Imagining Soviet Surveying

Last week I wrote about some of the apparent differences between how the US and the Soviet Union used satellites for mapping and geodesy. The Soviets seem to have been slower to operate dedicated satellites in both areas, with no apparent explanation. Though it’s dubious to use US intelligence estimates as evidence of what the Soviets were actually doing, they do at least shed light on some of the possibilities.

Two CIA reports from from the pre-satellite era, in 1954 and 1957, suggested that if the Soviets had made a connection across the Bering Strait between their own domestic surveys and the North American Datum, missiles launched from near the Bering Strait would have a Circular Error Probable (CEP) of 300–500 feet. Without the connection between datum, the error would be closer to 1,000 feet. without any additional surveys of US territory By making observations of an upcoming solar eclipse and gaining access to the equivalent measurements from US or Western European sites, the CIA predicted the error in intercontinental position could be reduced to about 500 feet from anywhere in the Soviet Union.

These estimates assumed that the target could be located on high-quality American maps, which the analysts presumed were available to Soviet planners. But what if the targets were secret sites not plotted on any maps? A Studies in Intelligence article (“Spy Mission to Montana”) from 1995 revealed that the CIA and Air Force tested those conditions as the silos for Minutemen ICBMs were being built in 1962. A three-person team, two from the CIA and one from the Army Map Service, made covert observations of the sites under construction from their rental car. Dodging both site security and the official survey being done by the Air Force’s 1381st Geodetic Survey Squadron, the covert team proved that observations could be made with a CEP of 600 feet when maps at 1:250,000 scale were available and a CEP of 200 feet with 1:62,500 scale maps.

Did the Soviet Union make a secret measurement of the Bering Strait or send its agents to survey the locations on American missile silos? The answer is probably somewhere in the files of the KGB or GRU.

An Atomic Rolodex

Los-Alamos-Rolodex_CoverOne of Steven Heller’s Daily Heller newsletters this week featured an interview about a new book from Blast Books and the Center for Land Use Interpretation. Los Alamos Rolodex prints a selection of business cards from companies that served the Los Alamos nuclear lab in the 1960s and 70s. As publisher Laura Lindgren explains, “everybody wanted to do business with the nuclear industry—from cryogenic corporations to cleaning supply companies … The shift in logo and typographic designs from the mid-’60s to the late ’70s tracks along with technology changes—often in the mid-’60s logos are sparked with atomic power; in the ’70s cool digital data technology images and typography provide the excitement, as in the mainframe computer tapes in the 1970 UDAC card logo and the 1978 Interactive Computers card.” The book looks like a really interesting slice of both design and national security history to complement atomic postcards and nuclear test participation certificates.

Cry Havoc and Let Slip the Sleds of War

Transportation_and_RevoltGeographer Jacob Shell’s Transportation and Revolt is a treasure trove of stories of animals in war. Having begun with culls of Belgian courier pigeons by anxious German occupiers during the world wars, Shell moves in quick succession to describe the mobilization of American mules to fight the 1945–49 Greek Civil War, elephant-riding Kachin rebels in 1980s Burma, and camel-borne troops in Algeria, Sudan, and the American West. This use of beasts of burden in war, revolt, and smuggling, Shell argues, are all examples of “subversive” transportation networks that threatened ruling political regimes – a category which also includes Britain’s nineteenth-century canal workers and the New York waterfront’s cargo handlers (or shenanoges).

Shell’s examples of subservice mobility are interesting, as is his general concept, though I’m not convinced when he tries to show that it was a subversive connection with social disorder that led to official disinterest or opposition. (The exception is the most local and most detailed case, that of New York cargo handling.)

My favourite example from the book, though, has to be the mobilization of the dog sled, starting with the sled-borne warriors of Siberia. Between 1697 and the early nineteenth century, the Russian Empire fought a long series of inconclusive campaigns against Siberian peoples like the coastal Chukchi who used dogsleds both for strategic mobility and in battle. Using their sleds the Chukchi could move entire villages out of the path of Russian expeditions, hunting seal on the ice and avoiding the enemy entirely. In the case of the 1697 invasion of Kamchatka, Siberian peoples including the Yukagir, Koryak, and Itelmen used their sleds like ancient chariots: one man driving and the other wielding a bow and arrow.

Nor was the eighteenth century the end of the dog sled’s military utility. Towards the end of the book, Shell points out that sleds were part of the transport of uranium to make the first atomic bombs. In the first stage of moving uranium from the mines near Great Bear Lake in the Canadian North, “dog sled teams, driven by Dené mushers, moved sacks of the pitchblende ore from the seams to Port Radium. Next, wooden barges carried the material down the Great Bear River to Fort Norman, then up the Mackenzie River to Waterways, Alberta. The material then went by rail to the Eldorado refinery plant at Port Hope, on the north shore of Lake Ontario.” That made the humble dog sled one of the first links in a chain whose last connection was the B-29 bomber Enola Gay, one most advanced warplanes of the Second World War.

TERCOM, System and Symbol: Part Three

From Part Two

Direct Action’s bombing of the Litton factory in Rexdale and its assembly line for cruise missile guidance systems reflected discomfort on the Canadian left with the way the Cold War was heating up again after the years of détente and what they saw as an insufficient willingness to put distance between Canada and US foreign policy. Neither it, nor the anti-cruise missile protests, put much specific emphasis on the technology involved. For them, TERCOM itself was just one particular articulation of the US military-industrial complex.

In fact, TERCOM in general was a dead end in military guidance. No other weapons used the same guidance technique, and both the Tomahawk and conventional ALCM used GPS as their primary guidance as soon as practical. I have yet to see any reference to a commercial spin-off from TERCOM either. The idea was more or less a one-off as a guidance technique. It survived into the twenty-first century in only one niche, the nuclear-armed Tomahawk. Because, unlike GPS or other radionavigation systems, there were no outside signals to be jammed or spoofed, TERCOM-assisted inertial navigation remained the sole guidance system on the nuclear-armed Tomahawks even after the conventional versions switched over using to GPS. The last nuclear-armed Tomahawks were only retired in 2013.

On the other hand, TERCOM did demonstrate the value and cost of good mapping, charting and geodetic data. It wasn’t the first system to make use of it – every US strategic bomber and ballistic missile relied on mapping and geodetic information to some extent – but it was the first to demand not just knowledge about Point A and Point B but also about the terrain along the way. That requirement put the Defense Mapping Agency in a bind, forcing it both to go into overdrive and to triage its TERCOM processing work. Recognising that that sort of crash project couldn’t be repeated for every new weapon, the deputy Secretary of Defense issued Program Decision Memorandum 85 (PDM-85) in 1985, which required early military department to “fund with its own resources the cost of unique earth data products.” Though Larson and Pelletiere wrote in the late 1980s that the rule was proving unenforceable, it was a mark of further recognition that this type of information was a critical war weapon.

After the Cold War ended, the Defense Mapping Agency was merged with many of the intelligence community’s imagery creation and analysis office to create the National Imagery and Mapping Agency (NIMA). In 2003, NIMA was renamed the National Geospatial-Intelligence Agency (NGA), a change that reflected the increasing conceptual consolidation of these kinds of information under the umbrella of geospatial intelligence (GEOINT). The term, as the US Geospatial Intelligence Foundation explains, was only about as old as the agency’s new name. But while The Atlantic’s Marc Ambinder could title as story about the agency in 2011 “The Little-Known Agency That Helped Kill Bin Laden,” NGA was pretty deeply embedded in the US national security establishment. More than thirty years after DMA started weaponizing its digital terrain elevation data (DTED), the idea that the military might not only demand detailed maps of its targets but also the underlying data, to transform into a three-dimensional computer model, a physical mockup, or – bringing us right back to the first uses of the DTED – a flight simulator profile (which was, after all, the first use for, back in the 1970s), is old news.

Source Notes: US cruise missiles are pretty widely discussed, so a lot of these posts were cobbled together from a lot of sources. Jay L. Larson and George Pelletiere’s Earth Data and New Weapons (available from DTIC here) was very useful for understanding how the DMA supported TERCOM, and is one of the few places to mention PDM-85. The explanation of how satellite stereophotogrammetry is done comes mostly from the NRO’s internal history Hexagon Mapping Camera Program and Evolution (as reprinted by the Center for the Study of National Reconnaissance). Information about Canadian protests against cruise missile testing and the Litton bombing in Toronto come from John Clearwater’s 2006 book Just Dummies: Cruise Missile Testing in Canada. Ann Hansen, one of Direct Action’s members, published a memoir after he release from prison. Direct Action: Memoirs of an Urban Guerilla offers more but similar details about the Litton bombing and reprints Direct Action’s communique.

TERCOM, System and Symbol: Part Two

Back to Part One

GPS was only one among many guidance technologies that saw their first use in the first Gulf War. One of the others was Terrain Contour Matching (TERCOM), the guidance system behind a trio of American cruise missiles: the sea-launched Tomahawk, Air-Launched Cruise Missile, and ground-launched Gryphon. As Ingrid Burrington reported for the Atlantic earlier this year, GPS provoked some strong feelings that included an axe-wielding attack by peace protesters. TERCOM attracted some of the same attention.

None of the three cruise missile systems were uncontroversial, though the Gryphon attracted the lion’s share of the protests in the European countries where it was to be based (just google “Euromissiles crisis” or “Greenham Common” for a sample). In Canada, US flying tests of the ALCM over northern Alberta as a proxy for the terrain of Siberia led to substantial public protests. Some went beyond vigils, marches, and speeches. Two Greenpeace members climbed the Peace Tower on Parliament Hill in Ottawa to unfurl a banner saying “No cruise, Greenpeace.” An art student, Peter Grayson, threw red ink on a copy of the Canadian constitution on display at the National Archives. In Prince George, BC, an ALCM was burned in effigy, near Wandering River, Alberta, Greenpeace launched a net carried by balloons during one test as a “cruise catcher.”

Canadian involvement with cruise missile development was also the spark for an act of violent protest reminiscent of the Kjoller-Lumsdaine “Harriet Tubman-Sarah Connor Brigade” attack on GPS.

Direct Action in Rexdale
One of the companies building guidance packages for cruise missiles was Litton Systems Canada, the local subsidiary of Litton Industries. On the night of October 14, 1982, the Etobicoke Police Department received a telephone call reporting that there was a blue van filled with explosives parked outside the Litton Systems factory in Rexdale, Toronto. The van, which was filled with 250 kg of dynamite stolen from the British Columbia Highways Ministry earlier that year, was the work of Direct Action, a radical British Columbia group who had already blown up four power transformers on Vancouver Island to protest BC Hydro’s development there. Three of what became known as the “Squamish Five” after their arrest – Brent Taylor, Juliet Belmas, and Ann Brit Hansen – drove out to Toronto to carry out the bombing.

Their target, Litton Industries’ Plant no. 402, had already been the site of protests against Litton’s cruise missile production contract. Direct Action took pains to avoid any casualties from their bombing. They placed an orange box with two sticks of dynamite and a warning on the van’s hood, then phoned in a warning message to the police. Despite their precautions, the bomb in the van detonated early, putting nine people in hospital. In their post-bombing communiqué, the group apologized for the injuries before going on to condemn Canadian complicity in the nuclear arms race.

Direct Action’s members soon moved on to other activities, firebombing a series of stories in the Red Hot Video pornography chain and planning the robbery of a Brinks armored car, before they were captured by the RCMP in early 1983. Their attack on the Litton factory had effectively no effect on the work done there. John Clearwater, whose book on the subject documents the eighteen flight tests and the related protests in painstaking detail, reports:

the explosion did not stop cruise missile navigation systems production. In fact, the area in the factory, well insulated from shock and vibration, remained unharmed. The test pedestals on which the delicate instruments were calibrated remained mounted on their steel and concrete columns sunk far into the soil beneath the factory. After the blast, technicians simply rechecked the alignment of the test-beds with the stars and the earth’s axis to ensure the perfection of the TERCOM navigation instruments. Production was not halted. (Just Dummies, page 102.)

Hansen’s own memoir reported that the factory’s work was disrupted for only two days.

Direct Action’s communiqué taking responsibility for the Litton Systems bombing rooted their opposition to the cruise missile in opposition to nuclear war, as “the ultimate expression of the negative characteristics of Western Civilization,” whose “roots lie deep in centuries of patriarchy, racism, imperialism, class domination, and all others forms of violence and oppression that have scarred human history.” They condemned the cruise missile as one of set of weapons “designed for offensive first-strike use” (including the Pershing II, Trident, and neutron bomb), as well as Canadian involvement in US nuclear weapons production: not just Litton’s work on the cruise missile, but also launchers for Lance missiles by Hawker-Siddeley, hull cylinder torpedo tubes for Polaris, Poseidon, and Trident by Vickers, cranes to load Trident missiles by Heeds Inernational, and one component of the MX missile. In the context of Direct Action’s critique, neither inertial navigation nor terrain contour matching was of much specific interest.

Forward to Part Three

TERCOM, System and Symbol: Part One

On the opening night of the first Gulf War in 1991, seven B-52 bombers flew 14,000 miles from Barksdale Air Force base in Louisiana to launch the first attack in history with GPS-guided weapons. The bombers, whose operation was formally code-named SENIOR SURPRISE but quickly nicknamed “Secret Squirrel,” launched thirty-five Conventional Air-Launched Cruise Missiles (CALCM), each equipped with a single-channel GPS receiver. That made the CALCM was the first and only GPS-guided weapon fired during Operation Desert Storm, though GPS navigation was used by many attack aircraft.

SENIOR SURPRISE was, it turns out, the first step towards a era of cheap and easy GPS-guided weapons. But during Operation Desert Storm those thirty-five missiles fired were vastly outnumbered by their close cousin, the US Navy’s BGM-109 Tomahawk cruise missile. The two missiles shared a common engine, the Williams F107 turbofan, and they had been designed with a common guidance system in mind too. The CALCM’s were a variant of the US Air Force’s nuclear cruise missile: the nuclear version was designated AGM-86A, and it and the BGM-109 shared a guidance system built around Terrain Contour Matching, or TERCOM.

US Navy photo of a Block IV "Tactical Tomahawk" in flight. 021110-N-0000X-003 China Lake, Calif. Courtesy Wikipedia.

US Navy photo of a Block IV “Tactical Tomahawk” in flight. 021110-N-0000X-003 China Lake, Calif. Courtesy Wikipedia.

Today, we think of the first Gulf War as GPS’s coming-out party – which it was – but imagine it as the backbone of the many new weapons the US unleashed during the war – which it wasn’t. That begs the question, what actually was guiding US weapons during the first Gulf War, and why those systems never captured the imagination – for better or worse. After all, in 1992 two activists took axes to unlaunched GPS satellites because GPS was “military in its origins, military in its goals, [and] military in its development.”

TERCOM was the guidance system at the heart of the US cruise missile triad: an air-launched missile that extended the range and capabilities of the US bomber fleet (the ALCM); a submarine and ship-launched missile that gave US Navy ships a strategic nuclear weapon (the Tomahawk); and a ground-launched missile to be based in Europe as a counterpart to the Soviet intermediate-range nuclear arsenal (the Gryphon). Of the three, Tomahawk would have the most significant career. While the ground-launched Gryphon was fairly swiftly traded away in exchange for the Soviets demobilizing their intermediate-range missiles through the INF Treaty, the Tomahawk became the American military’s go-to weapon for punitive strikes during the 1990s. Tomahawks were fired against Iraqi military targets in 1993 and at Afghan and Sudanese targets in response to the East African embassy bombings in 1998, as well as used in conjunction with airstrikes in Bosnia (1995), Iraq (1998), and Serbia (1999). They remain, as journalist Mark Thompson put it in TIME magazine last year, “the curtain-raiser on U.S. military strikes since 1991’s Gulf War.”

(Since the Block III version was introduced in 1993, Tomahawk missiles without nuclear warheads have had GPS guidance in addition to the TERCOM system. The nuclear-armed version remains TERCOM only.)

How TERCOM Worked

Tomahawk flight diagram. From GAO/NSIAD-95-116 (1995), p. 17

Tomahawk flight diagram. From GAO/NSIAD-95-116 (1995), p. 17

Unlike GPS or other radionavigation systems, TERCOM did not provide continuous position information. Instead, it was an adjunct to the cruise missile’s inertial navigation system (INS), which used gyroscopes and accelerometers to measure the weapon’s movement. What TERCOM did was provide updates to the INS by comparing the measurements from the missile’s radar altimeter with stored information on the terrain the missile was flying over. Assuming this was a sufficiently “bumpy” area, the altimeter measurements should match only location – providing a precise position fix. By using TERCOM to update its location three times during flight, the cruise missile had far superior accuracy to a missile using only inertial navigation. (On non-nuclear missiles, a separate system known as Digital Scene Matching Area Correlation [DSMAC] offered even more precise guidance for the final stage of the flight.)

Making Maps for TERCOM

The terrain information which the TERCOM system used (called a “matrix”) was based on a database of Digital Terrain Elevation Data (DTED), which contained the elevation above sea level for all the points on a (roughly) 300 foot grid. Luckily enough, the Defense Mapping Agency (DMA) had already been collecting this information for a far more mundane purpose: to use in flight simulators. TERCOM, however, required much, much more of this information. The DMA had been involved in the early testing of TERCOM but the agency wasn’t involved in the decision to put the cruise missiles using the technology into full-scale development. DMA had only been creating TERCOM matrices by hand (so to speak), and now it was looking at the requirement for more than 8,000 of them. According to Jay L. Larson and George A. Pelletiere, “it took DMA five years of ‘crash’ effort to fulfill these requirements.” Demands for TERCOM matrices had to be triaged, with three-quarters of the effort going to the ALCM and the other quarter to the submarine-launched and ground-launched cruise missile programs. If the US military had gone to war in the interim, they would have been unable to use cruise missiles except in particular areas.

To extend the Digital Terrain Elevation Data database over the entire world, and particularly over areas like the interior of the Soviet Union, the DMA used photography collected by the United States’ reconnaissance satellites. The technical principle involved, stereogrammetry, had been well understood since the First World War. Start with two photographs of the same scene taken from slightly different vantage points, such as two moments along the flight path of a plane or satellite. Look at the two images, one with each eye, and move them until they like a single three-dimension image – just like how the human brain normally combines the images from each eye … if your eyes were several hundred meters apart.

If you know the distance between the points from which the images were taken (that hundred metre bridge of the nose) and the angles between the two viewpoints and an object in the image, you can calculate the size of that object. Add in the altitude from which the images were taken and you can calculate the elevation. Conventionally, stereogrammetry is done using a human brain to process illuminated transparencies or with mirrors and hardcopy prints. In the early 1960s computerized system like the Universal Automated Map Compilation Equipment (UNAMACE) took over, speeding up the process tremendously.

UNAMACE. From Hexagon (KH-9) Mapping Camera Program and Equipment (CSNR, 2012), p. 301.

Pictures of UNAMACE, from Hexagon (KH-9) Mapping Camera Program and Equipment (CSNR, 2012), p. 301.

Even once TERCOM matrices were readily available, the mission planning process remained long and complex. During the first Gulf War, it had to be performed at a specially equipped facility (one of the Navy’s two Cruise Missile Support Activities) and took between 24 and 80 (!) hours.

The US cruise missile program attracted a lot of controversy in the 1970s and 1980s, particularly from critics who argued that a new, flexible, and less detectable nuclear weapons delivery system could be destabilizing or even provocative, but TERCOM itself did not itself draw that much of the attention. The exception was up in Canada, where a guidance system subcontractor found itself the focus of a blast of “direct action” against the cruise missile.

Forward to Part Two

London’s Secret Tubes and Secure Communications Beneath the Streets

I had some time to kill at the Toronto Reference Library last week, and I was lucky enough to browse my way along to London’s Secret Tubes by Andrew Emmerson and Tony Beard. The book doesn’t circulate, so I wasn’t able to finish reading it, but it is chock-full of remarkable information about the military and civilian services that operated underground during World War Two and the Cold War. As well as covering the various headquarters that were built to resist aerial bombardment (like the Cabinet War Rooms, Admiralty Citadel, and the alternate Cabinet war room code-named PADDOCK) and the deep air raid shelters, London’s Secret Tubes discusses a lot of the communications infrastructure that was buried for protection. The British telephone system’s main exchange was in Central London, so the Post Office (who operated it) not only built an alternate connection in the suburbs but also add a ring of alternate connections that skirted the city so that calls could be rerouted around the edge if connections to the hub were broken. After the war the central interchange itself went underground, taking over the deep shelter beneath Chancery Lane as the Kingsway exchange (on which more here).

Most of a SIGSALY installation (at the National Cryptologic Museum

Most of a SIGSALY installation (at the National Cryptologic Museum)

London’s Secret Tubes isn’t a book about communications networks per se, but discussing the hardening to them makes it an inadvertent guide to just how many systems were operating in and through London. In addition to the telephone system, the undersea cables headed overseas – for telegraphy at this stage, not telephones – had their British final connections in central London. These too had to have backup connections in the suburbs, often in converted private homes. The British end of the encrypted voice radio link between Churchill and FDR was also in London, installed in the basement beneath Selfridges department store – which Emmerson and Beard report was only one of several department stores taken over by the government during the war.

Though incredibly important, these networks were less of a “system of systems” and more of an accretion of communications systems that gravitated towards the centers of economic and political power. And, from time to time, they hit their limits. My favourite example is one that historian Peter Hennessy shared in a 2009 lecture (paywalled here) and incorporated into the revised edition of his book Secret State. In 1962, when the British government was tracing out the communications links the prime minister would use in case of an imminent nuclear attack, the connection to be used if the prime minister was in a car was a radio broadcast from the system the Automobile Association (the UK equivalent of AAA) used to contact its service patrols. The prime minister would then make for the nearest phone booth to call Whitehall. Should the driver not have change for the phone call (an issue which really did come up in planning), the prime minister would simply reverse the charges and make decisions of national security via a collect call. The mind boggles.

(I’m not going to reproduce the communications diagram that includes the AA radio link, since it’s Crown copyright and would require permission [and a payment to get a clean copy], but you can see it at a good size on The National Archives website here. The AA link is in the top right corner, complete with a little drawing of a car. The file itself is DEFE 25/49 (1),”Nuclear retaliation procedures.”)