At 2:30 a.m. on Feb. 7, 2003, a Cuban gunboat quietly docked at the Hyatt Marina in Key West, Florida. It had entered U.S. waters, the harbor and the marina next to the U.S. Coast Guard station undetected by the Coast Guard or the U.S. Navy.
The four men on the boat -- Cuban border guards in full uniform and carrying sidearms -- strode up to the Hyatt front desk. The desk clerk, discovering they had no reservations, turned them away.
Through the night they wandered Key West before coming upon a police officer, to whom they surrendered. Fortunately, the Cubans wanted only to defect.
The event has been recounted with some laughter. But as the Coast Guard commander who related it to us said, “It was no laughing matter that they were on American soil for four hours, free to move a weapon inland and be gone.”
Neither the Coast Guard nor the Navy had any detection measures in place where the Cubans’ gunboat had docked.
If the Sept. 11 attacks revealed security problems with airplanes, the Key West incident reminded everyone that the situation was even worse with ships. “We never knew what was going on in the oceans,” explained Mike Krieger, director of information policy for the Department of Defense’s chief information officer. “Hence we never knew what was coming to shore.”
Port Security Defined
The Coast Guard’s priorities had been safety on the waterways and fighting the drug war on the seas, not port security. “Prior to 9/11, I don’t know that you could find anybody who knew what port security was,” a shipping industry executive said.
Even after 2001, some officials downplayed the threat of a ship’s being weaponized and used as a bomb like the 9/11 jetliners. “Use a ship as a weapon to bring a major bridge down?” one Coast Guard officer asked. “There are easier things to do.”
Others perceived a serious danger. In the late fall of 2005, at President George W. Bush’s direction, Navy and Coast Guard admirals convened in Colorado Springs to map out a formal plan to achieve global “maritime domain awareness.”
The vision was for the world’s navies, ports and shippers to collaborate to protect shipping, harbors and cities from attack. But developing such a “concept of operations” was considered a challenge so large, it would take 18 months.
That’s right -- 18 months just to create a plan. Meanwhile, U.S. ports and those of its allies would remain vulnerable. And U.S. warships and commercial vessels entering foreign harbors would remain blind to hidden perils, even though information about a dangerous ship, cargo or crew might well exist in another agency’s or company’s database. There was no way to compile everyone’s data to tell friend from foe.
Even as the admirals were meeting in Colorado Springs, nearly 2,000 miles to the east a Coast Guard captain was inspecting the new Department of Homeland Security’s New York Harbor Watch Center, where plasma screens displayed every dock, warehouse and pier. If an incident occurred, people at all the other harbor watch centers in the country could presumably see what was going on in New York and size up the threat to their own port.
Yet the Coast Guard captain was mystified. The displays didn’t show a single vessel. This was one of the world’s busiest harbors. Where were the ships?
The watch center, he learned, did not yet have that feed.
It was a sad reflection of U.S. port security at the time: The Coast Guard had data on ships, Customs on cargoes, and the U.S. Department of Transportation on crews. Shipping companies had some awareness of all those elements for their own fleets. But no one had the whole picture
Back in Colorado Springs, as the three-day meeting was wrapping up, and everyone seemed resigned to the 18-month timeline, Krieger told the generals they could start sharing information in half that time. What he knew was that they could start collaborating faster by making the problem the right size -- find the common ground everyone can agree on now, figure out what can be done, and get started.
“This resonated with me,” Rear Admiral Joseph Nimmich recalled. He decided to continue with the 18-month effort to develop the concept of operation, but give Krieger a small team of Navy and Coast Guard technicians, operators and contractors and let him push ahead, as well.
To co-sponsor the effort, Nimmich recruited U.S. Navy Rear Admiral Nancy E. Brown, the most senior officer responsible for all systems for the U.S. North American Command.
At its first meeting, in 2006, the team settled on a statement of what it intended to do: “Get some worldwide situational awareness of what’s on the seas.” But Krieger wanted to narrow the scope to a goal that could be met in six to 12 months. So the team came up with a plan, first, to see how much data could be shared from the agencies’ individual databases -- to create a picture of what was on the water.
It would be difficult, however, to make sense of shared information because the various databases used different names for the same object. And there were thousands of different objects.
The database owners had to agree, for example, that the terms “ship,” “boat” and “vessel” would mean the same thing across all systems, and then give every mention of a ship, boat or vessel in each database the same tag. That way, if you searched the tag, you could see every ship, boat or vessel, no matter what anyone called it.
Maritime Traffic Watch
With thousands of maritime object terms in their databases, getting agreements on tags and then executing could take months and millions of dollars. And that might add up to only a partial solution, not full awareness of maritime traffic for all the agencies that need it.
So the team asked itself, “What if we do this only for a few important terms like ‘vessel’”? If tests could show that any authorized person could search only for a few tags to discover all mentions of everything afloat, then each agency could see what all the others saw.
Thus did Krieger’s group narrow its scope from building worldwide awareness to sharing information to simply tagging and sharing a few key terms. And that well-defined challenge became what the group decided it could deliver in nine to 12 months.
The questions the team ended up asking were “simple in the extreme,” said John Shea, who was project manager. “When we search for a tag, can we find it? When we pull the data up into our native viewers, can we see it?”
In his day job, Shea was technical director for the Navy’s “C4I” network operations -- command, control, computers and communications -- and was already at work improving the Navy’s ability to track maritime traffic. His work with Krieger’s team allowed him to collaborate not just with fellow Navy technicians and contractors but also with others from the Coast Guard, the Department of Transportation and private companies as well. If the team was successful, it would save Shea time, money and work and also let the Navy take advantage of the eyes and ears of other networks.
By November 2006, the team was ready to show its results to the highest levels of the U.S. Navy and Coast Guard. In the boardroom of the chief of naval operations, Singapore Harbor was displayed from a single system on board a Navy vessel. There were some 500 “tracks,” or ships. At the push of a button, new data were added from the Department of Transportation, the Coast Guard and the Office of Naval Intelligence. Now 1,500 tracks appeared on the display, many far out of view, over the horizon.
In nine months, Krieger had delivered.
(William J. Bratton, who led the Los Angeles, New York and Boston police departments, is the chairman of Kroll, a risk consulting company. Zachary Tumin is a senior researcher at Harvard Kennedy School’s Belfer Center for Science and International Affairs. This is the first in a three-part series of excerpts from their book, “Collaborate or Perish!: Reaching Across Boundaries in a Networked World,” to be published Jan. 17 by Crown Business. The opinions expressed are their own. Read Part 2 and Part 3.)
Read more opinion online from Bloomberg View.
To contact the writers of this article: William J. Bratton and Zachary Tumin at email@example.com
To contact the editor responsible for this article: Mary Duenwald at firstname.lastname@example.org