picked up and scattered the plutonium dust, and the DNA's subsequent report noted: "The total extent of the spread will never be known." Yet there was only sporadic monitoring of villagers and no effort to determine what level of contamination was acceptable. As the DNA's report later noted, "The Spanish government had not established criteria for permissible levels, which is completely understandable because plutonium-producing facilities and nuclear weapons were non-existent in Spain. Significantly, there were no criteria in the United States for accident situations. The available criteria pertained only to plutonium processing plants." Thus, the DNA applied guidelines governing fallout from tests at the Nevada Test Site.

Following the cleanup effort, the AEC and JEN established a monitoring program for the villagers and their land, with the AEC providing the funds and JEN conducting the surveys. In November 1971, Wright Langham with Los Alamos laboratory visited Palomares to assess the program. He found that only 100 villagers (about 6 percent of the population at the time of the accident) had undergone lung and urine testing: 29 tested positive but were deemed "statistically insignificant." Air monitoring for plutonium dust had ceased two years after the accident even though high counts were occasionally obtained during periods of strong winds. Soil sampling was hampered by the fact that JEN had only one alpha spectrometer, which did not always work well. Langham reported that morale among the JEN staff for the task had diminished since the accident and that the United States ought to provide more money and equipment to keep the effort going. A DNA report completed in 1975 concurred: "Palomares is one of the few locations in the world that offers an on-going experimental laboratory, probably the only one offering a look at an agricultural area."

The monitoring program apparently continued at least through 1986. In 1985, at the instigation of Palomares's mayor Antonia Flores (who witnessed the accident as a child), the villagers who had been monitored were finally allowed access to their medical records, which, according to Francisco Mingot, the director of JEN's Institute of Radiobiological and Environmental Protection, were kept secret under pressure from the United States and, later, from the Franco dictatorship, which sought to avoid excessive concern. The cost of the AEC-funded monitoring effort from 1966 through 1986 totaled some $5 million. The total cost of the accident — excluding the aircraft, but including the search and decontamination effort and the settlement of more than 500 claims brought by the residents of Palomares — was estimated at more than $120 million.3

A little more than two years after the Palomares accident, on January 21, 1968, a B-52G bomber on a secret early-warning mission crashed on the ice near Thule Air Base, Greenland, following an uncontrollable on-board fire that quickly cut off the plane's electrical power.4 Six of the seven crew members were able to eject safely. After passing directly over the base, the plane hit the ice-covered North Star Bay at a speed of 560 miles (900 kilometers) per hour. The impact destroyed the B-52, triggering the explosion of its 35,000 gallons (132,500 liters) of jet fuel, as well as detonating the high explosives in all four of the B28 bombs it carried. That explosion propelled parts of the bombs — including plutonium, uranium, and tritium components — into the inferno. Bomb debris and plane wreckage burned for at least twenty minutes and covered an area 1,000-2,000 feet (305 to 610 meters) wide. The heat of the fire melted the ice which later froze, encapsulating some of the debris. One B28 secondary assembly (the thermonuclear part of the bomb) apparently melted through the ice and was later retrieved during an underwater survey.

A massive cleanup effort named Project Crested Ice (but known informally among workers as "Dr. Freezelove") involved more than 700 U.S. servicemen and Danish civilian workers from Thule, including U.S. specialists from more than seventy government agencies. Despite nearly impossible working conditions — total darkness until February, winds of up to 85 miles (137 kilometers) per hour, temperatures 28 to 70 degrees Fahrenheit (-33 to -57 degrees Celsius) below zero, equipment which functioned poorly (if at all) in the subzero temperatures, and intense pressure to finish the recovery effort before the spring thaw — within eight months of the accident, 10,500 tons (237,000 cubic feet [6,700 cubic meters]) of contaminated snow, ice, and debris had been collected in barrels and shipped to the Savannah River Plant for disposal (aircraft debris was sent to Oak Ridge for burial). Residual waste was allowed to melt into the bay with the spring thaw, on the theory that the large volume of water would sufficiently dilute the radiation to safe levels. The entire effort is estimated to have cost some $9.4 million.5

Within twenty years of the accident, some of the 500 Danish workers reported a range of debilitating ailments, including cancer and sterility, which they associated with their work on Crested Ice. Wearing no radiological protective clothing (face masks were often discarded as they restricted breathing in the harsh conditions), search teams (initially all American, later a mix of American and Danish workers) used radiation monitors to locate the debris and then retrieved it by hand. While Air Force personnel drove ice scrapers and loaders, the Danes filled barrels and other containers with debris, which were later dumped into a total of sixty-seven spare 25,000-gallon (95,000-liter) fuel tanks. Spills were all but unavoidable. After their shifts, workers and equipment were decontaminated "by simply brushing the snow from garments and vehicles," according to the official Air Force report. In only a few instances was clothing deemed too contaminated for continued use; in these cases it was discarded. To save time, nasal swabs were often used in lieu of urine samples, even though, noted an Air Force health physicist, "since everyone's nose ran profusely in this climate, there was a reasonable doubt as to the validity of this check."

In the years following Project Crested Ice, the wife of the Danish personnel manager at Thule began collecting names and medical histories from about 800 of the Danes involved in the effort. Many reported a broad array of ailments, including ninety-eight cancers. In December 1986 Denmark's Prime Minister Poul Schlueter stated that surviving Thule workers would be examined by radiological experts. Eleven months later, the Danish Institute for Clinical Epidemiology reported that Thule workers who had participated in Crested Ice experienced a 40 percent greater cancer diagnosis than a cohort of 3,000 Thule workers who were at the base before and after the accident and did not take part in the cleanup. A report released the same month by the Institute of Cancer Epidemiology found the cancer rate for Crested Ice workers 50 percent higher than that for the general population, but this study concluded that radiation exposure was not the cause.

In late 1987, nearly two hundred of the Danish workers sued the United States under the Foreign Military Claims Act for damages caused by their work during Crested Ice. This suit was disallowed by the Air Force, but the discovery process did dislodge hundreds of secret documents. While these shed no light on the workers' health problems, they revealed that U.S. Air Force personnel who worked alongside the Danes had not undergone any long-term monitoring, even though many of them would likely have been more highly exposed than the Danes. To date, the Air Force's refusal to release any information concerning the toxic or radioactive inventory of the bombs or the B-52 continues to stymie Danish hopes of pinpointing the cause of the workers' illnesses.


1 This discussion draws on Chuck Hansen, The Swords of Armageddon, vol. 8 (Sunnyvale, Calif.: Chuckelea Publications, 1995), appx. 3, pp. 135-41; Tad Szulc, The Bombs of Palomares (New York: Viking Press, 1967); Flora Lews, One of Our H-Bombs Is Missing (New York: McGraw-Hill, 1967); John May, The Greenpeace Book of the Nuclear Age: The Hidden History, the Human Cost (New York: Pantheon Books, 1989), pp. 148-54, 162-68; Edward Schumacher, "Where H-Bombs Fell in ?66, Spaniards Still Worry," New York Times, December 28, 1985, p. 2; Letter from Myron B. Kratzer, Assistant General Manager for International Activities, U.S. Atomic Energy Commission, to Jose M. Otero, President, Nuclear Energy Board, Madrid, Spain, December 2, 1968, located in the Department of Energy Archives, Record Group 326, DOS McCcraw Collection, Box 17, Job 1320, Folder MHS 3-9 (1968) March, Thule Incident; USAF Nuclear Safety, AFRP 122-1 Jan/Feb/Mar 1970, vol. 65 Special Edition (on Project Crested Ice), pt. 2, pp. 2-97; Peter B. De Selding, "A Broken Arrow's Dark Legacy," The Nation, June 25, 1988, pp. 888-891. [Back]

2 Thirty-eight U.S. Navy ships participated in the search for this bomb. Despite the fact that news of the missing bomb and the recovery effort became well known, the DOD refused for forty-four days to admit it had been lost. This led to absurd exchanges between the press and DOD officials in which the latter would make statements such as, "I don't know of any missing bomb, but we have not positively identified what I think you think we are looking for." Quoted in Anthony Lake, "Lying Around Washington," Foreign Policy, no. 2 (Spring 1971), p. 93. [Back]

3 Strategic Air Command, History and Research Division, Project Crested Ice: The Thule Nuclear Accident, vol. 1, SAC Historical Study 113 (1969), p. 39. However, Randy Maydew, a retired employee of Sandia National Laboratories who participated in the recovery effort of the missing bomb, has stated that this effort alone "cost more than $50 million" ($262 million in 1996 dollars). See John German, "Palomares ?Bomb Nuclear Four' - It Crashed, It Fell, It Sank, but (Whew!) It Never Blew Up," Sandia LabNews, January 19, 1996 (available on the World Wide Web at http://www.sandia.gov/LabNews/LN01-19-96/palo.html). [Back]

4 Scott D. Sagan, The Limits of Safety: Organizations, Accidents, and Nuclear Weapons (Princeton University Press, 1993), pp. 156-203. [Back]

5 Strategic Air Command, Project Crested Ice, p. 39. [Back]

Copyright © 1998 The Brookings Institution