The problem was that Boeing did not yet have a functional system. “If you go off and say I want to build something that can do everything, you’ll never build anything.” The prototype went up in Arizona, Project 28, and the results were not good. The work was sloppy; the cameras were out of focus; the satellite system had too much latency delay; the towers were not structurally sound; the radar had too much clutter. The Border Patrol hated it. “We never tested it. We just put it up, turned it on. Didn’t work!” Borkowski says. At that point, the “clue light comes on.”
Borkowski had come to DHS from the air force, where he had long experience dealing with military contractors. When he arrived and began to see what had been going on with SBInet, he realized he had a big mess on his hands. “I thought, Oh my God, this program is doomed.” Before long, however, Borkowski found himself trying to salvage what he could from the wreckage. Changes were made, and ambitions were scaled back. Then, after a yearlong cost-benefit analysis and reassessment, Secretary Janet Napolitano decided to kill the project.
Although he freely admitted that a great deal of time and money was wasted, Borkowski argued that CBP and DHS had made progress. “SBInet, for a whole lot of reasons,” Borkowski said, “does not make sense” as a universal solution to the border security problem. But the integrated tower systems that were built do have some utility, in the right terrain. The hard part in general is to find solutions that fit the particular challenges of different stretches of the border. Borkowski advocates targeting technological solutions toward specific problems, with clearly defined objectives, using equipment that has been field-tested and is ready to go. Not every sector needs every kind of technology. Not every mile of the border requires a fence. Not every bend of the river requires a surveillance tower. Ground radar simply doesn’t work in some areas, and drones are not a silver bullet. Nor, he made clear, will every border security problem be solved by the application of new technologies. The trick will be to get the border situation managed to the point that other resources, such as policy, such as comprehensive immigration reform, can pick up the difference. “That would cut off a lot of the traffic between the points of entry; in fact, at a certain point, you would only have the really bad people left, the drug smugglers and the terrorists.” Of course, there aren’t many terrorists wandering around in northern Mexico, and Borkowski made no mention of comprehensive drug policy reform.
After mounting a fairly persuasive critique of the idea that technology is the key to securing the border, as I was leaving his office Borkowski nonetheless remarked that CBP was beginning to realize that as an agency it needs to get all of its technological assets connected in one place, and that place would seem to be the Big Pipe.
—
Not many of the agents and officers I spoke with in South Texas knew anything about the Big Pipe; others had heard about it but weren’t able to offer any specifics. At GovSec, the big government security trade show in Washington, D.C., none of the vendors I spoke with had heard of the Big Pipe, not even the people who were selling IP surveillance cameras to state, local, and federal agencies. It was completely under the radar.
At its most basic level, the Big Pipe is a video distribution hub, a web portal that integrates video feeds from a variety of assets and makes them available via the DHS secure network and the Internet. When Kenneth Knight was in Brownsville coordinating air support for the Zapata funeral, one of his prime objectives, aside from providing air support to the security operation, was setting up the helicopter video feed, which was transmitted via direct downlink from the helicopter to a microwave antenna, which he had installed on the roof of the Border Patrol station for just this mission. From there, the signal feeds into an encoder, which then pushes the video back through the DHS1 secure network to the National Data Center in Washington and outward into the Internet cloud. When we sat together in Brownsville, Knight pulled up the Big Pipe portal on a Border Patrol PC and logged in, and within a few mouse clicks we had the helicopter’s video feed on the screen. In this case, the whole point was to make a God’s-eye perspective available to whoever might need it—operational supervisors in a situation room or people in other locations who were unable to attend the funeral.
Several weeks later, in Washington, we went back over the architecture of the system. The core, as with any web application, was the server, which hosts mission feeds originating in assets such as the Predator—or a Guardian, the maritime version of that drone. Predators, unlike traditional metal-framed aircraft, have a composite skin, so one can put a parabolic dish inside the aircraft, which permits the transmission of extremely high definition video over a Ku satellite system to the National Data Center, from which point the data can be pushed out through the Big Pipe to wherever it needs to be. Other sensors, such as the helicopter-mounted camera I saw in Del Rio, or the camera in a scope truck, transmit their feeds via microwave downlink. Still others might be hooked directly into a router, such as an IP video surveillance camera in the Atlanta airport or on the side of a building in Washington, D.C.
One example that Knight gave me was a Guardian mission in the Bahamas in which the drone encounters an unidentified watercraft. All the individuals, no matter where they might be, who have an interest in this mission can be logged in to the Big Pipe, each one watching the same video feed in real time. Using a split screen, everyone can also be viewing a moving map, a maritime chart that shows the drone’s position, in real time, relative to the target and anything else that happens to be in nearby waters. In addition to watching the video simultaneously, they would all be communicating by means of a chat log. The drone operator might not be able to identify the particular ship that’s being “rigged,” but a Coast Guard analyst who is also online could pipe in and identify the craft without having to wait for the pilot to verbalize what he thinks he’s seeing down there on the water. Perhaps it’s a shrimper, or a type of vessel known to be used by drug traffickers. The Coast Guard might even have intelligence on that very craft. Whatever happens, all the video, the chat logs, the geo-spatial coordinates, the flight path, flight points, the line of sight, sensor points, marine surface radar pictures, and all the other metadata associated with the mission are recorded and fused, synced to the video, and bundled in a mission package that typically includes a geo-spatial page containing a graphical representation of the flight path, with video that can be rapidly cued by clicking at any point on the path. Click at a given point in the flight path, and you see what the camera was looking at right then, and you have access to the audio, the flight log, altitude, directional orientation, air speed—everything. After the mission is complete, the OAM generates this mission package and hands it off to the agency customer, perhaps JIATF South, the Joint Interagency Task Force South, which monitors suspect aircraft and marine vessels in the Caribbean and nearby waters.
Knight spoke quickly, running through examples and illustrations, development strategies, and plans for the future. Far from being merely a Homeland Security video server, it was clear, the Big Pipe aimed to be a comprehensive mission data distribution system, whatever that data might be, though surveillance video was often the most crucial component of the operational picture.
“We’re doing some really cool shit,” Knight told me when we first met. “The military does some of the same stuff, but they can’t do what we do. They work in the classified world; we actually cross domains.” The distinction was a crucial one. A military version of the Big Pipe would never be made available via the Internet to civilian law enforcement or to FEMA during a natural disaster or to some other partner outside the military domain. “We’re literally paving the way. The cool thing is that a lot of the stuff you’re seeing with the Border Patrol, like the RVSS towers, the mobile surveillance systems, all that gets integrated with this. So now we have worldwide distribution capability for situational awareness and exploitation purposes.”
Given that the system was created from within the OAM, it was perhaps natural that Kni
ght focused initially on getting the sensor data off the aircraft and into the network. That also happened to be the hardest piece of the puzzle. The infrastructure for the unmanned aerial systems was relatively easy, however, because Predators can use the government’s Ku satellite network. Microwave downlink from other aircraft has been more difficult. “Wireless just doesn’t work very well for us,” Knight told me, “because the bandwidth isn’t there.” Consequently, BMS microwave antennas are going up. “We’re putting these things in strategic locations around the country. They reach out about seventy-five miles,” Knight said. “We’re closing the gaps.” The next stage was IP video, which can capture the feed from any IP-enabled camera that is made available to the system: CBP surveillance video at ports of entry, at shipping ports, checkpoints, airports, train stations, on the sides of buildings, anywhere. At that point, sitting in his office in Washington or at any registered computer anywhere in the world, Knight could click from a TSA surveillance camera pointed at the security line at the Atlanta airport to a feed originating from an aircraft such as a P3, a helicopter, or a Predator monitoring the Red River flood. Click, scroll, click; just like that. Getting more coverage would seem to be a matter of logistics as much as anything, making contact with other agencies, making sure that systems talk to one another, ensuring that laws and regulations are respected, programming firewalls, and such.
Supporting disaster response to something like the spring floods in the Midwest would be an ideal mission for a drone and the Big Pipe. The Predator, because it can remain airborne for so long, can use its synthetic aperture radar, or SAR, to map the flood zone of an entire state in one flight. Prior to that year’s flood season, a Predator had been doing just that, and with the SAR providing change detection based on previous pictures, actually highlighting the differences as the situation on the ground evolves. This capability can give response teams on the ground crucial intelligence about flood patterns or ice breakage in real time. And, as it happens, the Big Pipe was born during a national disaster, in Corpus Christi, Texas, when the OAM was working to provide air support to the phalanx of agencies responding to Hurricane Katrina. When Knight started, he only had five feeds, but it was working, and extending the capacity of the system was mostly a matter of building the right systems architecture. That was no small task. “I built this thing, and then I just looked at it and said, Oh, my God, what did I just do? I’ve created a monster.”
From that original do-it-yourself server in Corpus Christi, the evolution of the Big Pipe has been rapid. It is now supported by CBP’s Office of Information and Technology and has dedicated developers who work to ensure that it will function on every conceivable platform, from old x86 PCs in a sheriff’s office somewhere using a slow Internet connection to a PDA or smartphone on a 3G mobile network. A custom content management system allows Knight to control access with whatever granularity he desires, giving the customer only what he has a need to see.
Although it is part of Customs and Border Protection, the OAM plays a support role for an array of other agencies—other Homeland Security agencies such as ICE, the TSA, the Secret Service, and FEMA, as well as the DEA, the FBI, NASA, NOAA, the Coast Guard, JIATF South, NORAD, and the U.S. Geological Survey. Knight told me he had at least one hundred DoD partners, though he cautioned that in accordance with the Posse Comitatus Act the military receives only limited access to the Big Pipe, on a temporary basis, in operations such as disaster response. He said he worked with the National Guard, state-level authorities such as the Texas Department of Public Safety, local police, and sheriff’s departments. Then there was the Air and Marine Operations Center, in Riverside, California, which integrates the Big Pipe into its core mission of monitoring and securing the entire national airspace, as well as the marine environment extending one hundred nautical miles offshore.
In the business of border security, the potential for the Big Pipe to provide the architecture of a common operating picture for a supervising agent, sitting in a control room like the one I saw in Brownsville, was pretty clear. Rather than just staring at a flickering sequence of video feeds from RVSS cameras, one begins to imagine a geo-spatial interface on a large screen showing every RVSS tower in a sector, along with every ground sensor and the ports of entry, with their hundreds of cameras. The radios used by the Border Patrol are all GPS enabled, so it’s no great leap to integrate them as well, so that anytime an agent cues the radio it shows up as a blip on the screen. Add the mobile surveillance systems like the scope trucks, the aircraft, and the riverboats, and you would have all your fixed infrastructure and all your mobile assets captured and represented graphically on a digital map, in real time, with the ability to click in and see what’s happening from any point in the system. The supervising agent in charge would know where all his agents are, and where the aircraft are, and he could see at a glance where they were in relation to one another.
Such a system would approximate the command and control software used in war zones around the world. But Knight wasn’t just talking about a specific operational zone like the Rio Grande valley sector; he was aiming at a much larger domain, fusing the national air radar picture with the coastal marine surface radar picture, and adding in all the CBP feeds along the border, and the video surveillance cameras in the ports, along with airborne video sources, as well as surveillance assets in metropolitan areas, so that the scope of operations expanded to the widest possible extent. This broad spectrum of surveillance is really what Knight seemed to have in mind when he spoke about total domain awareness—an operating picture that encompassed pretty much the entire United States. At that point, the potential for real-time persistent surveillance of a target comes into view. Total domain awareness means the ability to apply these tools anywhere in the United States.
Knight came back to his example of the Guardian drone following a target at sea. As that target vessel approaches Miami, the Guardian, which as an unmanned vehicle is not permitted to fly into the airspace of a major metropolitan area, hands off the target to a P3 aircraft. Then, as the vessel enters the port, it gets handed off to fixed video cameras, whereupon ground personnel can also play a role. One platform can’t do it all; the air assets can’t stay airborne forever, the still cameras can’t move, “but if you start putting all these camera systems together, you’ve functionally closed the gap.” Over time, Knight said, “we might be able to provide persistent surveillance coverage of a target, moving from outdoors to indoors in a major metropolitan area, from the airport to downtown city streets.” All through the Big Pipe. Knight laughed. “It’s a lot like what you see in some of these—what was that show called, 24? It’s the same principle behind it. But we’ve been working on it since before that show came out. At least I have been.”
As I listened to Knight talk about his vision for the Big Pipe, Borkowski’s skepticism about the functional payoff of a common operating picture continued to resonate with me. It wasn’t clear, for example, that a fully robust Big Pipe could have prevented the gun that killed Jaime Zapata from ending up in the Zetas’ arsenal. But judging from the logistical expertise that I saw demonstrated by Customs and Border Protection at the World Trade Bridge in Laredo, and by the surveillance infrastructure CBP is building along the southwest border, I had little doubt that Knight’s ambitions were not only technically possible but almost logically inevitable.
Yet what was most striking to me, as I reflected on the full scope of the Big Pipe and the ambitions of its creators, was the extent to which the border has become a laboratory, a controlled environment in which new security techniques can be perfected and where military tactics can be adapted for domestic application. The mission of securing our national borders has become indistinguishable from a new and still emerging understanding of what constitutes homeland security. It was hard to avoid the conclusion that the border itself was slowly expanding to fill the entire continent. Before long, we all may find ourselves inside the Big Pipe.
—
West from Del Rio the border grows wide as the Rio Grande recedes behind private ranchland. There are a few points of approach, although there is at least one place where you can cross the international boundary with impunity, Amistad Reservoir. Not long after I said farewell to my guides in the Border Patrol, my sons and I spent a pleasant morning reeling in black bass under the expert eye of a fishing guide by the name of Ray Hanselman. I had no idea we were in Mexico until Ray mentioned that the spot we were fishing was called Lobo Canyon. Lobo Canyon? I asked.
That’s right, said Ray.
In Mexico? I asked.
Yep.
I just stared straight ahead, blinking. There wasn’t much I could do about it at that point, but I couldn’t help but think about that couple on Jet Skis who came under fire from pirates on Falcon Lake, in Zapata County, just a few hours south. The epidemic of kidnapping was no joke.
Don’t worry, said Ray. I have all the permits we need.
Farther west, past Comstock, the highway swings within sight of the border at the mouth of the Pecos and a few miles later at Langtry, where in 1896 Judge Roy Bean staged a world heavyweight boxing championship on a sandbar in the Rio Grande.
That was a time of optimism and expansion. The Indian wars were over, and the western reaches of the state were opening up. What’s left now is largely ruin—crumbling towns, rotting remnants like Cedar Station, Pumpville, Dryden. People come out here to hunt, the railroad still runs through it, and there are still a few working ranches, like the one my family has operated since the 1970s, despite the drug traffickers who sometimes threaten to turn our long perilous road down to the river into a smugglers’ superhighway.
Texas Blood Page 21
