When Secrets Crash

July 1, 2001

When a passenger airliner crashes, investigators from the National Transportation Safety Board quickly arrive on the scene to try to determine what went wrong. Press conferences and press coverage follow. The NTSB Web site notes that media are briefed at least once a day by one of the board members accompanying the investigating team and that a public affairs officer maintains contact with the media. Viewers of the nightly news often see aerial images of the crash site. The flight and airplane involved will be precisely identified by the airline and NTSB. Eventually, the public can expect a detailed report on the conclusions.

Things can be very different when the crash involves a military aircraft–particularly if it is an airplane whose existence or mission the United States has not yet acknowledged or that carries particularly sensitive equipment. Over the years, a variety of secret intelligence and military aircraft have crashed, and the specifics of US government responses have varied–sometimes as the result of the different circumstances of the crashes, other times as the result of different rules for dealing with the press queries concerning classified programs. However, preserving secrecy has been a constant objective.

Often times, details of the crash and investigation will emerge only many years later, after the existence and mission of the aircraft have been acknowledged and documents have been released in response to Freedom of Information Act requests or as a result of government declassification programs. The U-2, A-12 Oxcart, SR-71A, and F-117A all are aircraft whose existence was at one time a tightly held secret but which suffered crashes.

Spyplane on Display

By September 1959, the U-2 had been flying operational missions for more than three years. It had survived Soviet attempts to knock it out of the sky with surface-to-air missiles and MiGs. At the time, of course, its espionage mission was an unacknowledged one. It was, the US government declared, an airplane used for high altitude weather research and was operated by “weather reconnaissance” squadrons. It was a cover story that few believed; in May 1957, the London Daily Express wrote of the U-2’s espionage missions behind the Iron Curtain. However, weather reconnaissance was Washington’s story, and it was sticking with it.

One of the weather reconnaissance squadrons, whose covert designation was Det. C, was located at the US naval air station at Atsugi, Japan. Since 1957 that detachment had been flying missions over the USSR and China, photographing the Klyuchi ICBM test area in June 1957 and monitoring Chinese troop movements in the fall of 1958. By fall 1959, despite flying some actual weather reconnaissance missions in an attempt to add credibility to its cover, political problems were beginning to inhibit U-2 operations. Those operations were difficult to conceal. Atsugi was a busy airbase, with a variety of Japanese military and civilians on the base. US military deployments and movements in Japan were followed closely by outside observers.

On Sept. 24, 1959, Thomas L. Crull was flying a newly arrived U-2C, Article 360, on a local flight, heading back to Atsugi after setting an altitude record. As the U-2’s fuel ran low, the airplane suffered a flameout–forcing Crull to make a dead-stick, wheels-up landing at the Fujisawa glider strip, 10 miles from Atsugi. Crull emerged unhurt, but his airplane overran the runway and slid onto the grass.

Letting the airplane simply sit there unguarded was not an option. A short time later several security personnel, apparently wearing loud Hawaiian shirts and packing large revolvers, showed up and began to order the growing crowd at gunpoint to stand away from the secret aircraft. The tactic proved counterproductive as it only led to extensive publicity about the crash landing. Eventually, the airplane would be packed off to the US, repaired, and returned to service with Det. B in Turkey.

From there, that airplane would make its final flight. It came on May 1, 1960, and its pilot was Francis Gary Powers. Powers was flying high over Sverdlovsk, USSR, when his U-2 came under attack by some 14 surface-to-air missiles. The U-2 broke apart, but Powers parachuted down safely and was captured, given a trial, and sentenced to 10 years in a labor camp. He was freed in 1962 in an exchange for the Soviet spy, Rudolf Abel.

Less than a month before Powers’s fateful flight, another U-2 had made a crash landing, this time into a rice paddy in Thailand. In contrast to the Atsugi incident, the only publicity in this case was an article in a local newspaper reporting on the crash of a jet airplane. Because the area was inaccessible to large vehicles, the airplane could not simply be hauled out of the rice paddy. Instead, it had to be cut into pieces. Then, with the assistance of local villagers, those pieces were hauled by oxcart to a place where they could be loaded on trucks. One night, the trucks carried the dismembered aircraft through Bangkok to Don Muang airfield. There, it was loaded onto a C-124 cargo airplane and flown back to the US. The CIA, to show its appreciation for the villagers’ efforts, provided $500 to build a new school.

A Different Kind of Oxcart

On May 26, 1963, the New York Times carried a front page story under the headline, “New Test Delay May Doom RS-70,” which reported that, according to authoritative sources, “the first prototype for the Air Force’s RS-70 reconnaissance bomber will not be flight-tested until September at the earliest.” Intended to fly at 2,000 mph, the airplane might not fly at all, the paper reported, as a result of the repeated delays that plagued the program.

What the Times did not report, and apparently did not know, was that the CIA was already testing another reconnaissance airplane that was projected to fly at speeds greater than Mach 3, at altitudes of up to 100,000 feet, and with the equipment to photograph huge expanses of territory. This airplane was the result of a 1958 decision by President Eisenhower to authorize development of aircraft that would fly higher and far faster than the U-2 in the expectation that its speed and altitude would make it invulnerable, if not invisible, to Soviet air defenses. Nor did the paper report that one of these top secret A-12 aircraft, which had been developed under a program designated Oxcart and looked unlike anything that had ever flown, had crashed just two days earlier.

On May 24, 1963, Kenneth S. “Dutch” Collins was making a subsonic engine test flight, flying very slowly just above a solid layer of clouds. He was accompanied by Jack W. Weeks in an F-101 Voodoo chase airplane. When Collins saw that Weeks’s F-101 could not stay up with his A-12, he told Weeks to continue on to the base alone. Shortly afterward, when Collins flew into the clouds, his A-12 suddenly stalled, pitched up, and went completely out of control-the result of an erroneous airspeed reading. Collins was able to eject safely from the airplane, which went into an inverted flat spin and then crashed 14 miles south of Wendover, Utah.

Because Collins was on a low-altitude subsonic flight, he was wearing a standard-issue flight suit instead of a pressure suit. The more conventional flying attire prevented him from facing a difficult set of questions from the truck driver who stopped to pick him up and then at the highway patrol office. From there, he contacted officials at Area 51 in Nevada, where the airplane was based, to let them know that their top secret airplane had gone down.

A combination of means was used to prevent unwanted attention and discussion among the local population as well as accurate press reports on the incident. Individuals at the crash site were requested to sign agreements committing them to remain silent about what they had seen. Two farmers, who arrived near the crash scene in a pickup, were told that the airplane had been carrying atomic weapons-which was not true but effectively curtailed their interest in getting any closer to the CIA’s secret spyplane. Meanwhile, the press was told a different and less alarming but also false story-that the airplane that crashed was a very unclassified Republic F-105 Thunderchief. Even official records listed the crashed airplane as being an F-105.

Shattered Fighter

In addition to producing aircraft like the U-2, Oxcart, and SR-71, Lockheed’s Skunk Works produced the F-117A stealth fighter. In 1982, eight years after the experimental Have Blue program began testing the concept of a faceted aircraft to reduce radar cross section, Lockheed delivered the first of the new, odd-looking fighter­bombers. By July 1986, trade journals and writers had turned out a number of articles on what some called the “F-19” stealth fighter. The Testors company even produced a model of what the airplane was supposed to look like, but it bore no resemblance to the real thing. That fact undoubtedly pleased those working on the secret program.

On July 11, 35-year-old Maj. Ross E. Mulhare, assigned to the 4450th Tactical Group, took off from Tonopah Test Range in Nevada and flew his aircraft into California airspace on what would prove to be his last flight. Mulhare, a 1974 graduate of the Air Force Academy, told his friends and members of his family in New Jersey that he flew F-5E fighter airplanes in mock combat missions against pilots from Tactical Air Command. From April 1978 to March 1980, he had flown such missions from Nellis AFB, Nev., the official home of the 4450th. That was followed by F-15 assignments in the US and overseas. In August 1985 he joined the 4450th. The group was an F-117A squadron and Mulhare was one of the squadron’s pilots.

Shortly before his flight, Mulhare was overheard telling a colleague that he was tired and “couldn’t shake it.” Despite his physical condition, Mulhare took off at 1:13 a.m. Pacific Daylight Time–such late night flights were intended to prevent discovery of the airplane’s unique shape–and proceeded westbound into the eastern portion of the San Joaquin Valley. He flew down the eastern side of the valley toward Bakersfield. At about 1:45 a.m., Mulhare’s airplane went into a steep dive and smashed into a hillside about 17 miles northeast of that city, just inside the Sequoia National Forest. Mulhare was killed.

The physical damage to the aircraft was such that one of the crash investigators described it as “without exception … the worst crash I have worked.” He went on to observe that while there was only light fire damage to the airframe, “the structural breakup was almost absolute” and that ” ‘shattered’ may best describe the aircraft after impact.” As a result, identification of special components was frequently impossible.

The crash also started a moderately intense ground fire, which spread through the surrounding hills, eventually burning 150 acres of range. While the aircraft fire had gone out by itself, the range fire had to be controlled by the forest service, an effort not completed until about 16 hours after the crash. Local fire and police were first on the scene. At 3 a.m., authorities began assembling a “divert” team at Tonopah. It arrived at the crash site around 11 a.m.

In the wake of the crash, Air Force spokesmen had little to say. The head of Air Force public affairs said the airplane had only one crew member and “was definitely not a bomber.” Air Force officials at Nellis acknowledged that Mulhare had not been a member of the base’s aggressor squadrons, which emulated Soviet air combat tactics in order to train USAF pilots. An Air Force spokesman also acknowledged that Mulhare was a member of the 4450th Tactical Group but said that all information about the unit was classified, and he could not discuss any of it.

The Kern County sheriff’s office, whose jurisdiction included Bakersfield, did relay some further information from the Air Force–telling reporters that the “whole area has been restricted, including the airspace above the crash site” and that “there will be military aircraft in the area and anyone entering the area will be dealt with appropriately by the Air Force.”

The airspace restrictions called for low-flying aircraft to remain about six miles away from the crash site and other aircraft to maintain altitudes of more than 5,000 feet when within that radius. While civilian aircraft were kept away from the crash site, there were plenty of military helicopters arriving and departing. The Air Force brought in officials and other personnel from Edwards AFB, Calif., and Meadows Field in Bakersfield. As many as four helicopters at a time were in operation from Meadows Field. A helicopter gunship was observed circling the crash site the day following the crash.

At ground level, armed sentries carrying M-16 automatic rifles kept unauthorized visitors away. Not even firefighters were permitted within the guarded perimeter, which one paper described as a “ring of steel.”

At the crash site investigators collected evidence and evaluated the remains of the aircraft for clues to the cause of the tragedy. Then came the task of cleaning the site and leaving no pieces of the highly classified aircraft for scavengers, the media, or others to find. A clean-up team moved out a thousand feet from the last of the recognizable debris and then dug and sifted all the dirt in the area.

On July 23, controlled explosive charges were detonated on the hillside to free pieces of the aircraft buried as the result of the crash.

To mislead anyone who might try to search the area for pieces of the F-117A, the recovery crew had the remains of an F-101A Voodoo, one that had crashed and been stored at Area 51 for over two decades, broken up. They returned to the crash site and scattered the debris throughout the area. On Aug. 7 the Air Force announced it had withdrawn its guards from the crash site and would no longer restrict access to the area.

The very next day, a reporter and photographer from Bakersfield’s KERO-TV were transported to the crash site by helicopter. They later said they didn’t expect to find anything because they assumed the Air Force had cleaned the area thoroughly. But to their great surprise, they found countless pieces of debris scattered within 100 to 150 feet of a dirt helicopter landing pad built by the Air Force. They filled three bags with the material, and it was displayed on the station’s Friday evening news broadcast. They then turned the bags over to an Air Force public affairs officer. An Edwards spokesman said the debris would be examined as a precaution, but that there were no immediate plans to return to the crash site to recover more.

Another F-117 Death

On Oct. 14, 1987, Maj. Michael C. Stewart was flying his F-117A on a night training flight over Nellis. About three-quarters of the way into his mission, local air traffic control radars showed the aircraft descending to the left of the flight path. The aircraft crashed shortly after, at 8:33 p.m., into scrub desert terrain, broke up, burned for a short time, and exploded. Stewart was killed.

The extensive investigation that followed produced information on maintenance, the condition of the pilot, transcripts of recorded communications between Stewart and ground control, and testimony from Lt. Col. Roger C. Locher, leader of the search team. The ultimate result was a detailed 322-page report with 27 sections.

In contrast, information provided to the media by the Air Force was sparse. A decision in favor of declassification, which would take place a little over a year later, had yet to be made, and the world at large was still unaware of the airplane’s shape or actual designation. Neither the Air Force nor the Pentagon was going to help out. Air Force officials at Nellis issued a sketchy five-sentence press release about 2 p.m. on Oct. 15, only after news agencies had called the base for information. In Washington, the Pentagon observed, “There is a plane that is missing. … That is all that we are saying.”

Even though Mulhare’s July 1986 crash had taken place outside of Nellis and Stewart’s airplane crashed inside it, the latter proved the more difficult of the two to locate. At the time of the Air Force’s press release, a USAF search may have just located the airplane.

The Air Force started its search on the night of the 14th, using a C-12 aircraft carrying four pilots wearing night vision goggles. The airplane surveyed an area about 45 miles north of Scotty’s Junction-an area between Goldfield and Tonopah-based on a Forest Service request for confirmation of a fire at that location. At approximately 1 a.m. on Oct. 15, the search team secured the use of an H-3 helicopter and spent another two-and-a-half hours searching before retiring.

The search resumed at 6:15 that morning, and the airplane was finally located early that afternoon-45 miles to the northeast of Scotty’s Junction. Locher, leader of the search team, later noted that the aircraft could have been located much earlier if they had had access to a variety of existing information–including the observation of a pilot of a flash in the area of the crash and the detection of a hot spot in the same vicinity by a US satellite (presumably a Defense Support Program infrared sensor).

Recovery at Sea

Lt. Col. Daniel House and Maj. Blair Bozek, took off from Kadena AB, Japan, on the morning of April 21, 1989, in an SR-71A, the Air Force airplane that had supplanted Oxcart in 1968. Their mission was to perform peripheral reconnaissance of Southeast Asia. Not long into their flight, they experienced a series of problems that forced them to bail out about a half-mile off the coast of the Philippines. Fortunately, they were rescued in good condition by Filipino fishermen and eventually made contact with the US authorities. At times, their experience became surreal. It included standing in flight suits to make a call from a town’s only public telephone.

The airplane, however, had no parachute to brake its fall. When it smashed into the water, both engines sent the sensors and other equipment through the airplane’s upper surfaces. Those items were distributed across the ocean’s bottom at varying distances from the primary wreckage.

By 1989, the SR-71A’s existence had been acknowledged for 25 years. It remained the most advanced reconnaissance aircraft in the world, by a large margin. It carried optical, radar imagery, and signals intelligence sensors as well as defensive systems to allow it to operate over hostile territory. It was not an airplane that the US would want to allow material exploitation specialists in Moscow or Beijing to have in their hands.

On their way to Clark AB, Philippines, House and Bozek had the helicopter in which they were riding fly over the area of the crash. A P-3 also conducted search operations, as did a couple of naval vessels. Sonar operations on April 29 and 30 located the debris. USS Beaufort, a 280-foot salvage ship, equipped with 10- and 15-ton cranes, was directed to the site to extract the wreckage, as well as locate the sensors and defensive systems. Navy SEALs were aboard, since the recovery operations were conducted near a portion of the Philippine coast controlled by the Communist New People’s Army.

On May 2, both SR-71 engines were lifted out of the ocean and swung over onto the Beaufort. Two days later, salvagers brought up many of the sensors. The forward fuselage section was recovered May 7 and the main structure was raised the next day.

Shootdown

Another F-117A crashed March 27, 1999, but this crash was quite different from those which took the lives of Mulhare and Stewart. The airplane did not crash in the western United States, but in northwest Yugoslavia, near Novi Sad. The cause was not fatigue or pilot error but hostile action-specifically, a Serb­launched surface-to-air missile.

The most significant contrast was that the pilot was able to bail out and survive. Search and rescue teams were dispatched on specially equipped HH-60 Black Hawk helicopters and HH-53 Super Jolly Greens on a clandestine recovery mission. The helicopters were protected by a contingent of fighter aircraft as they headed toward the crash site. Fortunately, they were able to rescue the pilot, which produced a “huge sigh of relief,” according to the Pentagon’s chief spokesman at the time, Kenneth Bacon.

There was no hope of recovering an airplane downed in hostile territory, but to the surprise of some, the Air Force made no attempt to bomb the wreckage into oblivion. By 1999, of course, the existence of the F-117A had been acknowledged for more than a decade, and stealth fighters often appeared at air shows. A 1988 CIA assessment had concluded, “The Soviets likely have a good understanding of US stealth programs and technology from successful Western technology acquisitions.”

Senior Pentagon officials argued that it was no longer necessary to protect the F-117’s 1970s vintage low observable technology or its infrared targeting system. At a Pentagon briefing, then­Maj. Gen. Bruce Carlson, the Air Force’s director of operational requirements, observed that if Serbia passed some of the airplane’s technology to Moscow, the effect would be “minimal.”

Others were less sanguine. Destruction of the wreckage would, according to some analysts, have prevented reverse engineering of the sensitive technology carried on the airplane and the radar absorbent materials. An anonymous Air Force official was reported to say, “It’s our normal practice to bomb the wreckage when there is sensitive equipment on the aircraft.” A pilot who expressed surprise that the remains were not bombed wondered if the US had the coordinates of the wreckage site.

Within a week, the wreckage site was visited by a Russian trade delegation to Yugoslavia, and materials and system components were salvaged. What, if any, benefit Moscow might have gained remains to be seen.

Jeffrey T. Richelson is a senior fellow and consultant of the National Security Archive in Washington, D.C., and author of nine books on intelligence and military topics. This is his first article for Air Force Magazine.