Top Secret: Project Azorian
In history's most ambitious marine salvage operation, the CIA organized and implemented the recovery of the forward section of a lost Soviet ballistic millile submarine in 1974 -- all without Russian knowledge.
“If you go back there it would mean war.” Those words, spoken by a Soviet naval officer, demonstrated the concern of Soviet officials as they learned that the Central Intelligence Agency had salvaged part—and possibly all—of a Soviet ballistic missile submarine. The CIA had in fact raised a portion of the sunken missile submarine K-129, using a cover story that actually attracted world-wide attention, and under-taken with Soviet ships closely watching the lift ship without knowing that the salvage operation was underway.
During the first week of August 1974 a portion of the wreckage of the K-129 Soviet ballistic missile submarine was raised three miles from the ocean floor up into the hull of a U.S. salvage ship. Although only a 38-foot section of the submarine was recovered, Project Azorian was unquestionably the most ambitious and the most audacious ocean engineering effort ever attempted.
The engineering challenge was to recover an object, estimated to weigh up to 2,000 tons, from more than 16,000 feet. Previously, the only object known to have been picked up from such a depth was a satellite package, or “bucket,” recovered by the U.S. Navy bathyscaph Trieste II on April 25, 1972. That object weighed several hundred pounds. And, prior to Project Azorian, the deepest ocean salvage of a ship was the raising of the U.S. submarine Squalus from a depth of 245 feet off the New England coast in 1939.
The engineering challenge was to recover an object, estimated to weigh up to 2,000 tons, from more than 16,000 feet.
The Trieste II recovery and the Squalus salvage effort were undertaken in the “open,” with the latter being reported in real time in the press. Even though the Trieste II effort was not publicized, the necessity for surface support by the floating dry dock White Sands and an ocean-going tug made it obvious that a deep recovery operation was under way.
Project Azorian was carried out under intensive press scrutiny because the “cover” for the salvage was a seafloor-mining project sponsored by the notorious Howard Hughes. Thus, the salvage of the K-129, besides being of unprecedented scope and depth, was conducted in the public view and with intensive Soviet naval surveillance and with the Soviet Embassy in Washington, D.C., having been previously notified by an anonymous source that the United States was planning to salvage a submarine.
The story of Project Azorian—incorrectly called Project Jennifer in the press—has been told and retold in the press and in books since it was first publicly revealed in February 1975. However, this book contains the accurate account of the location and salvage of the remains of the K-129, with details that have never before been revealed in the open literature. It is the first published account to be based on interviews with key participants in Project Azorian, official documents not previously available, extensive commentary by Soviet naval officers directly involved with the K-129, and photographs and other illustrations not previously seen in the open press. Perhaps most significant, this book explains for the first time information derived from analysis of the actual acoustic signals received from the events causing the sinking of the K-129, which were recorded by a U.S. Air Force hydrophone system in March 1968.
Moments after midnight on the night of February 24–25, 1968, the submarine K-129 got under way from her moorings on the western side of Avacha Bay on Kamchatka Peninsula. The K-129 went to sea with 98 officers, warrant officers, and enlisted men. In that remote, Far Eastern part of Siberia, the bay, with the city of Petropavlovsk on its eastern side, was the second-most-important—if not the most important—Soviet naval complex in the Pacific region. Only the naval complexes in and surrounding the port of Vladivostok were larger.
The Kamchatka base—usually referred to simply as “Petro” by Western intelligence—was isolated from the Siberian mainland, accessible only by sea and air, with highly sensitive communications going through seafloor cables. The importance of the Petro complex—including the submarine facility at Rybachiy—could be seen by almost 40 submarines based in Avacha Bay in February 1968—possibly more than any other submarine base complex of any navy. The K-129 was one of these submarines. Completed in 1960, the K-129 was a first-line submarine.
The submarine was a Project 629A craft, the Soviet designation for the submarine design. In the West, the submarine class was called “Golf II.” The K-129 carried three liquid-propellant R-21 ballistic missiles (NATO SS-N-5 Serb) with a range of up to 755 nautical miles, twice that of the submarine’s earlier R-13 missiles. The R-21 could be launched from keel depths down to 165 feet, providing enhanced survivability for the submarine, with the craft traveling up to four knots at the time of launch. The warhead had an explosive force of one megaton—about 65 times that of the explosive power of the atomic bombs used against Japan in August 1945.
The morning before sailing, the K-129 had slipped her moorings to the barge that formed part of the “floating base” at Rybachiy and proceeded to the nearby missile facility to load three R-21 missiles and two Type 53-58 torpedoes with nuclear warheads. With the submarine loaded with stores and weapons for a 70-day patrol, at 11 pm the division commander, Rear Admiral Viktor Dygalo, boarded the K-129, made a final inspection of the submarine, wished the officers luck on the mission, and departed.
Shortly afterwards, with the pulsating beat of her diesel engines, the K-129 headed toward the entrance to Avacha Bay, passing the guard boat from which Dygalo waved a last farewell. At about 1:30 am the submarine passed the Three Brothers Rocks. As the K-129 entered the waters of the North Pacific, Captain 1st Rank Vladimir Ivanovich Kobzar ordered the submarine to dive. He was the last man to leave the bridge, according to Soviet Navy custom. The lookout position at Zarosshaya confirmed that the submarine had submerged. That was the last direct contact that the K-129 would ever have with the shore.
Kobar turned his submarine south, probably carrying out several drills and attempting to evade any U.S. surveillance submarine watching the Petropavlovsk base complex, in this case the USS Barb (SSN 596). The K-129 would remain submerged throughout the planned 70-day patrol except periodically—on a set schedule—to surface or at least broach the main deck to place the craft’s high-frequency antenna insulator on the starboard side clear of the water when the 35-foot antenna was rotated into the upright position. Once above the water, electromagnetic energy could be fed to the antenna. If the insulator was not clear of the water, the transmission could possibly cause the antenna coupler to short out and possibly to explode.
A reasonable estimate was that the K-129 would maintain an average speed of 4 to 5 knots while sailing submerged on snorkel—using her diesel engines for propulsion and battery charging—for at least 20 hours in every 24. This procedure was dictated by Moscow headquarters. Operating on snorkel in heavy seas could cause physical discomfort if not major problems for the crew, particularly when the snorkel intake periodically shut down to avoid flooding and the diesel engines sucked air from within the submarine. Crew fatigue and illness would follow such sustained operations in heavy seas.
The primary concern for submarine commanders was not speed, but avoiding detection by U.S. aircraft and surface ships, and maintaining the maximum charge on batteries so the submarine could remain fully submerged and on electric (quiet) propulsion for a sustained period if U.S. forces were encountered. Snorkeling was found to provide the maximum opportunity to remain undetected by aircraft radar and visual searches. When the submarine was snorkeling, the SIGINT team could use its radar intercept equipment for the early detection of U.S. surface ship or aircraft radars to provide ample time for the submarine to cease snorkeling and fully submerge.
Thus, en route to her patrol area the submarine would sail fully submerged on electric motors at about three knots for only a few hours each day. This submarine period was useful for training and to “clear baffles,” the term used for submarine maneuvers to determine if another undersea craft was trailing it, “hiding” in the blind area behind a submarine where self-machinery noises could mask the sounds of another submarine.
While there were no physical barriers to the submarine’s transit toward Hawaii, there were several U.S. anti-submarine “barriers” to a submarine escaping undetected into the depths of the broad Pacific. By 1968 the U.S. Navy was operating nuclear-propelled submarines as “gatekeepers” off major Soviet submarine bases in the Arctic and Western Pacific. Keeping in international waters, the submarine off Petropavlovsk in February 1968 was the USS Barb. But she apparently was out of position and did not detect the departure of the K-129.
Next was the SOSUS—the U.S. Navy’s Sound Surveillance System. It was developed to detect Soviet submarines at long ranges when they operated at speeds at which their propellers produced strong, low-frequency cavitation. (Cavitation is the formation and collapse of bubbles produced in seawater by high-velocity propellers.) The prototype full-size SOSUS installation—a 1,000-foot-long line array of 40 hydrophone elements moored in 1,440 feet of water—was deployed off Eleuthera in the Bahamas by a British cable layer in January 1952. After a series of successful detection trials with a U.S. submarine, the Navy decided by midyear to install similar arrays along the entire U.S. East Coast; the first array became operational in September 1954 at Ramey Air Force Base, Puerto Rico. Subsequently, the decision was made to extend the system to the U.S. West Coast and to Hawaii. The first Pacific area SOSUS array had been installed off the southern California coast in 1957; by 1968 the U.S. Navy had deployed at least eight SOSUS arrays in the Pacific area.
These early SOSUS line arrays were positioned on the sea floor at locations that accessed the deep sound channel and were oriented at right angles to the expected threat axis. Their individual hydrophone outputs were transmitted to shore processing stations called “naval facilities”—or NavFacs—via multiconductor, armored cables. At the NavFacs the acoustic signals were processed to create a fan of horizontal “beams,” each of which represented the composite sound signal from a small angular sector—on the order of two to five degrees wide—oriented in a particular azimuthal direction.
Narrow-band, time-frequency analysis in the spectral region from 5 to 150 Hz was performed on these multiple beam outputs. The ability of narrow-band frequency analysis not only to discriminate against broadband ocean noises, but also to identify characteristic frequencies associated with rotating machinery was the key to detecting and classifying targets. While the detection ranges of SOSUS are rarely discussed, there is credible evidence that in 1966 a nuclear submarine was detected at a range in excess of 3,000 nautical miles.
On February 26, while transiting south, the K-129 did transmit a “burst” radio message at midnight (261200Z)—to acknowledge that the submarine was at sea. The message was compressed into a burst transmission so the duration may have been less than a second; it consisted of the submarine’s call sign and a two-letter code. That message was received by Soviet military radio stations in the Far East and immediately relayed to naval headquarters in Moscow.
This transmission also was intercepted by U.S. listening stations in the Western Pacific area. A Naval Security Group intercept station indentified the submarine as the K-129, actually using her “side” number—722—that earlier had been painted on her sail when training in the Kamchatka area. The ability to identify the specific submarine was probably accomplished by “fingerprinting” the submarine’s radio transmitter on earlier operations that were connected to visual sightings by a U.S. submarine or a maritime patrol aircraft that could visually see her side number.
There is a possibility that when Captain Kobzar turned his submarine eastward along the 40° parallel he was scheduled to make another “burst” transmission, but there is no evidence that such a message was sent. Another message was expected to be transmitted from the K-129 at midnight on March 7–8 (071200Z). Admiral Dygalo said that the submarine had been directed to send a message when she was at approximately mid-point in transit to her station. This missed communication of March 7–8 “rang alarm bells throughout [Soviet] Pacific commands,” according to former U.S. naval intelligence officer Lee Mathers. Soviet sources stated that naval communication centers in the Far East reviewed their records in an effort to find an off-schedule transmission from the K-129—without success. Nearly constant messages requiring the K-129 to immediately establish communications were transmitted over the next 48 hours by the Pacific Fleet and Kamchatka commands, again without success.
The K-129 was “quiet.” On March 8 the Kamchatka commander apparently asked permission from Moscow to initiate a search for the submarine. Subsequent attempts to elicit communications from the submarine were unsuccessful. Thus, by 10 pm on March 8, according to Dygalo, the Soviet Pacific Fleet commands had “hit the panic button.”
The exodus of Soviet ships and submarines from Petropavlovsk was observed by the U.S. nuclear-propelled submarine Barb. The authors of Blind Man’s Bluff related that the Barb’s commanding officer, Bernard M. (Bud) Kauderer,
had never seen anything like it. Four or five Soviet submarines rushed out to sea and began beating the ocean with active sonar. The submarines would dive, come back to periscope depth, then dive again.
The Soviets made no effort to avoid detection, no effort to hide. Their cries filled the airwaves, shattering the air . . . with unencoded desperation.
As Barb and other U.S. surveillance craft listened, it was clear that the Soviets had no idea where to find their submarine.
Indeed, the Soviet search forces were looking within about 350 nautical miles of the entrance to Avacha Bay in the event that the K-129 had sunk shortly after her first and only radio transmission. The searching submarines subsequently sailed southward to longitude 40° N, and then turned eastward to cover the expected track of the K-129 up to her mid-transit checkpoint position.
The Soviet Navy’s frustration and lack of knowledge about the K-129 was obvious as ships, aircraft, and submarines freely communicated. In particular, the five submarines carried out the “position checks” with burst transmissions, revealing the standard track followed to reach the Hawaiian Station and marking the progress of the submarine searches.
The Soviet ships and aircraft found no trace of the missing submarine. The large, civilian-manned research ships Akademik Sergei Vavilov and Petr Lebedev, remained in the search area for a while longer. No trace of the K-129 was found by the Soviet search forces, nor was there any indication of the submarine’s fate detected by the Soviet acoustic detection system with planar arrays in the Pacific. This system, called Cluster Lance by NATO, included arrays placed near the entrance to Petropavlovsk.
But the death sounds of the K-129 in fact had been detected. The U.S. Navy cable ship Albert J. Meyer (T-ARC 6), manned by a civilian crew, was operating in the Eastern Pacific at latitude 29° 32'N and longitude 147° 06'W on March 11. The Myer was carrying out acoustic surveys for a planned SOSUS array site and had deployed a hydrophone to the seafloor, 4,000 feet down. The hydrophone recorded a series of major acoustic events. The first signal appeared to have been produced at about midnight with the second following exactly six minutes after the first. Later in March, the U.S. Navy began looking into the meaning of the massive Soviet search effort and the Myer’s acoustic detections as well as other possible sources of information on the March 11 “incident,” as the U.S. Navy labeled it.
There was, however, another U.S. acoustic source in the Pacific area—a series of hydrophones operated by the U.S. Air Force Technical Applications Center (AFTAC) to detect Soviet nuclear detonations. In 1947, because of concern over Soviet nuclear weapons development, then—Army Chief of Staff Dwight D. Eisenhower directed the Army Air Forces (soon to become the U.S. Air Force) to establish the capability to detect atomic explosions anywhere in the world. The new organization was formally activated on April 1, 1948, eventually evolving into the AFTAC agency.
Starting with aircraft configured to detect the fallout from nuclear explosions, the Air Force deployed devices to identify such detonations on the surface, underground, in the atmosphere, in space, and underwater. This requirement led to the Air Force installing seafloor monitors that terminated at the islands of Eniwetok, Midway, Oahu, and Wake. AFTAC also received an acoustic feed from a single hydrophone in the Navy SOSUS array that terminated at Adak, Alaska. These Air Force sensors detected the sounds emanating from the K-129 at distances from about 700 nautical miles (Midway) to 1,930 nautical miles (Eniwetok).
On May 14 Navy acoustic and intelligence specialists met in Washington, D.C., with AFTAC officials and asked them to search their data records for explosions or implosions, or evidence of radioactive debris or other activity that might indicate a submarine in extremis. Thereby U.S. naval intelligence learned of the AFTAC detection of two acoustic signals some six minutes apart. By comparing the detection times of these acoustic events recorded at four AFTAC sites and the Adak SOSUS array, Air Force technicians in the Washington suburb of Alexandria, Virginia, determined that the K-129 was lost within two nautical miles of 40° 06'N and 179° 57'E, and—as will prove suggestive—the initial event occurred within one second of 1200Z on March 11—precisely midnight on board the K-129. The event occurred at a distance of 1,590 nautical miles northwest of Pearl Harbor (Oahu) on a bearing of 320 degrees.
The wreckage of the K-129 was estimated to be resting at a depth of some 16,800 feet. The U.S. Navy had several systems that could be used to search for the remains at that depth. Most notable was the Trieste II, a bathyscaphe (Greek for “deep boat”). The Trieste was essentially a 6 ½-foot-diameter steel sphere, which carried up to three personnel and instruments, attached to a large gasoline float. The gasoline, which is lighter than water, would bring the craft back to the surface after the release of up to nine tons of steel pellets that served as ballast. Acquired by the U.S. Navy in 1958, the Trieste dove to the deepest location in the oceans, the Marianas Trench, reaching a depth of 35,840 feet in 1960. Subsequently, the Trieste was extensively modified and became known as the Trieste II. She had located the remains of the USS Thresher at a depth of 8,400 feet during dives off the U.S. East Coast in 1963 and 1964. But the Trieste II had two major limitations: First, she had very limited horizontal mobility—she was best described as a “deep-sea elevator.” Second, the craft was surface-based, transported to a dive area in a towed dock and openly placed in the water in preparation for a dive. Thus, she was readily visible to all in the area.
Another U.S. Navy capability for deep-ocean search was the use of cameras and sonar devices towed at great depths by the civilian-manned research ship Mizar (T-AGOR 11). That ship could tow search devices to a depth of 16,800 feet, but, again, her purpose was well known to the Soviets and fully observable to nearby ships and aircraft.
In 1968 another deep-ocean search capability became available to the Navy: the USS Halibut. The Halibut had been the first—and as events evolved—the only nuclear-propelled Regulus missile submarine. When the Regulus cruise missile was retired in 1964, the one-of-a-kind Halibut became available for other roles.
In early 1965, Dr. John Craven, the chief scientist of the Polaris—Poseidon submarine missile program and head of the Deep Submergence Systems Project (DSSP), was called to the Pentagon to discuss the Sand Dollar program. This was a highly classified or “black” Air Force-Navy program that sought to identify and, when possible, recover objects from the ocean floor. These were objects that the Soviets had “dropped” into the sea such as missile reentry vehicles, satellite packages, weapons as well as some objects similarly dropped by the United States. During the Cold War the recovery of such objects by the opposing side could provide valuable technical and even operational intelligence about the enemy. (Despite a lack of “stealth,” the bathyscaph Trieste II did recover several small items from the ocean floor, some of Soviet “origin.” On April 25, 1972, she lifted an item of several hundred pounds from a depth of 16,400 feet.)
While there were means available to possibly locate objects on the ocean floor, such as the Trieste II and the Mizar, as noted above, these were surface-operated vessels; what was required was a platform that could operate clandestinely, underwater, to search for lost objects and to possibly recover them. By this time, Craven, as the head of DSSP, had initiated the development of several systems to provide such capabilities. In the “white” world of DSSP was a program to construct two deep-submergence search/recovery vehicles that could operate to 20,000 feet to locate and recover small objects. They would be carried into the target area clandestinely on board a nuclear-propelled submarine, which could launch, recover, and replenish the 50-foot vehicles. And, in a collaborative effort between DSSP and Admiral H. G. Rickover’s nuclear power directorate, the Navy was building a nuclear-propelled search and recovery vehicle, the NR-1. This 146-foot, 393-ton submersible could be towed underwater by a submarine but was primarily surface supported. (In the event, the 20,000-foot search vehicles were not built; the NR-1, with a 3,000-foot operating depth, was in service from 1969 to 2008.)
Craven, fascinated by the Sand Dollar list of targets on the ocean floor, and supported by the director of the Defense Intelligence Agency and his scientific advisory board, initiated a black seafloor search and recovery program. He later outlined the three capabilities that had to be developed:
The most immediate was a means of conducting a clandestine search in the deep ocean. We had a particular interest in finding and recovering Soviet missile reentry bodies and guidance systems. . . . The second immediate requirement was to develop a capability for manned inspection and recovery of small and intermediate-sized objects from the deep ocean floor. The third was the capability to clandestinely place divers on the seafloor for the recovery of objects that required some form of handling.
These efforts obviously would require an underwater “platform” to carry out such missions, one that could not be easily detected by Soviet surveillance activities. This meant a submarine, and the USS Halibut was the obvious solution. The 350-foot, 4,775-ton submarine was available following the retirement of the Regulus missile. Although relatively new, completed in 1960, the Halibut was considered too large, too slow, and too noisy and lacked advanced sonar to be effectively employed as a torpedo-attack submarine.
With Craven secretly funding the work, and being careful to make minimum changes to the external appearance of the Halibut, the submarine was modified at the Pearl Harbor naval shipyard during a $70 million “overhaul” that lasted from February to September 1965; additional modifications were made at the Mare Island (California) shipyard and the Keyport (Washington) naval base. Her new features included:
• darkroom photographic facility
• facilities to enable the submarine to stow, release, tow, and recover “Fish” towed sensor devices —with winch and cable
• sail structure increased in height to house additional surveillance antennas
A thruster device was installed atop the Halibut’s hangar to improve maneuverability, but it was not used. It was found to be too noisy and was not needed. The Fish was a towed “body” about 12 feet long weighing two tons that contained cameras, strobe lights, and sonar for detecting seafloor objects. A tunnel-like chute installed in the bottom of the Halibut’s large bow hanger—labeled the Bat Cave permitted the launching and retrieval of the Fish. The development of such devices that could successfully operate and survive at depths to 20,000 feet was exceedingly difficult. Developing lights, cameras, and sonar to operate at great depths was difficult and very expensive. These costs as well as those for other black programs were “hidden” by Craven in the Polaris—Poseidon and DSSP white programs.
The Fish was designed to be towed for about six days, and then be winched back up into the submarine—cruising at a depth of about 200 feet a process that took several hours. For the K-129 mission the tow cable would be about 25,000 feet long. The Fish depth could be easily and accurately controlled by reeling cable in or out, and with slight changes in the Halibut’s speed. Once the Fish was back in the Bat Cave the camera film would be extracted from the Fish for on-board developing and its batteries would be recharged. The Halibut would normally carry two of the Fish.
The nuclear “spy sub” reached the search area some 1,600 nautical miles northwest of Oahu and, using satellite navigation, planted a series of acoustic transponders. Launched through the submarine’s bow torpedo tubes, these devices came to rest on the ocean floor where they would each transmit a “ping” when interrogated by the Halibut. Each “ping” was coded so that an acoustic grid could be mapped on board the Halibut to determine the area being photographed by the Fish. The accuracy of the grid was about 500 yards at the depth of three miles, with transponder responses being received by both the Halibut and the Fish. Within 36 hours the acoustic grid was established on the ocean floor.
Then began the tedious task of launching and towing the Fish. Sonar images transmitted through the Fish tow cable were carefully studied in the Bat Cave as technicians sought to identify potential targets on the ocean floor.
The Halibut’s search phase employed the Fish-mounted sonar to identify several seafloor objects of potential interest within the area. Preparations were made to begin using the Fish-mounted camera when there was a failure in the cable slip-ring assembly in the sail structure. The Halibut sailors, always innovative, jury-rigged a solution that enabled the search to resume. The photographic phase of the search continued with the Fish periodically reeled back into the submarine to recover the camera film. Sometime later, as the film was being developed, a photographer came running into the captain’s stateroom saying, “I think we have found our target.” The remains of the Soviet submarine had been located.
What the Halibut photos showed was the wreckage of the K-129 lying on her starboard side. The submarine had broken into two sections, probably when she struck the ocean floor. The K-129 had fallen to the ocean floor—a distance of some three miles—at a sink rate of about 12 knots. This estimate is based on actual data from the sinking of the U.S. target submarine Sterlet (SS 391) in 1969, which had been fitted with appropriate instrumentation, and analytical modeling.
The forward section included the missile compartment. The after section, about 100 yards away, contained the engineering compartments and stern torpedo tubes. The first few feet of the submarine’s bow—ahead of the pressure hull—had also broken off. The photography revealed that the submarine’s snorkel intake mast, a periscope, and antenna were in the raised position, indicating that the K-129 was at periscope/snorkel depth when the “incident” occurred.
Significantly, the after portion of the sail structure was torn away. The missing portion of the sail contained the two aftermost R-21 missile tubes; the tubes and their missiles were gone. The forward missile tube remained; while damage was visible, the cap to the No. 1 missile tube appeared intact, indicating that the missile could have survived the mishap and sinking. The forward section—which would be the target of Project Azorian—was 136 feet long.
Historically, all submarine salvage operations were undertaken with the use of cables or chains passed under the stricken submarine. This procedure required the use of large pontoons or specialized lift ships rigged for handling cables. Divers were required for these operations, limiting submarine salvage to depths of a few hundred feet at most. Before 1974 the deepest submarine salvage effort had been the raising of the USS Squalus (SS 192) from a depth of 245 feet off Portsmouth, New Hampshire, in 1939. The U.S. Navy in 1968 was developing the Large Object Salvage System (LOSS) to recover a submarine—with the use of divers from a depth of about 1,000 feet.
Even if divers were not required, such salvage procedures would be difficult if not impossible in the heavy seas of the northern Pacific, even in summer, and such operations could not be clandestine. The Soviets would certainly keep watch on such activities by communication intercepts, by long-range aircraft, and even by surface ships.
Thus, a totally innovative approach to raising the K-129 was required, one that could lift a submarine from a depth beyond all existing or even proposed salvage systems. And the effort would have to be totally denied to Soviet surveillance. Such an endeavor was far beyond the capabilities and investment resources of the Navy although, it was thought, the Navy could benefit greatly from the access to such a submarine, its weapons, cryptographic equipment, and possibly even its important publications.
By July 1969 the CIA had initiated a program to develop the means to recover the K-129 missile and possibly other equipment. The CIA task force for Project Azorian was established on July 1, with John Parangosky as head of the team. He had held key roles in the agency’s development of the U-2 and A-12 spyplanes and reconnaissance satellites. His deputy was Navy Captain Ernest J. (Zeke) Zellmer, a senior CIA officer who had served in submarines during World War II.
Parangosky and his team immediately began to examine possible methods for raising the remains of the K-129. At least four basic lift concepts were considered: (1) “brute force”—using massive winches on surface craft and wire ropes, (2) a “drill-string”—a three-mile “string” of connected pipes from the surface with a “claw” at the end, (3) “trade ballast/buoyancy”—buoyant material that would be carried down to the submarine using excess ballast that would be released after the buoyancy material was attached to the submarine, and (4) “gas generation”—creating the necessary buoyancy “at depth” to lift the submarine. Ironically, the drill-string was discarded early in the discussions for two reasons: because it was too difficult to envision how the massive length of interconnected pipes could be employed, and because the weight of the pipe itself would be too great for a salvage lift of the stricken submarine.
By July 1970— after a year of pondering methods to lift the K-129’s remains—“brute force” was clearly the favored system. At an ExComm meeting on October 30, 1970, the concept of lifting the estimated 1,750-ton “target object” from 16,500 feet was described, to be accomplished by mounting heavy-lift winches on a surface ship 565 feet long with a beam of 106 feet. About this time the probability of success of the operation was estimated at about 10 percent. The estimate would continue to rise, as would costs, but the promise of recovering a Soviet nuclear-tipped missile and possibly its guidance technology continued to justify the program.
Meanwhile, another approach to the salvage was being developed, employing the drill-string or pipe-string technique. On November 3, 1969, CIA officials met with Curtis Crooke, vice president of engineering of Global Marine of Los Angeles. Crooke later recalled that two CIA officials showed up at his office: “They walked in my door and closed it, and my office door was never shut. They wanted to know if my company could build something to lift something of so many tons and in about 15 to 20,000 feet of water.”‘
The CIA had approached Global Marine because by 1969 the firm was operating several deep-sea drill ships, primarily in support of the oil industry. The firm had pioneered the concept when it took delivery of the pioneer drill ship Cuss I, the converted 260-foot Navy barge YFN 730. The Cuss I was innovative because it could drill into the ocean floor without anchoring the vessel by the use of thrusters for dynamic positioning. She was able to keep a precise location based on sonar bearings to moored buoys. Such a positioning system would be required if the ship were to maintain position in very deep water, beyond the depths in which anchoring was possible. The Cuss I like other drill ships—was distinguished by the tall pipe-assembly tower, resembling an oil-drilling derrick, and the stacks of 60-foot “double-length” pipes on her deck. During March–April 1961 in National Science Foundation– sponsored Project Mohole—the Cuss I drilled into the earth’s crust off the coast of Guadalupe, Mexico. The ship successfully drilled 600 feet into the crust in water 11,680 feet deep.
Three other deep-sea drill ships were converted by Global Marine, and in 1967 construction was begun on the 3931/2–foot Glomar Challenger. Completed in August 1968, she would operate under a long-term contract from the National Science Foundation. This ship was built with thrusters for dynamic positioning that could enable her to maintain a precise location for several days while drilling seafloor core samples. Over the next 15 years, the Glomar Challenger would drill cores around the world, the deepest water being 20,483 feet. In 1970, in a remarkable example of station keeping and deep-sea drilling, the Glomar Challenger drilled a hole in the ocean floor at a depth of 10,000 feet off New York City, withdrew the drill bit, and maintained position precise enough to enable a new bit to be placed in the same drill hole. Significantly, the technique did not rely on holding the ship in a precise location, but in knowing where the bottom of the free-swinging drill-string (the drill bit) was located in relation to the hole. A sonar transponder lowered through the drill pipe was able to locate a 16-foot-diameter cone that had been drilled into the seabed.
The firm’s success in seafloor drilling provided the obvious path to deep-sea salvage. After the initial CIA discussion with Curtis Crooke, additional members of the firm were brought into the “fold,” including the firm’s chief engineer and principal naval architect, John R. Graham. By March 1971, Global Marine and the Sun Shipbuilding and Dry Dock Company in Chester, Pennsylvania, had completed the design of a specialized lift ship.
Derived from seafloor oil-drilling technology, the basic concept was developed by Crooke and his colleagues John R. Graham and James F. McNary. Their design was subsequently summarized as
a deep ocean mining ship in which heavy mining equipment can be raised between the vessel and the ocean floor. The ship includes a large well in the center thereof which passes through and is enclosed by the hull. This well is closable across the bottom by moveable gates. The vessel also includes a pipe handling system for moving mining pipe sections between a storage position and the drill string extending from the vessel to the mining machine and to support the mining machine.
Two vertically movable, tiltable legs, one located at the forward end and one located at the aft end of a large open well internal to the hulls of a vessel for deep ocean mining operations for docking and undocking a subsurface mining vehicle and for raising and lowering the mining vehicle into or out of the well. Each leg includes a panel at the lower end thereof for engaging pins on the mining vehicle.
It is not clear who—within the CIA or elsewhere—came up with the idea of having Howard Hughes “sponsor” the salvage effort under the “cover” of a seafloor-mining venture. CIA officials knew that there was a group within Lockheed developing the concept of seafloor mining for mineral nodules. The cover story was a stroke of genius. The CIA contacted Hughes and the billionaire recluse immediately offered his assistance. The Hughes empire was the perfect “front” for the endeavor: it was a collection of privately owned corporations, not responsible to stockholders or to the Securities Exchange Commission. And Hughes was known for undertaking unusual projects.
The purpose of the Glomar Explorer “required a design incorporating unique solutions which were well beyond the state-of-the art in numerous engineering and scientific disciplines, particularly mechanical engineering.” Much of the specialized design features and the equipment of the ship were derived from the offshore oil—drilling industry, a field pioneered by Global Marine.
Beyond the lowering of a “capture vehicle” or “claw” at the end of a pipe-string and then recovering the submarine, the system would have to raise the capture vehicle, submarine hulk, and pipe-string up through an open well. There would be strong dynamic forces at work in the North Pacific even in summer, and it would be necessary to hold the ship in an exact position over the three-mile pipe-string. As the K-129 was raised it would be necessary to ensure perfect alignment with the opening of the docking well or moon pool. And, of course, the recovery had to be unobservable by outsiders. The specific engineering features in the ship that were considered unique and representative of major advances in mechanical engineering design were:
• A massive gimbaled platform to isolate the suspended load from the ship’s dynamic pitch and roll; thus, regardless of the ship’s motion, the platform on which the pipe-lowering system was mounted—would remain stable.
• Hydraulic/pneumatic heave compensation system to prevent the ship’s heave (vertical) motion from dynamically affecting the suspended load (i.e., pipe-string).
• Hydraulic hoisting system to lower and raise massive loads via the pipe-string.
• Pipe-handling gear to convey pipe sections to the hydraulic heavy-lift system.
• Massive enclosed center well—moon pool—capable of being flooded and pumped dry, with sliding bottom doors or “gates.”
• Docking system to permit a massive, 2,000-ton capture recovery vehicle—with or without target object—to be mated with the ship in a dynamic seaway.
• Dynamic positioning system using bow- and stern-fitted thrusters to enable the ship to maintain station in a seaway.
These features that had previously been incorporated in drill ships were provided on a substantially larger scale in the Glomar Explorer. For example, the gimbaled platform’s outer ring was 40 by 40 feet, with four gimbal bearings that were unique in size and design, each with a capacity of 5,000 tons, i.e., able to support up to 20,000 tons—the lift system pipe-string, capture vehicle, and submarine. Similarly, the heave compensation system—essentially a giant spring—used two massive, hydraulic rams to mitigate the vertical motion of the ship while the pipe-string was suspended.
The hydraulic/pneumatic hoisting system had an 8,000-ton capacity, the expected weight of a 17,000-foot pipe-string, the capture vehicle, and the target object. The system was designed for a constant lifting/lowering speed of 18 feet of pipe per minute, although actual operations would be conducted at a slower rate. The ship had a deck stowage for 17,000 feet of pipe, assembled in 60-foot lengths, weighing in total about 4,250 tons. The handling system enabled the pipe to be easily and continuously moved to the hoisting system day or night, under most weather conditions.
The docking system for the capture vehicle was another highly innovative feature. It had to stabilize a 4,000-ton load suspended from a single point (i.e., the pipe-string) in a dynamic seaway and to hoist it into the narrow confines of the ship’s center well. The system had two semirigid structural arms or “docking legs” that could be lowered beneath the hull, at either end of the docking well, to engage massive “pins” at both ends of the capture vehicle and guide it up and into the center well. During docking and undocking the 200-foot docking legs could tilt up to seven degrees fore and aft, facilitating recovery in a seaway. When not in use these docking legs were retracted vertically and protruded upward, through the main deck, fore and aft of the pipe-hoisting system.
The Glomar Explorer was a huge ship: 618 feet, 8 inches long with a beam of 115 feet, 8 ½ inches. Originally, the ship was to have had a beam of just under 106 feet to allow passage through the Panama Canal’s 110-foot locks when the ship sailed to the Pacific. Charles Cannon, a naval architect at Global Marine, later recalled that the 10-foot increase in beam, “was all driven by lack of stability, and a fear that we still didn’t have a good handle on topside weights like the heave compensator gimbaled platform, pipe-handling equipment, A frame, hydraulic hoses, etc.” Thus, stability was enhanced by increasing the beam, but at the cost of requiring the ship to travel to the Pacific around Cape Horn, about a 50-day voyage.
The ship’s size, especially the moon pool, was dictated in large part to incorporate the various special features needed to recover the sunken K-129. Sized to accommodate both the 2,000-ton capture vehicle and a submarine hull segment weighing up to 2,000 tons and approximately 136 feet long, the moon pool was 199 feet long and 74 feet wide with a minimum vertical clearance of 65 feet. Two massive doors or “gates” closed the bottom of the moon pool. These gates were each 9-feet thick, 80-feet wide, and 98-feet long. They were motor driven and slid along tracks to close the bottom of the moon pool. Air was pumped into the gates, compressing a hard rubber seal, to force them upward to help seal the bottom of the moon pool so that it could be pumped dry.
At this time, the massive claw or capture vehicle was being assembled in the floating hangar known as HMB-1—the Hughes Mining Barge No. 1. Constructed by Lockheed in Redwood City, California, the capture vehicle that would grasp the remains of the K -129 was designed specifically to match the forward section of the submarine, a length of some 136 feet, weighing up to 2,000 tons. The “grabbers” of the capture vehicle had to be designed to align perfectly with the hull, with three points on the submarine designated as alignment points. This alignment scheme was based on the Halibut photographs of the submarine’s forward section. The term “grabber” was used to include the eight beams and davits that protruded down from the capture vehicle, all of which were hydraulically operated.
The core of the capture vehicle was a strongback or “spine” that was composed of two massive steel beams, 179 feet long and 31 feet wide. The total weight of the capture vehicle, with all of its components, would be 2,170 tons dry and 1,864 tons in water. Mounted on the beams were a variety of sensors, transducers, thrusters, cameras, and lights, along with the grabbers that would lift the submarine. Water jets were mounted along the sides of the grabbers and davit arms to help evacuate the soil and silt so the davit tips could penetrate deeper into the soil and pass under the submarine, which was embedded in the bottom. It was critical that the grabbers pass beneath and not pierce the submarine’s hull and thus cause damage to both the submarine and the lift arms.
There were eight grabbers—five that would be on the port side of the submarine and three that would be on the starboard side. Also on the starboard side were two beams that held and opened a steel containment net that would be deployed to restrain the hoped-for surviving missile should it begin to slide from its severely damaged tube. Mounted on the strongback were 26 lights to illuminate the work area and the wreckage, 12 cameras to provide pictures of the recovery operation to the control vans, sonar, pressure-proof spheres to house the electronics, eight hydraulic thrusters to help align the capture vehicle, and two 15-horsepower electric yaw thrusters to ensure that the capture vehicle did not rotate and “unscrew” from the pipe-string. The two 15-horsepower thrusters were to ensure that if the ship turned to port or just from its own weight there would not be a tendency for the pipe-string sections to begin unscrewing from its own weight and the motion of the capture vehicle!
During near-bottom operations, an acoustic positioning system gave commands to the eight hydraulic thrusters to position the capture vehicle over the submarine irrespective of the position of the ship three miles above. For this part of the operation the Glomar Explorer’s positioning system would sense the “bias” in the positioning system of the capture vehicle and commanded the ship’s thrusters to maintain the ship directly over the claw and align with it. The fully outfitted capture vehicle was supported by a three- point, hinged bridle that was attached to the end of the pipe-string by the pipe length called the Dutchman.
At each corner of the capture vehicle was an extendable “breakout leg.” These were huge cylinders, ten-feet in diameter, with a 40-foot maximum stroke, that could telescope to act like jacks to lift the (captured) submarine out of the silt. At the base of each leg was a “cookie cutter” that would extend to dig into the silt to help firmly anchor the breakout legs. Using pressurized water pumped through the pipe-string, the breakout legs could exert more than 2,000 tons of breakout force. Once the capture vehicle with the K-129 within its grasp was free of the bottom, the legs would be released and would remain on the ocean floor.
As the capture vehicle was assembled, it was given the nicknames “Crab” and “Clementine,” the latter soon becoming synonymous with the device. The capture vehicle was assembled within the HMB-1, which had been built by the National Steel and Shipbuilding Company in San Diego. Like the other key components of Project Azorian, the HMB-1 was a unique endeavor, intended specifically for a single mission: to provide a concealed space to assemble the capture vehicle and then, by submerging the barge, allowing the Glomar Explorer to moor above the barge and life Clementine up into the moon pool without being observed by others.
The Hughes Glomar Explorer got under way from a position off Long Beach on June 21, 1974, sailing northwest from the coast of southern California. There were 178 men on board. The ship’s crew, now commanded by Captain Thomas J. Gresham, stood bridge watches, navigated, manned the engine rooms, cooked, baked, served meals, did the laundry, and performed other tasks to keep the ship operating. While this crew performed the daily ship’s operations, the recovery team under the CIA mission director prepared for the unprecedented salvage operation.
The recovery team would operate the ship’s lift system, man the mission control and other specialized vans, handle the remains of the K-129, and perform a multitude of other functions. In addition to the CIA specialists, the mission crew included engineers and technicians from Global Marine, Honeywell, Lockheed Ocean Systems, and the Mechanics Research Institute; the last was a CIA cover company. At least one naval officer was on board: Captain Fred Terrell, the director of operations. There were also 10 scuba divers from various agencies on board to assist in undocking and recovering the capture vehicle and, hopefully, the target object.
On July 4—Independence Day—the lift ship arrived at the recovery site at 1:01 pm local time. President Nixon had departed Moscow on the previous day. It had been a 13-day voyage at an average speed of eight knots. There were no other ships on the horizon. The wreck of the K-129 had been located within six nautical miles of the intersection of latitude 40° North and the longitude 180°—the international date line. Upon arrival at the location, the Glomar Explorer immediately began deploying transponders on the ocean floor to establish a bottom navigation grid. Several of the transponders had flaws and were rejected before the ship was able to establish a six-transponder grid. Subsequently, the ship’s crew deployed two wave-rider buoys and calibrated the automatic station-keeping system.
The moon pool gates were slid open on July 8 and testing of the capture vehicle’s main components continued, first the cameras and then the hydraulics, water jets, davits, and other components were checked. But further tests were put on hold because of typhoon Gilda approaching the area. Heavy seas and strong winds plagued the ship until July 13.
By July 14 the rough weather subsided sufficiently to consider beginning to lower the capture vehicle on the following day. But on the evening of the 14th cracks were discovered in the forward and after docking leg structures, a serious problem. Repairs were essential and canvas screens were rigged on deck to protect men from the wind as they worked on the structures, raised above the ship’s main deck. The work, which involved extensive welding, took 72 hours to complete. And, the weather began to worsen as tropical storm Harriet approached.
Accordingly, the bottom gates to the moon pool were closed, and it was pumped out amid six-foot waves on July 16. Preparations were made to leave the recovery site if necessary. The weather did not worsen and on July 17 the ship was advised by U.S. surveillance activities that a Soviet missile range instrumentation ship was en route to the site. She was expected to reach the Glomar Explorer about 4 Al on July 18.
The Glomar Explorer’s arrival at approximately the same site as earlier Glomar-operated ships did not go unnoticed in the headquarters of the Soviet Pacific Fleet at Vladivostok. Then—Captain 1st Rank Anatolyi Shtyrov, the fleet’s deputy head of intelligence, had been responsible for dispatching the Soviet ships that had earlier observed the survey ship Glomar IL Shtyrov again brought the U.S. ship activity in the North Pacific to the attention of the fleet commander, Admiral Nikolai Amel’ko, who ordered a Soviet ship to undertake visual surveillance of the Glomar Explorer’s operations.
On the morning of July 18, “the fog burned off and out of the mist came this . . . what looked to me like a white, beautiful ship [with] very odd radomes on top and some funny looking antenna aft. Our security people were pretty excited at the time.” The shadowing ship was the Chazhma, a large missile range tracking ship. The Soviet ship had departed Petropavlovsk about June 15 to support a Soyuz/Salyut space event and was en route back to her base from a position near Johnston Island when she was directed to observe the American activities. The ship was fitted with a massive radome, tracking devices, and antennas, and had a helicopter hangar and flight deck aft. With the ship playing the role of intelligence collector, or AGI in NATO parlance, a Kamov Ka-25 (NATO Hormone) helicopter was soon pulled out of the hangar. As a precaution against the helicopter seeking to land on the Glomar Explorer, the mission director ordered that a stack of crates be placed on the ship’s helicopter deck to preclude that possibility.
The Chazhma initially closed to within approximately two miles of the Glomar Explorer. At 2:30 PM the Soviet ship further closed the distance to one mile, and crewmen began taking photographs of the U.S. ship; the helicopter took off. The Ka-25 carried a camera and began circling the ship while a crewman snapped photographs from all angles.
At 4:30 pm on July 18 the Chazhma started blinking a signal lamp message to the Glomar Explorer that was difficult to read because of the lighting conditions. The Soviet ship then passed 500 yards astern of the U.S. ship and signaled that she would communicate further. The Glomar Explorer responded by signal lamp, “I am going to communicate with your station by means of international code signals.”
For several hours the Glomar Explorer attempted to respond to the Soviet ship’s communications, and at 6:47 pm the Chazhma radioed that she was ready for the U.S. ship’s message. The Glomar Explorer answered, “We have no message. Understand you have a message for us.” The Soviet ship replied, “Stand by five minutes.” A short time later the Chazhma transmitted, “We are on our way home and heard your fog horn. What are you doing here?”
The Glomar Explorer answered, “We are conducting mining tests—deep-ocean mining tests.” The Soviet ship then asked, “What kind of vessel are you?” to which the Glomar Explorer replied, “Deep-ocean mining vessel.” When asked how long the American ship would be there, the answer was “We expect to finish testing in two or three weeks.” The Soviet ship signed off with “I wish you all the best.” The Chazhma departed the recovery area about 9 pm on July 18, en route to Vladivostok. She had been relieved of her shadowing assignment because she was low on provisions.
The weather cleared sufficiently on July 19 for the moon pool to be flooded and the bottom gates opened. System checks were carried out, and on July 21 the Glomar Explorer began to lower the capture vehicle on the pipe-string at a rapid rate. The ship’s Lift Log book noted, “01:27 UNDOCKED—very smooth. . . . 10:00 STARTED LOWERING CV.” Divers checked out the capture vehicle while men in the control vans kept careful watch on dials, pointers, and television screens. The recovery operation had truly begun.
On the morning of July 22 the Soviet naval seagoing tug SB-10 arrived on the scene to watch the Glomar Explorer. The Glomar Explorer’s captain repeatedly warned the tug to keep her distance, by radio and by flag hoists. When the tug persisted in closing on the drill ship day after day, and at night, Captain Gresham radioed Global Marine headquarters asking that a formal protest be made to the Soviet government, through diplomatic channels, to stop the tug’s harassing of the seafloor mining operation. Eventually the tug backed off, but remained in the area.
The lowering of the capture vehicle did not go smoothly. The plan had been to move pipe at the rate of six feet per minute, but the actual rate was much slower because the system encountered numerous problems. Most were minor problems. For instance, on July 22 there was a breakdown in the pipe-handling system that halted operations for almost two hours. Furthermore, the grabbers closed on themselves the capture vehicle would have to be brought back up to the ship and hopefully, repaired. Problems continued.
At 25 minutes after midnight on July 26 the command-control van reported that the capture vehicle’s sonar had contact with the seafloor. At this stage the pipe-string consisted of 211 “doubles”—the capture vehicle was at a depth of 12,690 feet. It was about 3,600 feet above the seafloor.
Suddenly, at 3:45 am on the 26th a cable broke on board the Glomar Explorer, dropping a 30,000-pound counterweight attached to the pipe conveyor onto the deck—sending a shock vibration through the entire ship. The Lift Log Book noted that it “made a hell of a bump.” That stopped operations until 1:24 am on the 27th. Later that morning, at 8:15, another cable broke and there was a hold at 15,008 feet until 5 pm; there was another hold at 7:52 pm with another pipe-handling problem. The deputy mission director for recovery reported, “The heavy lift system is operating marginally.”
Meanwhile, at the end of the pipe-string, the capture vehicle was being maneuvered in an effort to locate the target object—the K-129. The capture vehicle—mounted sonar detected the submarine and, at 7:52 pm on the 28th, the Lift Log Book recorded, “ARRIVED OVER T.O. [Target Object].” As the closed-circuit television cameras revealed the target object, engineer Feldman was in the control van:
It was very impressive how clear it was . . . we could see all sorts of detail.
We could see crabs crawling about . . . we could see rattail fish.
We could see much detail of the sub; in fact we were able to immediately pick out the towing eye at the bow, which was one of our aligning points and also the crack at the aft end [another alignment point].
We could see the missile tubes . . . what was left of them; they were pretty well crushed up.
Someone said they could see something down in one of the tubes . . . perhaps a warhead.
The capture vehicle was rotated by use of the hydraulic thrusters to align with the submarine. Pipe running was continued. At 1:02 pm on July 29 the capture vehicle was 217 feet from the seafloor. At 11 pm it was 90 feet from the seafloor.
On July 30 the decision was made on board the Glomar Explorer to move the capture vehicle about 800 feet “down hill,” away from the target, to permit a full, 60-foot “double” to be added to the pipe-string. This was done to allow sufficient pipe length in the ship’s pipe-string to attach hydraulic pumps. The capture vehicle’s movement began at 11:24 pm and was completed at 1 am on the 31st, when the last pipe was added to the pipe-string. Two hours later the capture vehicle was maneuvered back over the submarine. At 4:36 am the Lift Log Book noted, “Back Over T.O.”
Carefully the recovery team rotated the capture vehicle to align it with the K-129’s remains. The capture vehicle had been designed to align with three points on the submarine—a towering eye on the crumpled bow, a stanchion just forward of the sail structure, and a fracture in the hull at the after end of the forward section of the submarine. The capture vehicle was carefully aligned with the K-129. Although problems persisted, at 9:13 am on July 31 the log book recorded, “TOUCHDOWN.” With the exact alignment of the capture vehicle over the submarine, at 9:33 am the controllers began to dig the eight beams and davit tips into the soil beneath the submarine with the assistance of water jets attached to the davit arms. Numbers 2, 4, 5, 6, and 8 were on the port side, and numbers 1, 3, and 7, plus the missile containment system, were on the starboard side. The unbalanced arrangement was necessitated by the submarine lying on her starboard side, with the “missing” starboard grabbers leaving space for the sail structure. The beams and davit tips were pushed into the ocean floor with two million pounds of force. This was accomplished by offloading weight onto the pipe-string as the heavy lift system continued to lower pipe. That released the “stretch” in the pipe-string to increase the weight on the capture vehicle.
It was immediately obvious that the soil was harder than anticipated. Another one million pounds of weight was offloaded onto the pipe-string to drive the grabbers deeper into the soil. The decision to offload the additional weight to drive the capture vehicle’s “fingers” deeper into the soil was not supported or agreed to by the capture vehicle’s support contractor on board nor by many of the senior people. According to one observer, “I think we all became very nervous and very concerned.”
The beams were closed in sequence, followed by the davits, capturing the submarine in their grasp. At 4:18 pm the missile containment net was partially deployed, to contain the hoped-for missile in the No. 1 tube. It would be fully deployed immediately after breakout. At 11:48 pm the four breakout legs were extended to jack the capture vehicle and submarine off the bottom, obviating the need to stress the pipe-string, possibly to the breaking point.
On August 1, at 25 minutes after midnight, the breakout occurred. At that point the submarine’s remains shifted in the capture vehicle’s grasp, possibly breaking one of the capture vehicle’s beams.
The extending of the legs continued, and at 2:48 am the K-129 was completely free of the bottom—six years and four months after she had struck the ocean floor. At 5:15 pm the Lift Log Book recorded, “Legs OFF—ON the way UP.” The breakout legs had been jettisoned because they were no longer needed. Later it was learned that, about this time, beam No. 4, on the port side opposite the front edge of the sail, had cracked and completely broken off. But there was no discernable strain on the lift system.
Slowly, surely, the pipe-string was recovered, raising the capture vehicle and the submarine toward the surface. Problems persisted, but the pipe kept coming up. At 3:30 am on August 3 the Glomar Explorer had recovered 53 “doubles,” raising the submarine more than 3,000 feet above the seafloor. Also on the 3rd, in accordance with the Azorian cover plan, a radio message went out over an open commercial channel saying that the ship had a problem with the “nodule collector vehicle” and requested permission from the Navy to enter Midway Atoll to perform repairs to the vehicle. This would permit technicians—the B mission crew—to be flown out from the United States and board the ship.
At 6:25 am on August 4 the Lift Log Book declared, “Pipe going like a son-of-a-bitch.” But at 6:30 there was a shut down for 1 ½ hours to repair a hydraulic leak.
At 6:53 am the capture vehicle suddenly took a nose-down attitude as grabber No. 6 lost all pressure. The sail of the K-129 began to rotate. “We were in the galley, a bunch of us having an impromptu meeting [with] a cup of coffee and the ship just shook slightly like a small earthquake,” recalled Sherman Wetmore. “That wasn’t right—something had gone wrong somewhere on the ship.”
The Glomar Explorer’s massive heave compensator had reached its limit, indicating that a tremendous amount of weight had been lost. Could the pipe-string have broken? Could the K-129 have broken loose? The Lift Log Book recorded, “Heavy Lift shut down . . . CV took Nose down attitude to -2.7° trim . . . Davit [grabber] 6 came to 0 PSI [pounds-per-square-inch pressure]. . . . Possibility is the T.O. broke up or rolled over?”
The capture vehicle at that moment was 6,720 feet above the ocean floor.
There was confusion in the control vans. As engineers checked the gauges and controls, it was obvious that the pipe-string was all right and that the capture vehicle was still attached to it. The capture vehicle control van was advised that the pipe-string had lost a considerable amount of weight. Inside the van the indications were that all 12 closed-circuit television cameras on the capture vehicle were working and everything appeared normal.
About 20 minutes later the capture vehicle control van was advised that the pipe-string had lost a considerable amount of weight. Inside the van the indications were that all 12 closed-circuit television cameras on the capture vehicle were working and everything appeared normal.
About 20 minutes later the capture vehicle control center had figured out the problem. The television system used multiplex video signals whereby images from the 12 cameras on the capture vehicle were transmitted sequentially up to the Glomar Explorer. Ironically, the real-time transmission of the video was disabled in order to save film tape during what appeared to be a quiescent period. The system engineers were thus looking at images of something that was not there. When the video system was switched back to “real time” the cameras revealed that most of the K-129’s hull had fallen away.
Only the foremost 38 feet of the submarine remained within the grasp of the capture vehicle. The remainder of the target object—almost 100 feet—had fallen back to the ocean floor, with its one remaining missile, fire control system, and cryptologic material.
The decision had been made earlier that if the tug was in the area when the submarine was raised, the pipe-string would hold at a depth of about 500 feet during the daylight hours—too deep for divers on board the tug to investigate the operation. Slowly the capture vehicle and its remaining prize came up toward the open moon pool. As the claw reached a depth of a couple of hundred feet beneath the ship, U.S. divers were sent down to survey the situation. They had to make certain that there were no pieces of wreckage hanging down that could impede the closing of the moon pool gates; they took underwater cutting equipment with them on their dive, just in case. There were such obstructions.
On August 6 the SB-10 again maneuvered close aboard the salvage ship, coming to within 75 yards as the larger ship signaled the tug to stand clear. Meanwhile, the capture vehicle was raised slowly and carefully, although problems continued to occur with the lift system. At 5:11 pm the last pipe was retracted and the capture vehicle docking procedure began. At 9:17 pm the Lift Log Book recorded, “CLEMIE IS HOME! CV IS DOCKED WITH A PIECE OF T.O.!”
As if on prompt, at 9:35 pm the SB-10 again approached the Glomar Explorer, again coming within 75 yards. And again the ship signaled the tug to keep clear. The Soviet ship backed off, sounded three long blasts of her whistle, and sailed off toward the horizon, heading back to her base at Petropavlovsk. By 10:30 pm she had disappeared from the Glomar Explorer’s radar screen. Thus ended just over 13 ½ days of close surveillance of Project Azorian by Soviet naval ships. The CIA report noted, “One can only conjecture the reaction and chagrin of Soviet authorities when they later realized that two Soviet Navy ships were on the scene and, in effect, witnessed the recovery operation against their lost submarine.”
For 15 hours on August 7–8 the water was pumped out of the moon pool. As the capture vehicle and then the remains of the K-129 were revealed, crewmen installed wooden shoring to prevent the wreckage from rolling over. Immediately an inspection team in protective suits with respirators checked the K-129 remains for nuclear contamination. Radiation was detected, primarily from one or both of the nuclear torpedoes carried by the submarine. The CIA believed that one or both of the torpedoes’ high-explosive components had detonated without creating a nuclear explosion. The plutonium of the warheads was in a hydroxide form and thus there was little danger of airborne particulates; it was dangerous only if inhaled or digested. Whenever acetylene-cutting torches were used on the submarine during the next few days everyone in the area initially wore breathing devices because the plutonium could vaporize if heated and could then be inhaled. But that precaution was soon abandoned. Otherwise, personnel working on the submarine in the moon pool wore only industrial hard hats and disposable white coveralls.
As the wreckage was checked for radiation, a few mission specialists stood on balconies overlooking the moon pool and viewed the crumpled wreckage resting below them—amid flotsam, puddles, and a number of manganese nodules that had come up with the wreckage. As the capture vehicle released the wreckage, of the eight grabbers they saw that No. 4 and No. 5 were broken halfway up their length, and that No. 6 had its hydraulic cylinder torn away. It became obvious that the amidships section of the K-129 broke away and fell through the gap created by the missing No. 4 and No. 5, carrying away the containment net attached to No. 7 grabber and damaging No. 6.
As for the submarine herself, she lay crumpled at the bottom of the moon pool. Without a sail of after section, the wreckage bore little resemblance to a submarine. Still, although only 38 feet had been salvaged, to many of the crew the submarine looked huge. Raymond Feldman recalled, “Looking into the opening of the piece [of the submarine] that we had it looked pretty much like a dump. There were no bulkheads that I could see. There were no fixtures that I could recall . . . , I don’t remember seeing anything attached to the hull on the interior. There seemed to be a solid pile of material of maybe two or three feet in height that was very dense and very compacted. . . . There was nothing I could recognize as the interior of a submarine.” And, “Immediately exploitation of the sub was started . . . people enthusiastically going in and digging like they were at an archaeological site; it’s the closest I can describe it.”
While the outer hull of the submarine was torn up, the inner or pressure hull appeared relatively intact.
As mission specialists sifted through the debris that had been the bow section of the K-129, appalling sights—and smells—awaited them. Several mutilated bodies were found, one recovered in what remained of his bunk; he had apparently been studying a manual related to the submarine’s torpedoes when—in an instant—death came to all 98 men in the submarine. Details of the interior condition of the submarine are vague, but one other corpse was found that was relatively intact. The incomplete remains of other bodies were found in the debris, and smaller body parts had spilled out of the submarine and were found on the floor of the moon pool. At her resting depth the bodies within the submarine had been preserved because the near-freezing water and intense pressure had inhibited decay. Now, with the submarine on the surface and exposed to air, the bodies immediately started to deteriorate.
A “production line” of tables was set up in the moon pool where each item removed from the K-129 was meticulously cleaned, photographed, entered into a log, and placed in preliminary packages. Subsequently, items that were identified as having no intelligence value were broken into small pieces and prepared for disposal. One of the items removed from the submarine—that would later become significant in U.S.-Soviet relations—was the submarine’s bell. It had apparently been removed from the sail structure and stowed in the bow compartment when the K-129 went to sea.
Papers, documents, and the few manuals that were found were rushed to vans that had special paper preservation facilities and materials. Letters and photographs were also found.
The Glomar Explorer reached her berth at Long Beach on September 21, 1974. At Long Beach the specialized mission vans were removed and the crates containing material from the K-129 that would be the subject of further examination and assessment were brought ashore and shipped to an undisclosed location. There are indications that the location was the fabled Area 51, some 80 miles northwest of Las Vegas, Nevada. The location is cited in numerous accounts of U.S. government activities related to Unidentified Flying Objects (UFO). More to the point, Area 51 was the scene of testing for several top-secret projects. By 1974 it contained a variety of technical facilities that could certainly perform examinations of the K-129’s remains. The final resting place of the submarine’s remains that were deemed significant enough to retain were packed up and—according to the unofficial but creditable statements made to the authors of this volume—are stored in a nondescript building, known simply as Building 7717, set back from Trigger Avenue, on the naval submarine base at Bangor, Washington.
Even as the Hughes Glomar Explorer sailed toward Long Beach, officials at CIA headquarters in Langley, Virginia, and at Global Marine headquarters in Los Angeles evaluated the feasibility of returning to the site of the K-129’s remains to salvage the missile section and its hoped-for R-21/SS-N-5 missile and cryptologic material.
The CIA persisted and Project Azorian officially ended on October 20, 1974; the following day, Project Matador was initiated. Meanwhile, Global Marine, Lockheed, Honeywell, and other contractors involved in the Glomar Explorer and her lift systems began to rehabilitate the ship and to rebuild the capture vehicle. Not only would the damaged systems be repaired, but the grabbers and certain other components had to match the new configuration of the wreckage, that information being derived from the seafloor photographs taken by the Seawolf’s towed fish.
An unusual event occurred in December 1974 at the U.S. Navy’s holiday party for foreign naval attachés, held in Bethesda, Maryland. At the part the Soviet naval attaché approached the U.S. Navy captain who served as liaison officer for foreign naval attachés. The Soviet officer, reportedly, said that it was known that the United States had tried to raise the submarine K-129 and, “if you go back there it would mean war.”
Was the Soviet officer making an “official” statement? Had he had too much to drink at the party? And, did his statement reveal that the Soviets knew that the entire submarine had not been raised? Or that another salvage effort was being prepared? The conversation was reported to Admiral Inman the next morning.
Project Matador was short-lived.
Yet, the Americans failed. Only some 38 feet of the submarine were recovered. While that section did have intelligence value, the principal targets of the effort—an R-21/SS-N-5 ballistic missile and cryptologic material—and press reports continue to ascribe the recovery of the entire submarine or, at least, the missile and the crypto material.)
The recovery of two crushed nuclear torpedoes, a few documents, and some basic knowledge of Soviet submarine construction—for a dated, non-nuclear design—was a pretty thin “take” compared to the envisioned haul. Lee Mathers, a former U.S. Navy intelligence officer, wrote, “Azorian was a great gamble, displaying the actions of a confident country with the wealth and the will to make such a gamble if the potential gain would make the effort worthwhile.”
If successful, the effort would have potentially provided a long-term look into Soviet cryptologic systems and established a solid datapoint in our understanding of the Soviet technological development curve concerning both missiles and nuclear weapons. However such a gain—especially in the area of cryptographics—would be fleeting unless the [secret recovery] mission could be kept completely secure for a long, indeed an indefinite, time.
The total cost of Project Azorian was on the order of $500 million in early-1970s money, including about $200 million for constructing the Glomar Explorer. The cost of a manned mission to the moon at the time was on the order of $500 million. The CIA has carefully withheld all cost data on the project, even to the date of this book’s publication. Could such data be useful—several decades after the event—to a potential enemy? Or even to the Congress?
The cost was significant. Was the effort “worth it”? Here is Admiral Dygalo’s evaluation of the project:
I said it many times: one must judge fairly his adversary. It goes without saying that the USA made a breakthrough in technology and had [a] successful economy—both became exclusively thanks to highly qualified human resources. Just the fact that they managed to build a ship of 60,000 tons of displacement, to install equipment to sustain such a [submarine] load, to make provision of how to accommodate the submarine under the ship and finally lift it up. It seemed to us something unreal, fantastic, I can compare it with a mission to the moon in regard of technology and invested money. And another point—the ship was built in two years [sic] and we were fooled these two years—the disinformation was organized outstandingly.
Thus did the “enemy” judge Project Azorian. That judgment was shared by many Americans, and was “made official” in July 2006 when the American Society of Mechanical Engineers awarded its prestigious Historic Mechanical Engineering Landmark award to the Hughes Glomar Explorer. The citation noted,
The success of the Hughes Glomar Explorer proves that the impossible is, indeed, possible when talented engineers with the courage to take prudent risks are provided an incentive to stretch the state-of-the-art.
The Hughes Glomar Explorer and Project Azorian—history’s most ambitious ocean engineering effort—thus took its place among such remarkable American engineering achievements as the Wright Flyer III, the Hughes Flying Boat HK-1, the hydrodynamic research submarine Albacore (AGSS 569), the Saturn V rocket, and the Apollo command module. Project Azorian was certainly a remarkable and audacious engineering achievement—even if not fully successful.