Originally published in Scientific American.

August 2001

Taming the Killing Fields of Laos

Live bombs from the Vietnam War continue to kill people and hamper agricultural development in Laos. The cleanup project required deciphering decades-old computer files

By Daniel Lovering

IN AN OFFICE ABOVE THE dusty streets of Vientiane, the tranquil capital city of Laos, Michael Sheinkman watches accidents waiting to happen. With finger poised on computer mouse, he gazes at a monitor revealing a grainy black-and-white digital map of Laos. He clicks, and suddenly a constellation of tiny pink dots appears like a pox on the countryside. Each dot marks the likely site of an unexploded bomb.

The dots represent the legacy of one of the world’s most extensive bombing campaigns: they are U.S. bomb target coordinates from the Vietnam War era. U.S. forces dropped more than two million tons of bombs on the Ho Chi Minh Trail, the North Vietnamese supply route that snaked through the jungles of eastern Laos.

“The patterns show very clearly the targeting was on flat land and lines of communication,” says Sheinkman, an American geographer working on the extensive effort to locate and neutralize the unexploded ordnance. “The downside for the people of Laos some 30 years later is that flat land is at a premium.”

Land mines, perhaps more familiar as a threat, represent a mere 4 percent of the unexploded bombs in Laos. U.S.-made aerial bombs account for the majority of ordnance lurking in topsoil and bamboo thickets, although artillery shells, antitank rockets, hand grenades and other types of ordnance from China, France, Russia and elsewhere emerge regularly from hiding.

The human cost of these unexploded bombs is considerable: they have killed or maimed more than 10,000 people. In addition to their effect on public health, unexploded bombs greatly hamper the small and fragile Lao economy, especially by impeding agricultural development.

After the war ended in 1975, bomb disposal groups tried to clear swaths of land with marginal success. But it wasn’t until the mid-1990s that a more sophisticated, countrywide program called UXO Lao (the national office of which is in Vientiane) was launched by the Lao government with the support of international aid agencies. Finding and disposing of unexploded ordnance, or UXO, is the slow and exacting work of the program’s 628 bomb technicians and 23 foreign advisers.

For the past two years, Sheinkman, employed first by a government contractor called Management Support Technology and now by Federal Resource Corporation, both based in Fairfax, Va., has assisted the group in using U.S. bombing records and digital mapping technology to identify areas most likely to be strewn with unexploded bombs. His work is an outgrowth of U.S. funding that began three years ago, when the U.S. Department of Defense’s Humanitarian Demining Office began assisting UXO Lao with training and equipment. In late 1998 Sheinkman and his Lao advisees began charting old bombing runs that may have left unexploded ordnance in their wake.

Bomb Sniffing

THE DIGITAL MAPS splashed across Sheinkman’s office incorporate bombing data that had to be  painstakingly culled from electronic records kept by the U.S. military during the war. Roy Stanley, a U.S. Air Force reserve officer and statistician at the U.S. Department of Energy, began the project eight years ago, after a serendipitous discovery.

A part-time air force historian in Washington, D.C., Stanley had just finished archiving World War II combat records. He and his colleagues at Bolling Air Force Base’s History Support Office planned to write a history of air combat operations during the Vietnam War. He was browsing the office library and found an index of U.S. bombing records. “It had pages and pages of databases from the Vietnam War era,” Stanley marvels. “I thought, ‘This is just incredible.’”

The index, from a Rand Corporation study published in 1976, inspired Stanley to seek the actual magnetic tapes embedded with streams of arcane numerical codes used by the U.S. military during the war. Stanley realized that those tapes—if they still existed—should denote bomb and aircraft types, target coordinates and other information about U.S. bombing missions in Southeast Asia. Military analysts knew nothing of the antiquated records.

Stanley next checked closer to home, at the National Archives, in Maryland. He found open-reel tapes, some of them disintegrating, as well as correspondence confirming that many of the databases he sought had been erased during and after the war, as officials saw no reason to keep them. Thomas E. Brown, manager of archival services at the Electronic and Special Media Records Services Division of the National Archives, says some Vietnam air combat databases were never transferred to the National Archives from the Department of Defense. “I am certain that some databases related to air missions were destroyed,” Brown adds. Although the National Archives had informal policies aimed at preserving electronic records as early as 1968, they were not officially enforced until years later.

Despite these setbacks, Stanley did find two useful tape databases at the National Archives that had been listed in the Rand index: the Combat Activities File had details about missions flown in Southeast Asia from October 1965 through December 1970, and the Southeast Asia Database documented missions flown between January 1970 and August 1975. These intact databases were created on IBM System 360 and System 370 mainframe computers using software called the National Military Command System Information Processing System 360 Formatted File System, or NIPS. Developed for the government by IBM in the 1960s, NIPS did what database software does: it created, structured, maintained and revised data files. But the details of missions over Laos—on average, a planeload of bombs every eight minutes for nine years—were coded to save space because of the limited storage capacity of mainframes at the time. “What you have is nested data,” Brown explains. “You would have a fixed field with information about sorties and additional fields identifying each leg of the mission.”

Further hindering Stanley’s attempt to understand the databases was the scant documentation of changes in database codes. “Sometimes codes were reused and the data processors just assigned new values to the codes,” Brown says. “These changes were not always incorporated into the systems manuals, and early versions of the manuals were not always saved.” In the Southeast Asia Database, for example, the National Archives has a manual dated 1975, but it is uncertain whether the information applies to earlier years. So simple coded data included nested information, and the codes themselves may have variable meanings.

In 1994 Stanley landed a $10,000 grant from the Department of Defense to complete a feasibility study of the data, still unaware that it would eventually help UXO Lao save lives. He enlisted the aid of Management Support Technology, the government contractor that eventually employed Sheinkman in Laos, to decipher the databases. The contractor then sought the help of a former IBM programmer who worked on the original database. The programmer created software to convert
the NIPS data to a plain-text file readable by a modern personal computer. These efforts revealed another stratum to be excavated. “Once we got the data out, the data itself was encoded,” says Skip Jacobs, a government contractor who assisted Stanley in building the new database.

In 1996, almost four years after his search began, Stanley unearthed his Rosetta stone. In a vault at Hickam Air Force Base in Hawaii, where archival material from the war was stored, he found computer printouts with codes for Vietnam-era databases. “That helped us solve 85 to 90 percent of all the codes,” Stanley says. He and his colleagues funneled the data into a modern, searchable database.

In the course of converting data, Stanley found what he calls “typos, hiccups and some plain wrong” information that may have made its way into the command chain during the war. For instance, the AIM-4 Falcon, a 140-pound air-to-air missile, was mounted on U.S. aircraft, including the F-102 and F-4D/E, but the Combat Activities File database reports it was used on the F-100 and F-105, among others. Some errors may have been the result of poor data-entry techniques and inconsistent codes. “You’re still talking punch cards,” Jacobs says. “You’re talking key error or key entry errors that are in there.”

(An unexpected by-product of this research was a partial explanation for the 1998 reports by CNN and Time that contained erroneous information about a secret U.S. mission in Laos during the Vietnam War called Tailwind. The news agencies alleged that U.S. forces used deadly nerve-gas bombs against defectors in Laos. Stanley says the bomb code used in the report, CBU-15, temporarily stood for a standard cluster bomb but was also used to denote an experimental gas bomb, the testing of which was halted in 1970.)

Stanley’s work attracted the attention of U.S. military officers in 1995, and the U.S. Humanitarian Demining Office went on to commission Management Support Technology to prepare the ongoing, larger study for use in Laos.

Deploying the Data

AT THE UXO LAO National Office in Vientiane, Sheinkman enters Stanley’s data into a geographic information system. The program plots as pink dots the target coordinates linked to the bombing database. The resulting maps are printed and used by bomb technicians equipped with handheld GPS units, which are capable of pinpointing coordinates on the ground via navigational satellites. The maps, which also include empirical data collected in the field, show bomb technicians what they might find as they scour villages and rice paddies. But they cannot determine precisely what remains from the war. “We have no idea what exploded and what did not,” Sheinkman admits. And areas that look clear on the map could nonetheless be ridden with other types of explosives. Encouragingly, the data used so far have been accurate. “We’re finding [the bombs] pretty much where [the coordinates] say they are,” Jacobs says.

Although the bombing database has been used only in Laos—the country hit hardest by U.S. air strikes—data exist for Cambodia and Vietnam as well. A Vietnamese delegation visited the U.S. last year to discuss using the database.

Some commentators say it could take a century to clear Laos of its unexploded bombs. One UXO Lao adviser cites clearance projects in Europe, where bomb technicians still find unexploded ordnance from World War I. With Stanley’s historical detective work as a start, however, perhaps Laos may more swiftly end its fight against the ghosts of war.


DANIEL LOVERING is a Bangkok-based correspondent for the French news service Agence France-Presse. He recently completed a Pew Fellowship in International Journalism at the School for Advanced International Studies at Johns Hopkins University. Lovering is a graduate of the Columbia University Graduate School of Journalism and Oberlin College.



Making War in Peace. Slide set. Mennonite Central Committee, Akron, Pa., 1986.
Aftermath: The Remnants of War. Donovan Webster. Vintage Books, 1996.
A History of Laos. Martin Stuart-Fox. Cambridge University Press, 1997.
Shooting at the Moon: The Story of America’s Clandestine War in Laos. Richard Warner.
Steerforth Press, South Royalton, Vt., 1997.
GIS and Cultural Resource Management: A Manual for Heritage Managers. Paul Box. United Nations Educational, Scientific and Cultural Organization, 1999.[/vc_column_text][/vc_column][/vc_row][vc_row][vc_column width=”1/1″][vc_separator type=”large” style=”normal” color=”light” align=”center” margin_top=”5″ margin_bottom=”25″][/vc_column][/vc_row][vc_row][vc_column width=”1/1″][vc_column_text]Days of Destruction

Finding the bombs makes it possible to detonate them safely CHANTHAVONG INTHAVONGSY can easily identify her prey on this day. In Champassak, a quiet province in southern Laos, Inthavongsy shines a flashlight into the tail section of what she recognizes to be a U.S. MK-81 aerial bomb sitting broadside in a bed of dead leaves. She peruses the bomb, then declares it safe to transport. Of course, not all unexploded bombs are so easy to handle. Some bombs have sensitive mechanical fuses and must be destroyed where they are found. Such diagnoses are easy for the 22-year-old, who leads a team of four technicians that frequently find the 250-pound bombs nosed into the ground or sitting in full view on the jungle floor. U.S. bombing data and reports from villagers help to guide teams such as Inthavongsy’s that are taking part in UXO (for “unexploded ordnance”) Lao, a countrywide program.

UXO Lao finds scores of different types of U.S. aerial bombs, some of which weigh a hulking 2,000 pounds. The millions of unexploded antipersonnel cluster bombs, locally nicknamed “bombies,” in Laos can be particularly nettlesome because of centrifugal fuses that were designed to arm the bomb as it fell through the air. Technicians need to examine such fuses closely to determine whether the bomb is armed. Although UXO Lao is familiar with most U.S. bomb designs, some are still classified by the U.S. government. Bomb technicians have also discovered several large aerial bombs that they have not been able to identify.

Inthavongsy’s colleagues hoist the rusted bomb into the back of a GPS-equipped Toyota pickup truck and head for a demolition site a few miles away. At the site, a remote jungle grove with a nearby protective log bunker, the crew follows a routine procedure to destroy the battered MK-81 bomb. Its goal is to avoid a socalled high-order detonation, in which the bomb explodes at full force. Instead the team will attempt to set off a low-order detonation that will quickly burn the bomb’s contents and produce only a relatively small blast.

As a precaution, Inthavongsy clears the area of villagers and takes cover in a bunker in preparation for a bone-rattling blast that could still blow a five-foot hole in the ground.

To elicit a low-order explosion, the bomb technicians use a device dubbed a baldrick (named for a character on the British television program Black Adder, whose “cunning plan” for almost any problem involved a small explosive). The baldrick, a small aluminum tube crammed with 80 grams of plastic explosive and capped with a thin sheet of copper, pierces the outer casing of a bomb and ignites the plasterlike mixture of TNT and other explosive material inside. This action splits the bomb’s casing and burns away its contents almost instantaneously, doing relatively little harm to the surrounding area. “The munition is usually attacked explosively from the outside using a charge sufficiently violent to cause a chemical reaction but not hard enough to cause [a high-order] detonation,” says Sidney Alford, a British explosives engineer who develops bomb disposal techniques used in Laos.

A member of Inthavongsy’s team places the baldrick 80 millimeters from the bomb and inserts a blasting cap wired to an electrical switchbox in a faraway bunker. After a countdown by Inthavongsy, a zipping sound rings out from a hand-cranked switchbox as an electrical charge shoots down the wire and unleashes a thunderous boom. Pieces of shrapnel shriek through the air. Moments later Inthavongsy and her colleagues emerge from the bunker and return to the site. The demolition is successful. The ground is charred, and the grass flickers with small fires, but the area is largely intact. —D.L.