Blank Spots on the Map Read online

  As far as spy satellites are concerned, these are only the beginning. The National Reconnaissance Office’s still-classified files overflow with the code names and security compartments associated with secret spacecraft. There are the TRUMPET/ADVANCED JUMPSEAT satellites, designed to intercept signals from radars, surface-to-air missiles, and other electronic devices. The MERCURY/ ADVANCED VORTEX constellation of near-geosynchronous satellites conduct eavesdropping operations, while the MAGNUM/ADVANCED ORION and MENTOR spacecraft pluck other wayward signals emanating from the ground. The QUASAR constellation acts as a data relay system between ground controllers and other satellites. And then there are the MISTY/ZIRCONIC spacecraft, a class of stealth satellites that have proven just as adept at hiding from amateur satellite observers as they have from congressional oversight committees. In sum, the “other night sky” is a mind-boggling collection of colorful and highly classified spacecraft.

  The other night sky extends down to the earth. For every satellite in Earth orbit, there are giant corporations like Lockheed, Boeing, and Northrop with contracts to research, develop, and deploy the spacecraft. There are legions of soldiers and contractors whose job it is to operate the spacecraft from a global network of control facilities and downlinks, from the Pine Gap facility in Australia’s outback to Onizuka Air Force Station in the heart of California’s Silicon Valley, and from Thule Air Base, near the North Pole, to Diego Garcia, the restricted island in the Indian Ocean. The sole purpose of agencies like the National Reconnaissance Office and U.S. Space Command, not to mention the satellite arms of the Navy and the National Security Agency, is to coordinate, organize, and oversee all of these spacecraft. Although most classified satellites are invisible to the unaided eye, their earthly footprint is tremendous.

  More than fifty years of black spacecraft design has also meant creating a vast bureaucratic, cultural, and social architecture of secrecy and deception. Building secret satellites means creating enormous black budgets, hidden factories and obscure contracts to task their development, ultraclassified security compartments to protect the “product,” and a history of disinformation and outright lying to protect their secrets. “Overhead assets,” as the National Reconnaissance Office and Pentagon call them, are not only technological marvels but social and cultural things.

  The first problem with identifying black satellites in the night sky is learning how to tell the difference between something interesting, such as an ONYX spacecraft, from something more quotidian, like a DirecTV satellite. The place to start is an online catalog called space-track.org. Produced and maintained by U.S. Strategic Command (USSTRATCOM), Space Track is a collection of data describing the motion and location of all the satellites and significant pieces of space debris that the Department of Defense keeps active tabs on. Classified satellites, however, do not appear in the catalog.

  The Space Track catalog is a map to the unclassified night sky, the “known knowns” in Earth orbit. If something appears in the night sky and it’s not in the catalog, that’s a strong indication that the unknown object is a classified spacecraft.

  Using Space Track is a little creepy. First of all, the domain name is registered through a company called Domainsbyproxy. com, whose business is to anonymously register Internet addresses. To access the data on Space Track, a user has to create an account with the Defense Department, promise not to distribute the information anywhere else, and submit to a policy that includes the following:

  WARNING! This web site contains data and information provided by the U.S. Government. If you are not authorized to access this system, disconnect now. You should have no expectation of privacy. By continuing, you consent to your keystrokes and data content being monitored.

  A Space Track account gives access to lines of data sometimes called “Keplerian elements,” more commonly known by the acronym “TLEs,” for “two-line elements.” A TLE is a series of numbers that looks like this:

  ISS (ZARYA)

  1 25544U 98067A 04236.56031392 .00020137 00000-0 16538-3 0 5135

  2 25544 51.6335 341.7760 0007976 126.2523 325.9359

  15.70406856328903

  To the uninitiated, a TLE looks like a mysterious code. The top line (ISS) designates the catalog or common name of the orbital element. In this case, ISS stands for “international space station,” often the brightest thing in the night sky besides the moon. The first entry on both subsequent lines, 25544, is the United States Space Command’s satellite catalog number for the object—the military designation for the satellite (00001 was Sputnik). The letter “U” after the number means that the object is unclassified. The second number, 98067A, denotes the object’s international designation, also known as its COSPAR (Committee on Space Research) number. In this case, the ISS was launched in 1998 (98), it was the sixty-seventh successful launch that year (067), and it is the first object from that launch tracked, so its COSPAR number is 98067A.

  From here, things get complicated as the numbers begin describing the object’s orbital characteristics. One number describes the last observed south-north equatorial crossing of the satellite, others the inclinations, eccentricities, perigee, mean anomalies and motions for the object, and so forth.

  The Space Track resource is amazing, a testament to what Niels Bohr meant with his notion of an “open world.” The Space Track catalog, like much else in the satellite-observing hobby, is a holdover from the Moonwatch days. Back when civilian observers acted as a de facto satellite-tracking service, NASA’s Goddard Space Flight Center provided observers with known orbital elements, enabling the amateurs to check their accuracy and to identify any satellites the government had no records for. Once the military developed tracking systems far more advanced than teams of volunteer hobbyists, there was no reason per se to keep making the tracking data public, but the tradition of making the elements public remained intact. Nonetheless, the Space Track catalog is constantly under threat of becoming classified: There are plenty of people in the Air Force and NRO who don’t at all like the idea of an average civilian with a Web browser having access to the same data that they do.

  Ted Molczan uses the Space Track catalog as a backdrop to process his observations and those of other satellite hobbyists like Russell Eberst and Mike McCants (who post regular reports to a Listserv called SeeSat-L). When his colleagues observe an “unknown” object, Molczan compares the sighting to the catalog. If there are no obvious correlations, he tries to match the data to a classified object in his own catalog. When he finds one, he updates the TLE files he maintains for the black spacecraft. Using these methods, Molczan generates phenomenally accurate records.

  “In addition to accurate positional data, [some observers] estimate visual magnitude,” he explains. In other words, some observers will describe the brightness of a witnessed satellite using a scale derived from astronomy, where Jupiter is about mag -1, the stars in the Big Dipper are around mag 2, and the faintest visible star on a moonless night might be around mag 5.5. For Molczan, these observations reveal even more: “We try to develop a ‘standard magnitude’ by normalizing data as if they’re all at a thousand kilometers from the observer. Brightness follows the inverse-square law. We can take mag observations amassed over time and recompute them as if they’re at one-thousand-kilometer range. Satellites show phases just like the moon and planets.” If Molczan can determine a standard brightness from multiple observations, he can use that data to estimate the size of an object: Larger objects are intrinsically brighter than smaller ones, because they have more surface area to reflect light. Another variable is the payload capacity of the rockets used to launch spy satellites: Molczan estimates the size and mass of an object by studying the performance characteristics of rockets like the Titan IV (published for the benefit of commercial companies launching satellites).

  Using these methods, Molczan and his colleagues figured out that a new generation of KH reconnaissance satellites took to the skies in 1992: The newer reconnaissance birds were larger than the
ir forebears but were simultaneously about a magnitude dimmer. By paying close attention to the perturbations of an orbit due to atmospheric drag and solar radiation pressure, Molczan can calculate the area-to-mass ratio: He can learn how dense an object is and how much it weighs. A skilled observer can derive a surprising amount of information from nothing more than little streaks in the night sky.

  The bulk of this work happens on a Listserv for satellite observers called SeeSat-L. Although anyone can join the forums, it’s a tough place to be a newbie. Most posts contain nothing more than strings of numbers, mathematical descriptions of the previous evening’s sightings.

  When he has some spare time, a satellite observer in Texas named Mike McCants compiles SeeSat observations into a simple text file he calls “classfd.tle.” The file contains only twenty kilobytes, about the size of an e-mail, but it serves as a near-complete map to the other night sky. With classfd.tle, anyone with access to a computer can generate predictions of reconnaissance satellite overpasses for any location in the world. The data is usually accurate to within a fraction of a second.

  Import the classfd.tle file into a freeware application like Orbitron, and the information blossoms with details of almost two hundred secret orbits. At the top of the list is a leftover piece of debris from a Japanese spy satellite named Information Gathering Satellite 4A (IGS 4A). The list ends with an entry for XSS-11r, the rocket booster left over from a joint Lockheed/Air Force Research Laboratory project called the Experimental Small Satellite, a “micro-satellite” that could rendezvous with other spacecraft for “inspection” (a mission widely understood to mean that the XSS was, in fact, designed to intercept and destroy other satellites).

  “It’s a good night for observing satellites,” says Molczan as he loads up the Obsreduce software he developed to process observations on his desktop computer. He also uses the software to make predictions about where a satellite will be in the sky, then plots its path with a dry-erase marker on a laminated page in his star atlas. We’re going to focus on three objects this evening: First up is an object the observers call 05683A. The number is a “pseudo-international designation,” Molczan explains. The object doesn’t appear in the Space Track catalog, meaning that it’s a classified object. Like a few other objects in orbit, the observers don’t know where it came from or what launch it’s associated with. They invented the “pseudo-international designation” as a naming scheme for objects like this one: “05” represents the year of the first sighting, and “683” is the day of year it was first sighted plus five hundred (because there are far fewer than five hundred satellites launched each year, the extra numbers prevent an amateur pseudo-international designation from being confused with an official designation). So object 05683A is an unidentified classified object first documented on the 183rd day of 2005. It seems to be lurking in a geosynchronous transfer orbit; it’s probably a spent rocket body. The second object we’re planning to look at is USA 129, a KEYHOLE-class imaging satellite, one of the NRO’s “crown jewels,” as Molczan puts it. The last object is a special request. I’ve asked to look at an object called 9928C, also known as USA 144deb. Molczan humors me: He spent the evening observing it a few days ago, and its elements are up-to-date. Observing USA 144deb tonight will have little value in terms of the data collected. But I get the impression that he welcomes my interest in it. If Molczan has a trademark satellite, it is USA 144deb. The only problem is that USA 144deb isn’t, strictly speaking, a spacecraft. It’s probably a decoy, meant to disguise a spacecraft Molczan can’t track.

  Even with access to the U.S. Strategic Command’s exhaustive unclassified catalog, coupled with the highly accurate map of the secret night sky compiled from decades of collective observation, there remains a small class of objects that continues to baffle the world’s most proficient amateur observers. A class of stealth satellites code-named MISTY continually thwarts Molczan. After more than fifteen years of trying to track the black spacecraft, Molczan has come to the conclusion that it’s personal.

  Molczan’s cat and mouse game with the MISTY program began on February 28, 1990, with the launch of space shuttle Atlantis Mission STS-36. The shuttle’s pilot for the classified mission was a man named John N. Casper, whose name I’d first encountered doing RESUMINT on astronauts. Casper’s NASA biography explains that after attending the USAF Test Pilot School at Edwards Air Force Base, he joined and later commanded the 6513th Flight Test Squadron. Translation: Casper was a Red Hat—he flew Soviet MiGs at Area 51 before going on to become deputy chief of the Special Projects Office at the Pentagon; Casper spent his career in the world of classified research, development, and procurement.

  A blurb in Aviation Week & Space Technology piqued Molczan’s interest in the mission. The trade magazine reported the project’s designation as AFP-731 (for Air Force Project 731), an “advanced reconnaissance satellite to be used by the Central Intelligence Agency and the National Security Agency,” and that it was “a ‘combination’ spacecraft carrying both digital imaging reconnaissance cameras and signal intelligence receivers.” Molczan’s trained eye noticed another unusual fact about the mission: the shuttle would fly into a 62-degree inclination, the steepest shuttle orbit before or since.

  Molczan assumed that STS-36 was another KEYHOLE-class object and that he’d eventually be able to track it, “but being impatient and wanting as much data as I could get, knowing launch date and time, it was easy to determine that it wasn’t going to be in range of most active observers. We were out of luck for weeks.” The mission wouldn’t be visible from his Toronto balcony, so, he said, “I needed to find observers in the north.” So with a bit of “inspired phone-calling,” Molczan recruited teams of agreeable amateur astronomers at Yellowknife in the Northwest Territories, at Whitehorse in the Yukon, and a third team in Alaska, then held a series of informal training sessions to teach the astronomers the basics of satellite observing. As the launch approached, Molczan calculated “look angles” for each of the teams, predicting where they could expect to see the object in the sky.

  “Things went really well,” Molczan explains, “It went into the orbit that it was supposed to be in from Aviation Week. The whole idea of tracking it was to refine the estimated orbits that I’d predicted. That all worked like a charm.” The observers in Whitehorse and Yellowknife supplied useful data. In the United Kingdom, Russell Eberst made some very precise observations. “A few days of this, and the guys up north asked if I had all the data I needed. Remember that it was about forty degrees below outside.” Molczan concluded that he did, even though it would be weeks before he’d actually be able to see the object from his home. Then something unexpected happened: “About a week goes by and a press release comes out from the Soviets saying that the satellite may have blown up.” An article in Aviation Week explained that the “apparent failure of the $500-million AFP-731 imaging reconnaissance satellite launched by the space shuttle Atlantis Feb. 28 is a serious setback in the U.S. strategic intelligence program,” and that “the apparent failure of the AFP-731 imaging spacecraft was the third in a series of major Western space failures in the last month.” Responding to the Soviet press release, the U. S. Defense Department issued a statement that “Shuttle Mission 36 achieved its goal associated with a classified Defense Dept. program. Hardware elements associated with the mission are expected to reenter the Earth’s atmosphere.” Molczan, like everyone else, assumed that the mission had been another billion-dollar boondoggle. He was in for a surprise.

  That November, eight months after AFP-731’s apparent explosion, Molczan remembers, “I get this message from Russell Eberst in Scotland, and others, and I realize we’ve seen precise unknowns on the same nights.” The European observers saw an especially bright object that none of them could identify. “Russell refined the orbits, and we expected it to match something already up there,” Molczan says. Assuming that the unknown object was a wayward rocket body or a forgotten Soviet spacecraft, Molczan checked the numbers against th
e Space Track data and his own records of classified orbits. He couldn’t ID it, so he turned to another set of records: objects that they’d tracked and subsequently lost. Using the observations of the unknown object, Molczan precessed the orbits back in time and “lo and behold, they lined up on the seventh of March, the day the Russians said that the satellite exploded.” The satellite “had been exceedingly bright and was still bright, especially considering its height of about eight hundred kilometers.” Molczan, Eberst, and other observers continued tracking the object for several weeks, reporting their sightings to one another on a BBS board, the precursor to the World Wide Web. From time to time, the press snooped in on the BBS conversations, and a reporter named Todd Halvorson at Florida Today, the local paper for Cape Canaveral, wrote up an article about the sightings. Pretty soon, the wires picked up the story. The New York Times eventually wrote that “amateur satellite trackers in Canada and Europe reported they had spotted the spy satellite in a higher orbit than anyone suspected, an indication that the craft was not only working but also highly maneuverable.”

  “Soon after it became public in the press it vanished again,” Molczan recounts. “We had a hunch that it was going to make some adjustment, but I fully expected it to be no challenge to think of.” Molczan assumed that the spacecraft had made a simple adjustment and that a straightforward search of the known orbital plane would recover the object. That didn’t happen. “It was just gone.”

  Over the following years, theories abounded in online conversations about what had happened to AFP-731. One theory was that the satellite went into a different orbit to support the first Gulf War. A rumor circulated that the bird had been “sacrificed” for the war. When he heard the word “sacrifice,” Molczan thought that it was transferred into a lower orbit where it could take higher-resolution images. By doing so, he thought, it had probably entered an orbit that induced a lot of drag and that it had depleted its propellant in the maneuver. It would be “sacrificed” because the lower orbit guaranteed that it would burn up in Earth’s atmosphere far sooner than intended. Another theory held that the National Reconnaissance Office put the bird into a higher orbit, which would provide longer dwell times over Iraq but compromise its longer-term mission. Another form of “sacrifice.” Molczan and the other observers conducted planar searches for both hypothesized orbits but came up empty-handed. AFP-731 was gone. A decade later, Molczan realized what probably happened.