As a computer-science doctoral candidate at the University of Southern California’s industrial engineering department, Praveen Paruchuri titled his 2007 dissertation “Keeping the Adversary Guessing.” At its core was a randomization algorithm, based on game-theory modeling, that makes it harder for criminals to stake out, map, and elude security patrols. With the nurturing of professors and grad students at USC’s Center for Risk and Economic Analysis of Terrorism Events (CREATE), Paruchuri developed the algorithm into a software program that allows for the “intelligent randomization” of security deployments – including canine patrols and vehicle checkpoints – and keeps the bad guys guessing just where and when they’ll run into the police.
The software, nicknamed ARMOR, proved so successful at increasing the interception of drugs and other illegal goods funneling through Los Angeles International Airport that it’s now being used to schedule federal air marshal patrols. It’s also being put into service at other airports and ports throughout the nation.
ARMOR was a feather in the cap of CREATE, one of a dozen university-led Centers of Excellence operated by the U.S. Department of Homeland Security (DHS). The centers’ remit is to enhance the nation’s security by tapping into the brainpower of a large, multidisciplinary set of scientists and engineers. Their wide-ranging tasks include bolstering port security, bomb detection, and food safety, and mitigating the effects of natural disasters. Beyond ARMOR, the centers have inarguably produced a string of other successes, including tracking down the source of a major salmonella outbreak and helping to predict which coastal areas of the Gulf of Mexico were most at risk from BP’s Deepwater Horizon oil spill. “We’re doing very well with these centers,” says Matthew Clark, director of DHS’s University Programs. “They’ve been very efficient.”
Perhaps. But successful outcomes don’t guarantee continued federal funding. At a time of intense political pressure to reduce the deficit, the White House wants to cut the centers’ budget by 20 percent, to $40 million, and the House has gone along. Senate appropriators would maintain funding at $50 million. If the two chambers split the difference, the result would be a 10 percent cut.
DHS’s Science and Technology Directorate, of which the university programs are part, has long had a rocky relationship with Capitol Hill. DHS itself has routinely been criticized as bloated, inefficient, and wasteful. Created by Congress in a post-9/11 rush to improve America’s defenses after the 2001 terrorist attacks, it comprises a variety of agencies, including the U.S. Customs and Borders Protection, Transportation Security Administration, Federal Emergency Management Agency (FEMA), and U.S. Coast Guard.
A 2008 congressional report said DHS blew $15 billion on initial payments for contracts that were either canceled or well over budget. A 2006 GAO report accused DHS staff of wasting up to $2 billion through the misuse of government credit cards. Then there was the antiterrorism data-mining software system, nicknamed ADVISE, that the agency canceled in 2007 after spending $42 million. It was axed after it was discovered researchers had inappropriately used live personal data to test it. Today, the centers stress that data mining is strictly verboten. “That’s kind of a four-letter word for us,” explains Timothy Collins, managing director of the Command, Control, and Interoperability (C2I) Center at Purdue University, which works on visualizing data and is co-led by Rutgers University, which handles data collection and analysis.
Since the centers were initially created, their number and structure have changed several times, and Congress has expressed some doubt that they were giving DHS sufficient bang for the buck. The House has asked the Government Accountability Office, Congress’s research arm, to evaluate DHS’s method of doling out millions of dollars in research money to its 12 centers. “It’s asking, ‘Are you running the centers in the best way possible?’” observes Gerald Epstein, director of the Center for Science, Technology, and Security Policy at the American Association for the Advancement of Science.
The fact that universities don’t compete for individual projects doesn’t mean the centers are uncompetitive, argues Michael Bruno. He directs one of them, the Center for the Maritime, Island, and Remote and Extreme Environment Security at the Stevens Institute of Technology in New Jersey. Competition among campuses to gain a center is fierce, he says, so it’s not as though unqualified schools have won the appointment. In addition, other schools with expertise in areas useful to a center often get involved in the research as partners. A major effort at his maritime center involves using multiple layers of sensors – underwater acoustic, high-frequency radar, and satellite – to detect, identify, and track ships, both on the high seas and in busy ports and harbors. Every boat has its own individual wake and sound signature, much like a fingerprint. Using these types of sensors to identify them is novel, however. “We are breaking new ground with the passive acoustic [underwater] sensors,” Bruno says. To do so, he is always looking for partner schools to help with the effort. “It’s not a case of ‘the haves continue to get,’” he notes.
Though the centers number just 12 in all, they encompass many other universities beyond the ones at which they’re based, as well as private businesses, think tanks, and national laboratories. USC’s CREATE, for example, has 14 partners, including Georgetown, Howard, and New York universities, Rand Corp., and the Universities of Virginia, Wisconsin, and Hawaii, Hilo. The centers work on an assignment basis and between them have 200 to 300 projects going at any one time. Unlike much federally funded research, which is usually basic, they do only applied research on a fast-turnaround basis.
“I tell my centers that I want Nobel-level research, and I want it tomorrow,” Clark explains. They also have to be ready to react rapidly to events. After pet food ingredients imported from China were found in 2007 to be inadvertently contaminated with melamine, Clark had the Center for Food Protection and Defense at the University of Minnesota do a quick assessment of the vulnerability of America’s human food supply to intentional poisoning from overseas. Short answer: very, because testing wasn’t keeping up with increasing exports from developing countries.
Most academics aren’t used to inventions-on-demand type research and schedules. Nevertheless, the centers claim that faculty members are eager to join their staffs. “This is putting [their research] into action, and they appreciate seeing it go from the chalkboard to the streets,” says Jack Jarmon, associate director of the Rutger’s center. “And for taxpayers, this is the best return on their investment we can give.”
Over the past five years, centers have received annual amounts ranging between $2 million and $4 million each. “But our mission is to maximize the return on our investment,” Clark says. That means the centers are strongly encouraged to seek additional funding from other federal agencies for complementary research. During the 2004 - 09 period, the centers brought in an additional $105 million in funding from other sources. Rick Luettich, director of the Center for Natural Disasters, Coastal Infrastructure, and Emergency Management, University of North Carolina, Chapel Hill, reckons that on average his center’s principal investigators get only a quarter of their funds from DHS, with the remainder coming from other agencies.
Moreover, Clark argues, the centers pay for themselves in other ways. A large salmonella outbreak in 2008 that sickened hundreds of people in 40 states was initially blamed on Florida tomatoes. But after the Minnesota center got involved, its research quickly determined that the ailment was showing up in
places where the tomatoes hadn’t gone. Ultimately, the culprit turned out to be jalapeño peppers from Mexico. That finding allowed a quarantine on Florida tomatoes to be lifted, saving that industry $220 million.
More recently, Chapel Hill’s Luettich was able to adjust modeling software his center originally developed to predict how coastal areas will be affected by storm surges and flooding. In its new configuration, the program helped determine which Gulf shorelines were most at risk of being coated with oil from the BP spill. “The transport of oil was a new phenomenon for us, but we did come up with some helpful scenarios,” Luettich says. And they helped guide containment crews placing floating booms to catch the oil before it hit land. “The advances we made will find other uses,” he predicts.
Stevens’s Bruno argues, “For a relatively small amount of money, these centers can have enormous impact. Even at the higher level [of funding] there was a feeling that it wasn’t adequate to achieve their full potential.” But if budget trends continue, that’s a potential that may soon be even further out of reach.
Thomas K. Grose is Prism's chief correspondent, based in the United Kingdom.