News this Week

Science  22 Nov 2013:
Vol. 342, Issue 6161, pp. 914

You are currently viewing the .

View Full Text

Log in to view the full text

Log in through your institution

Log in through your institution

  1. Around the World

    1 - Warsaw
    Climate Negotiators Pressed About Ocean Acidification
    2 - Taipei
    Yet Another Avian Flu Virus Found in a Human
    3 - Washington, D.C.
    Congress Boosts Chimp Retirement Funding
    4 - The Hague, the Netherlands
    Dutch Push for Open Access
    5 - Singapore
    Zoological Naming Authority Rescued
    6 - Zhangye, China
    City Bans 'Genetic Bombs'


    Climate Negotiators Pressed About Ocean Acidification


    Acidic oceans will likely erode coral reefs faster than they can build up.


    The argument for cutting greenhouse gas emissions often rests on global warming's impact on terra firma. But scientists are increasingly confident that climate change will disrupt the oceans as well. Researchers working with the International Geosphere-Biosphere Programme argued that case this week at a meeting of the United Nations Framework Convention on Climate Change in Poland.

    Marine waters are becoming more acidic as they absorb carbon dioxide from the atmosphere, and the researchers warned convention delegates that they had "high confidence" that increasing acidity will erode coral reefs in the tropics faster than they can build up. Mollusks and other shelled creatures will also suffer, although the extent of the economic harm is hard to predict, they further noted in a summary report released at the meeting. Reducing emissions could slow, but not prevent, the changes.

    The meeting, which wraps up 22 November, is intended to help delegates prepare for U.N. negotiations in 2015, which they hope will yield a major climate treaty.


    Yet Another Avian Flu Virus Found in a Human

    Scientists in Taiwan have identified the first known human infection with the avian influenza virus subtype H6N1. The female patient, hospitalized in May, suffered moderate but not life-threatening illness and recovered after antiviral drug treatment.

    H6N1 is common in wild and domestic birds worldwide and has affected Taiwan's poultry since the 1970s, causing minor symptoms. But it had never before been found in humans—even though the pre dominant H6N1 strain circulating in poultry in Taiwan since 2000 has a genetic mutation that might allow it to more readily latch on to human lung cells. "It is possible that people have been infected but not recognized due to their mild symptoms," says Ho-Sheng Wu of Taiwan's Centers for Disease Control in Taipei. The recent patient had had no contact with live poultry, and no close contacts had been infected. If the virus mutates further, it could threaten human health, Wu and his colleagues reported online on 14 November in The Lancet Respiratory Medicine.

    Washington, D.C.

    Congress Boosts Chimp Retirement Funding

    Golden years.

    More NIH research chimpanzees will be heading to Chimp Haven.


    Congress approved a bill last week that will allow the National Institutes of Health (NIH) to move forward with plans to retire most of its research chimpanzees. NIH decided earlier this year to phase out most NIH-funded invasive studies on chimpanzees and to retire all but 50 of its 360 research chimpanzees. But its spending on the federal chimpanzee sanctuary, housed at Chimp Haven in Keithville, Louisiana, was expected to hit a $30 million cap, instituted by Congress in 2000, by the end of November.

    Both the House of Representatives and the Senate approved a bill to lift the spending cap, allowing NIH to spend up to $9 million to $12 million a year over the next 5 years on the federal sanctuary if that is cheaper than keeping NIH's chimpanzees at research facilities (which it should be). The bill has now gone to President Barack Obama for his signature.

    The Hague, the Netherlands

    Dutch Push for Open Access

    The Dutch government wants to make all publicly funded research freely accessible within 10 years—and it believes the "gold" model, in which authors pay to be published in freely available online journals, is the way to get there. In a 15 November letter to the Dutch House of Representatives, State Secretary for Education, Culture and Science Sander Dekker announced measures to speed up the transition to open access (OA): The government will host a 2014 roundtable between scientific organizations and academic publishers and will require universities and research organizations to produce annual OA reports. It may make OA compulsory by law in 2016.

    Dekker's clear preference for "gold," favored by publishers because it protects their revenues, has irked some OA watchers. The Dutch government "fell for the publishing lobby's nocturnal fantasy," says OA advocate Stevan Harnad, who advocates a bigger role for "green" OA, in which authors can publish in any journal as long as they archive a copy of each paper in a freely accessible electronic archive.


    Zoological Naming Authority Rescued

    The International Commission on Zoological Nomenclature (ICZN) has found a new home in Asia and—at least for the moment—some financial security. Since 1895, the London-based ICZN has set rules for naming newly identified species and resolved disputes over names for animals both living and extinct. But the U.K.-based charitable trust that supported ICZN has run out of money. The National University of Singapore (NUS) and the Natural History Museum in London announced on 18 November that they will come to the commission's aid: While the editor of ICZN's Bulletin of Zoological Nomenclature will remain at the Natural History Museum, the commission's secretariat, with one full-time staffer coordinating its worldwide activities, will be based at NUS. The arrangement is intended to provide some breathing room so that ICZN can ponder ways to ensure its long-term survival. Vice President Daphne Fautin predicts that if the commission disappeared, the world of nomenclature would fall into "a period of chaos."

    Zhangye, China

    City Bans 'Genetic Bombs'

    GM cotton


    A new round of debate over genetically modified (GM) crops has erupted this month in China after the northwest city of Zhangye on 25 October announced a ban on the sale and use of GM seeds. The ban—part of the city's regulation to ensure food safety and to support organic crops—makes Zhangye the first city in China to officially outlaw the growing of GM crops.

    Environmental groups applauded the move, but it is at odds with a national policy that promotes the production of new GM crops. In defending the decision, Chen Kegong, Zhangye's Communist Party secretary, on 5 November told reporters the city will "unyieldingly uphold" the ban on planting "genetic bombs."

    China grows more than 4 million hectares of GM crops, mostly cotton expressing a pesticide protein, although approval for commercial cultivation of GM corn and rice remains delayed. Luo Yunbo of China Agricultural University says the ban doesn't reflect a broader change in national policy.

  2. Random Sample


    Some planetary scientists last week expressed concern over NASA's 15 November decision to cancel the Advanced Stirling Radioisotope Generator project. Dwindling supplies of plutonium-238 had motivated the space agency to invest in a more efficient isotope-based power source for deep space missions, but the U.S. Department of Energy resumed production of the radioactive element this year. Budget constraints also prompted the move, NASA said.

    'Panda of Indochina' Makes a Camera-Trap Cameo


    A last-gasp attempt to save the critically endangered saola may be paying dividends. For the first time since 1999, this antelopelike creature has been spotted in the wild. The new photo documentation—from a camera trap in Vietnam's Annamite Mountains—is a huge morale-booster, says zoologist William Robichaud, coordinator of the International Union for Conservation of Nature's Saola Working Group. It "shows we should never give up hope."

    The saola was the first large mammal discovered in half a century when it was described in 1992. But the ungulate gets caught in snares set for civets and deer, and its numbers have dwindled to a few hundred at most.

    Over the past few years, Vietnamese villagers have removed more than 30,000 snares from critical saola habitat. The secret spot where the new saola was spotted "will be completely locked down," Robichaud says.

    Epic Sperm Fight Wins 2013 'Dance Your Ph.D.' Contest


    The votes are in, and the top prize for the 2013 "Dance Your Ph.D." competition goes to … Cedric Tan, a postdoc biologist at the University of Oxford in the United Kingdom, who finished his Ph.D. there last year with a thesis titled "Sperm competition between brothers and female choice." His dance interpretation of that research illustrates the chicken mating process using a range of dance styles, from swing and water ballet—yes, in actual water—to modern jazz and what can only be described as cockfighting.

    The contest, now in its 6th year, is sponsored by Science magazine and AAAS (publisher of Science). Based on votes from previous winners and an independent panel of artists and scientists, Tan won both the biology category and the overall prize: $1000 and a trip—sponsored by HighWire Press—to screen his video at Stanford University in Palo Alto, California. Tan spent a year on the video, in the midst of a busy field season massaging male chickens to extract their sperm. "I assembled the team, trained them, forced tight skimpy attires on them (they complained a lot), forced them into the freezing cold lake (they hated me for it)," Tan explains by e-mail. According to the judges, this year was the strongest yet. Watch Tan's video, plus the $500 winners for the chemistry, physics, and social sciences categories at

    By the Numbers

    62% Portion of Europeans who feel that "science makes our ways of life change too fast," based on a European Commission public opinion survey published last week.

    1 to 2 million Additional babies to be born in China each year as a result of loosening the one-child policy, according to demographers (the present rate: 15 million per year).


    Join us on Thursday, 5 December, at 3 p.m. EST for a live chat with experts on a hot topic in science.

  3. Newsmakers

    Two Researchers to Round Out DOE Science Team



    President Barack Obama has added two academic researchers to his science team at the U.S. Department of Energy (DOE). On 18 November, he nominated chemical engineer Franklin "Lynn" Orr, a professor and administrator at Stanford University in California, to fill the newly created position of undersecretary for science and energy. The new post will oversee DOE's Office of Science, as well as fossil, nuclear, and renewable energy research programs and electrical grid and technology transfer issues. Orr directs Stanford's Precourt Institute for Energy and served as dean of the School of Earth Sciences from 1994 to 2002.



    Obama tapped physicist Marc Kastner of the Massachusetts Institute of Technology (MIT) in Cambridge to lead the Office of Science. He has served as dean of MIT's School of Science since 2007. "Obama made an inspired choice," MIT President L. Rafael Reif said in a statement, adding that Kastner is "ideally suited" to manage DOE's $4.6 billion research portfolio. If confirmed by the U.S. Senate, Kastner would replace William Brinkman, who left this past April.

  4. When Mice Mislead

    1. Jennifer Couzin-Frankel

    Tackling a long-standing disconnect between animal and human studies, some charge that animal researchers need stricter safeguards and better statistics to ensure their science is solid.


    Three mice had vanished. And Ulrich Dirnagl had a hunch about where they'd ended up: in the metaphorical dustbin housing animals—and there are lots of them—that line up at an experiment's starting line but are discarded before the finish. The paper that Dirnagl, director of the Center for Stroke Research at Charité University Medicine Berlin, was reviewing described how a new drug protected a rodent's brain after a stroke. The authors used 20 mice, half of which got the therapy. But mysteriously, only seven of the 10 treated animals appeared in a graph analyzing the results.

    "I wrote to the editor and said, 'I cannot judge this paper, I need to know where the three mice went,'" Dirnagl recalls. For 6 months, radio silence. Then, the editor responded. He'd heard from the authors, he told Dirnagl. The three mice, suffering from massive strokes, had died, and the authors had simply left them out of the paper. Extra analysis of their stroke drug, however, revealed that those mice had an important message to bear: The therapy harmed the brain rather than helping it.

    "This isn't fraud," says Dirnagl, who often works with mice. Dropping animals from a research study for any number of reasons, he explains, is an entrenched, accepted part of the culture. "You look at your data, there are no rules. … People exclude animals at their whim, they just do it and they don't report it." That bad habit, he believes, is one of several that plague animal studies.

    For years, researchers, pharmaceutical companies, drug regulators, and even the general public have lamented how rarely therapies that cure animals do much of anything for humans. Much attention has focused on whether mice with different diseases accurately reflect what happens in sick people. But Dirnagl and some others suggest there's another equally acute problem. Many animal studies are poorly done, they say, and if conducted with greater rigor they'd be a much more reliable predictor of human biology.


    It's hard to generalize, of course: Animal studies cut across a massive swath of biology, tracking everything from the activity of single molecules in a healthy organ to side effects of a new drug poised for human testing. And many who stake their careers on animal studies conduct them with care, judiciously weighing how to structure their experiments and chasing the science wherever their furry subjects take it.

    That said, even animal research that has a big effect on human drug studies—like the work Dirnagl reviewed—is governed by far fewer standards than clinical trials in people. There, volunteers are randomly assigned by computer to get a new drug or a placebo. Those running a trial are often blinded to who's in what category, preventing clinicians invested in a therapy's success from imagining hints of efficacy in patients they know are getting a new drug. And look up any clinical trial seeking volunteers and you'll see a long list of "inclusion" and "exclusion" criteria governing who can participate. If you have high blood pressure or if your cancer is being treated with a certain drug, you might be out of luck.

    Animal studies rarely follow these rules. For ethical and cost reasons, researchers try to use as few animals as possible, which can mean minuscule sample sizes. Unblinded, unrandomized studies are the norm. In Dirnagl's words, "the way we do our research with our animals is stone-age."

    From various quarters, there's pressure to change that. High-profile studies showing that preclinical results often cannot be reproduced are driving funders and researchers to seek solutions—as much to mend their public image as to guarantee sound science.

    The roots of bias

    Dirnagl's concerns were sparked around the same time as a friend and colleague's across the English Channel. A decade ago, Malcolm Macleod, a Scottish neurologist at the University of Edinburgh, went hunting for new stroke therapies. He wanted to find compounds that had looked good in animals but had stalled there and that might be worth testing in people.

    Macleod and his colleagues identified 603 drugs tested in animals, 374 of which had helped heal the brain. Of those, 97 had been tried in humans—and only one had worked. And that one, Macleod is quick to point out, wasn't tested because of animal data at all, but because it had already benefitted patients with heart attacks.

    Startled by this chasm separating experimental animals and people, Macleod turned his attention to what was going wrong. One possibility, he reasoned, was that the therapy wasn't tested properly in humans—say the dose was too low, or it was given too long after a stroke. Another was that human testing had been appropriate, but the animals were simply a poor model of human stroke. And the third was that the drug wasn't tested properly in animals to begin with.

    Macleod dug deeper. What he found alarmed him. Only 36% of the animal studies described randomly assigning animals to stroke treatment or placebo. Only 29% reported blinding. What's more, studies that didn't report randomizing and blinding—which was most of them—"gave substantially and significantly higher estimates of how good these drugs were," Macleod says. In one case, the effectiveness of a stroke drug was twice as high in the studies that didn't report randomizing as in those that did.

    Macleod then turned to other neurological ailments: Alzheimer's, multiple sclerosis, Huntington's disease, Parkinson's, and pain. In animal studies of potential treatments, the situation was, if anything, worse than in stroke, the measures that might dampen bias applied even less often.

    Many of these authors likely didn't recognize what Macleod perceived as lack of rigor in their studies because their mentors, and their mentors' mentors, had followed this same approach. "I was trained as an animal researcher," says Lisa Bero, now a health policy expert at the University of California, San Francisco. "Their idea of randomization is, you stick your hand in the cage and whichever one comes up to you, you grab. That is not a random way to select an animal." Some animals might be fearful, or biters, or they might just be curled up in the corner, asleep. None will be chosen. And there, bias begins.

    Macleod's work, published in a series of papers beginning in 2004, is complemented by other strands of evidence. A 2008 paper in the journal Amyotrophic Lateral Sclerosis described efforts by the nonprofit ALS Therapy Development Institute to retest more than 70 compounds that had eased symptoms in a mouse model of the disease. Not a single one panned out.

    It was what the ALS authors did next that was particularly interesting. At the end of their dismal replication efforts, they were left with a treasure trove of data on 2241 control animals—mice that hadn't gotten any active drug. The researchers randomly assigned mice to two groups, matched for sex, litter size, and other variables. Then they looked for differences in mean life expectancy—something they shouldn't see, because the two groups were essentially the same.

    What they found was telling. If the two groups contained just four animals each, there was a 30% chance that an illusory life expectancy gap would show up. With 10 animals per group, the risk dropped to 10%. "You can imagine 10 labs doing this experiment," says Shai Silberberg, a program director at the National Institute of Neurological Disorders and Stroke (NINDS) in Bethesda, Maryland. "One gets an effect, and they publish it." The other nine are much less likely to submit a paper. Suddenly, the literature is skewed.

    The numbers of animals that the ALS researchers used may sound small, but they're grounded in reality. A survey of 76 influential animal studies found that half used five or fewer animals per group.

    Bero recently examined animal research of statins for heart disease. At the International Congress on Peer Review and Biomedical Publication in September, she reported that work funded by industry was less likely to endorse the drug in question than work from another funding source, maybe because companies don't want to pour millions of dollars into testing a treatment in people that's unlikely to help them.

    Status quo revisited

    In Bethesda, Silberberg sits in a position of power, part of a committee advising the NINDS director on which of the most costly studies should be considered. About 3 years ago, Silberberg, who trained as a biophysicist in Israel and later the United States, grew more and more worried that the institute was greenlighting some projects that weren't based on solid science. He decided to do something about that.

    There were lots of avenues Silberberg could have followed, and he settled on animals. In part, he was responding to data like Macleod's, with its startling evidence of what he saw as entrenched biases in animal research. A slice of NINDS's budget is funneled to translating animal studies to people. Among other things, Silberberg worried about "poor patients [who] are exposed to things they shouldn't be."

    After lots of agitating and conversation within the National Institutes of Health (NIH), in the summer of 2012 Silberberg and some allies went outside it, convening a workshop in downtown Washington, D.C. Among the attendees were journal editors, whom he considers critical to raising standards of animal research. "Initially there was a lot of finger-pointing," he says. "The editors are responsible, the reviewers are responsible, funding agencies are responsible. At the end of the day we said, 'Look, it's everyone's responsibility, can we agree on some core set of issues that need to be reported' " in animal research?

    In the months since then, there's been measurable progress. The scrutiny of animal studies is one piece of an NIH effort to improve openness and reproducibility in all the science it funds. Several institutes are beginning to pilot new approaches to grant review. For an application based on animal results, this might mean requiring that the previous work describe whether blinding, randomization, and calculations about sample size were considered to minimize the risk of bias. "Sometimes the fundamentals get pushed aside—the basics of experimental design, the basics of statistics," says Lawrence Tabak, principal deputy director of NIH, who is coordinating these efforts.

    Another of NIH's ventures is at the National Institute of Environmental Health Sciences in Research Triangle Park, North Carolina, where toxicologist Kristina Thayer is looking for a way to grade animal studies, in part to guide regulators making recommendations about particular chemicals. For work that examines the hazards of bisphenol A, a compound found in many plastics, Thayer is experimenting with 15 "risk of bias" questions. Among them: Did the researchers randomly allocate animals to treatment groups? Did they know which animals were exposed to chemicals? Were experimental conditions the same across different groups of animals? "When you're looking at bias, it's not just yes or no," she says. "There can be different shades of gray, and there can be scientific judgment in there." The Environmental Protection Agency is also reconsidering how it evaluates animal data.

    Journals, too, are getting in on the act. In April, Nature released a checklist for authors and reviewers, requesting extra detail about scientific methods in life sciences papers. Among other things, the checklist asks whether the animals were randomized and the researchers blinded, and requests the criteria by which animals were dropped from the study—an effort to avoid the three missing mice Dirnagl encountered. Science Translational Medicine announced a similar initiative in June, and Science is considering the same.

    Some in the field consider such requirements uncalled for. "I am not pessimistic enough to believe that the entire scientific community is obfuscating results, or that there's a systematic bias," says Joseph Bass, who studies mouse models of obesity and diabetes at Northwestern University in Chicago, Illinois. Although Bass agrees that mouse studies often aren't reproducible—a problem he takes seriously—he believes that's not primarily because of statistics. Rather, he suggests the reasons vary by field, even by experiment. For example, results in Bass's area, metabolism, can be affected by temperature, to which animals are acutely sensitive. They can also be skewed if a genetic manipulation causes a side effect late in life, and researchers try to use older mice to replicate an effect observed in young animals. Applying blanket requirements across all of animal research, he argues, isn't realistic.

    Bass is just as concerned about the undercurrent that scientists aren't to be trusted. "A lot of what this argument is, is that there's this ethical flaw across the community, and we're going to correct it by mandating these laws," he says.

    Dirnagl agrees with this last point, even though he believes that new standards are needed. "A lot of the academic researchers, they are being accused of producing crap, complete crap," he says. "I think this is overshooting it, and it's even dangerous. … We need to properly discuss these quality issues." More importantly, "we need to teach them to the next generation." He tries to present his case with optimism, so as not to discourage or alienate his colleagues.

    Dirnagl also says he's cleaned up his own act, something that, for the most part, hasn't been particularly onerous. He marks the tails of all his animals with numbers and uses a number generator that spits out a list to help him randomly select mice. If during a surgery an animal's blood pressure drops below a certain level, Dirnagl excludes it, whether it's getting a new stroke treatment or not. He's starting to do what clinical trialists have done for years—run multicenter studies, where labs pool their animals to boost the experiment's reliability with greater numbers.

    One open question is whether such adjustments will help animal experiments hold up to scrutiny. "It's almost certain that we're not completely right" about what's worth changing and what's not, Macleod says, and that will need to be gauged over time. Ultimately, though, he believes better research standards will lead to a renewed trust in mouse models of disease. "I wouldn't be wasting all my time" on this, he says, if he didn't have faith that the mice had it in them to be auspicious guides—if only we could figure out the best way to use them.

  5. Missing the Mark

    1. Eliot Marshall

    The U.S. intercontinental missile defense system faces a crisis as Congress presses for an expansion, interceptors malfunction, and a basic targeting problem remains unsolved.

    Ground-based defense.

    U.S. interceptors (far left) are designed to hit an incoming warhead in space. Experts worry they won't pick out a real warhead from decoys.


    On the morning of 5 July, a long-range missile lifted off from the Marshall Islands in the Pacific and traced an arc into the northern sky. Five minutes later, a U.S. interceptor missile took off from California, 7500 kilometers away, and streaked south toward an encounter far above the atmosphere. When in range, the California-based missile was supposed to release a "kill vehicle" from its nose that would pick out a collision point, maneuver at a closing speed of more than 10 kilometers a second, and slam into the target. That didn't happen. The kill vehicle failed to detach—and missed.

    Similar tests of the U.S. ground-based missile defense system have been under way since 1999, more than one a year on average. According to the Pentagon's tally, eight have succeeded and eight have failed. But in recent years, the rate of tests has slowed—there were only three intercept attempts from 2009 to 2013, and all three were failures. The program's distress has given credibility to skeptics and alarmed some members of Congress, who have asked whether the system can be fixed.


    Ambition has always outrun technological reality in the U.S. missile defense system, a legacy of the "Star Wars" push that began 30 years ago for a shield against Soviet attack and is now seen as a defense against a limited or rogue missile strike. In the next 5 weeks, the United States will begin planning an expansion of its ground-based interceptor network, which now has two rocket-launching bases, in California and Alaska. Under pressure from Congress and private groups worried about the threat of a missile attack from Iran, the Pentagon will pick a location east of the Mississippi and two alternatives to study as a new base for the 18.3-meter interceptors.

    Far from welcoming a chance to expand their missile defense system, however, military officials worry about the cost of an Alaska-style base in the east—possibly as much as $3.5 billion to build, almost half the $7.7 billion annual budget of the U.S. Missile Defense Agency (MDA). They also worry about the distraction from a pressing problem, unsolved for 50 years: How do you create a system that can reliably single out and hit a target in space?

    Expanded missile defense.

    A U.S. National Academies study argues that adding a new interceptor base in New York would add to the reach of those based at Fort Greely, Alaska, protecting a wider area with a single shot against warheads from Iran (inside the blue line) and giving time for at least one follow-up shot (within red line).


    Navy Vice Admiral James Syring, MDA's director, has told Congress that the military doesn't have a "validated" need for a third base. He and Secretary of Defense Chuck Hagel have pointed to other items that do need urgent action. One is to learn why the 5 July test failed and work out what a repair will cost. The other is an earlier problem identified after a December 2010 failure of a ground-based interceptor. Until it is fixed, Hagel has said, the military won't acquire 14 new interceptors he promised to buy in March to expand western defenses (Science, 29 March, p. 1508). At the moment, the system is in limbo.

    But even a string of successful single-target interceptions will not address the system's deeper flaw. In the near vacuum of space, debris and decoys can fly in step with an incoming ballistic warhead, confusing sensors and complicating the job of identifying and destroying the target. Expert reviews have warned repeatedly that target discrimination in space has not been solved or even studied adequately. In testimony to Congress, Syring has identified it as a top priority. But it's not clear that a push to solve this decades-old conundrum can win funding in competition with the cost of fixes, the $1 billion for Hagel's new interceptors, and calls for a new base.

    "The budget implications of the current situation are enormous for MDA," says physicist Philip Coyle, an outspoken critic of the program who was director of operational test and evaluation for the Department of Defense from 1994 to 2001 and former scientific security adviser for the Obama White House. "Given the long list of repairs and upgrades that appear to be needed," he says, "some people think it is a lost cause."

    Get a little closer

    Even though Congress cut U.S. defense spending overall last year and may do so again this year, it instructed MDA in the 2013 defense authorization bill to start scoping out an East Coast missile base site. The main support for this plan comes from the House of Representatives, where the Republican majority voted to begin construction. But some eastern lawmakers of both parties also relish the federal investment and the new jobs. New York Senator Chuck Schumer, a Democrat, put out a press release in May promoting two upstate New York sites, saying base construction there "could create thousands of jobs and significant revenue in local communities." He added, however, that it was up to the military to decide if a base was needed. The Senate did not support construction, but agreed last year that candidate sites should be chosen in 2013 and an environmental impact study started as soon as possible.

    MDA duly issued a list of five potential sites in September; one of Schumer's was on it, along with others with congressional backing in Maine, Vermont, Ohio, and Michigan. By 31 December, the list will be narrowed to three finalists.

    What's to be gained by locating a new base in the east? The short answer, proponents say, is that you get more time to shoot at missiles coming from the Middle East and a chance for better targeting. Basing interceptors closer to Iran, the presumed aggressor in current scenarios, shortens the flight time between the takeoff point in the United States and the earliest contact with a warhead. (The base would add firepower, too—possibly 20 more interceptors.) In theory, the timesaving would allow for an efficient "shoot-look-shoot" tactic. After a first interception attempt, radar and other sensors would look to see whether the kill vehicle had hit the target and, if not, call for another. This could improve the aim of successive shots and avoid waste (at a stiff $75 million per interceptor). Because the California–Alaska system doesn't allow time for a second look at a warhead coming from the east, some say the current strategy is "shoot-shoot-shoot"—send four or more interceptors at each warhead.

    Champions of an expanded ground-based defense note that U.S. intelligence reviews predict Iran will have a long-range missile by 2015. "We do ourselves a great disservice by having half the country [the east] completely exposed to a potential missile attack," says Jeff Kueter, president of a group that favors putting weapons in space, the George C. Marshall Institute in Washington, D.C. Kueter sees no choice to be made between paying for better interceptors and building a new base: "You need to do both," he says.

    A year ago, the U.S. National Academies' National Research Council (NRC) argued for an eastern interceptor base in a report titled Making Sense of Ballistic Missile Defense. The chief co-authors of its classified and unclassifed sections were missile builder L. David Montague, a former president of the Missile Systems Division of Lockheed Martin, now a consultant in Menlo Park, California; and Washington, D.C., attorney Walter Slocombe, a former defense adviser for several administrations. Their panel concluded that the East Coast now is "poorly protected," and their long list of improvements included a new base in New York or Maine. "The key is, you must have shoot-look-shoot capability to increase the size of the battle space" and boost the chance of success, Montague says.

    A moving target

    The new base is needed, the panel said, because midflight interception is the only workable option for defending against intercontinental missiles. In principle, such a missile would be easiest to target and knock out early in its flight, when it is big, slow-moving, and hot. But over the years, schemes for such "boost phase" counterstrikes, including space-based lasers that could punch a hole in a missile's thin skin or fast interceptor rockets based close to a hostile nation, have proved impractical. Intercepting an intercontinental ballistic missile as it arcs through space is the best option left, even though it has its own enormous challenges: targeting and hitting a tiny warhead, separated from its booster, moving faster than a rifle bullet, and potentially surrounded by a cloud of debris or decoys.

    Syring has said that his first task is to fix the existing U.S. missile defense system so that it can pass a simple intercept test. He told Congress that a "production quality" flaw in an interceptor caused the first of the three recent failures and a guidance "design issue" caused the second. No report has yet come out on the third.

    The next priority, Syring has said in several hearings, is to solve the puzzle of target discrimination. Syring declined a request for an interview, but outside experts have described the problem using unclassified data. Physicist George Lewis of Cornell University was a co-author of an analysis of the challenge posed by decoys, published by the Union of Concerned Scientists in 2000. He notes that a network of satellite-borne infrared sensors and long-range warning radars give an interceptor its first, approximate heading. The early warning radars, operating at 400 to 500 megahertz, aren't capable of pinpointing the warhead by shape or size. They can pick out individual objects from thousands of kilometers away but cannot discriminate among objects less than about 10 to 20 meters long.

    High-frequency radars in the 10 gigahertz range, known as X-band, can resolve objects as small as 25 centimeters long, Lewis says, but most existing ones have less than intercontinental range. An exception is a huge system operated from Hawaii, called the Sea-Based X-Band Radar or SBX system, which has observed U.S. tests and missile launches from North Korea. The complex, power-hungry system is costly to use, however, and MDA keeps it idle most of the time, Lewis says. But he predicts that the military will soon get a green light to build a large X-band system on the East Coast, whether or not a new launch base is approved.

    No matter how good the tracking information from these external systems is, the kill vehicle must find its own final path to the target using onboard sensors. The device must pick out a warhead roughly 2 meters long, which may be embedded in a kilometers-wide field of other objects, and slam into it. The latest U.S. kill vehicle, known as the CE-II version, has never achieved this. The older version, CE-I, worked in carefully scripted early tests but now is in doubt, having failed in July.

    The NRC report says that the targeting problem can be solved. It recommends that the kill vehicle be capable of reporting back what its own sensors see on the final approach just before a hit or miss. This would provide more precise, close-in views of objects in the "threat package" and sharpen the aim of the next shot, if needed. It would help research, too. The CE-I and CE-II kill vehicles, however, lack a report-back capability, Lewis says, "and apparently that can't be fixed." They must be redesigned, flight-tested, and replaced to achieve this, which would take years.

    The NRC report also recommends extensive improvements in ground- and sea-based radar. For example, it suggests building five new powerful X-band radars by stacking two copies of an existing Army-Navy radar, AN/TPY-2, and locating them at strategic points in the United States and Europe. This plan also includes an entirely new set of more agile ground-based interceptor launchers.

    Others are skeptical that these expensive fixes would suffice, citing many ways in which targeting could still be thrown off. Lewis and Theodore Postol, a physicist at the Massachusetts Institute of Technology in Cambridge, summarized some of them most recently in a 2010 article for the Arms Control Today newsletter. Tracking radars could be flooded with signals, they wrote, "[i]f an adversary deployed thousands of wires on slightly different trajectories along with warheads." Lewis and Postol cite other tricks: A warhead could be enclosed in a reflective balloon to mask its shape; lightweight fake warheads could be released; and the upper stage of the launch vehicle could be cut into bits with explosives to add to the muddle. More powerful radars don't necessarily help, Lewis argues: "The radar isn't going to be able to tell you which balloon has the warhead in it. … They're all going to just look like balloons."

    IBM physicist Richard Garwin, a defense analyst and arms control advocate who has followed this issue for years and contributed to classified analyses of missile defenses (which he doesn't discuss), also challenges some conclusions of the NRC report. In a letter to its authors that he made public this year, Garwin says that he's not convinced they have a solution to the targeting problem, because "[n]either the unclassified nor the classified Report describes how this could be done."

    Not so easy

    NRC panel co-chair Montague says it's "not new news" that U.S. national missile defense is "flawed." But he thinks critics overstate the ease of deploying decoys. He regards the Union of Concerned Scientists analysis and others as "scholarly and naive," because "they base their assumptions on first principles, and there's a lot more than first principles going on." Take balloons. They're not so lightweight, Montague says. He's not sure how you wrap a warhead in a balloon or fix the package to a missile without an attaching part that would become a giveaway "tag" or signature that radars and other sensors could detect. "I've been involved in building more missiles than the rest of the free world combined," he says, "and I know that you cannot make gossamer [decoys or masks] that work."

    MDA officials have cited a few broad themes in their plan to address target discrimination, but declined requests to discuss them. Echoing a recommendation of the NRC panel, they have spoken about combining radar and infrared data from existing sensors to profile and catalog types of real and fake warheads. In testimony, Syring has said MDA is also experimenting with new infrared and optical sensors that might be placed on high-altitude drone aircraft to provide additional data for sorting out targets and decoys. And MDA is considering a common kill vehicle with new sensors to replace the current 1990s technology aboard both large and small interceptors, Syring has testified. But that's a long way off.

    Even if a fix is feasible, it may not be affordable. MDA's tab for analyzing this year's intercept failure, making repairs, and retesting the system could add hundreds of millions of dollars to costs that are already zooming, according to David Best and Cristina Chaplain of the Government Accountability Office. GAO reported in May that unplanned expenses associated with the redesign and testing of the CE-II kill vehicle after a failure in 2010 amount to more than $1.2 billion—and are still growing. Now, MDA is bracing to hear what it will cost to recover from the CE-I failure.

    If Congress doesn't reach an agreement soon on a 2014 budget, automatic cuts enacted in the sequester law could kick in, potentially taking $700 million out of MDA's budget. Congress could exempt MDA from the cuts, but only by sacrificing other programs. Syring faces some tough decisions in the next year. But he has made it clear that building a new, multibillion-dollar launch site is not something he wants to add to the must-do list.

  6. Missile Defense Made Practical

    1. Eliot Marshall

    When it comes to hitting targets, smaller missile defense systems appear to be far more reliable than the national system.

    Good aim.

    The Navy's Aegis system, using specialized radar and a small SM-3 missile, has intercepted targets in 28 of 34 tests.


    Small is beautiful in missile defense, to judge by recent testing results. In a matchup between a mobile, sea-based system called Aegis, built by the U.S. Navy to knock down short- to medium-range missiles, and the massive backbone of long-range missile defense in Alaska and California (see main story, p. 926), the little Navy one comes out ahead. The Army's mobile interceptor unit called THAAD (Terminal High Altitude Area Defense), designed to protect a relatively small area from medium-range missiles, has also scored well. Weapons analysts advise, however, that none of these systems has ever intercepted an intercontinental-range ballistic warhead, which travels faster than shorter range missiles and may be camouflaged by decoys.

    Aegis went through its paces most recently on 4 October near Hawaii, its 34th test since 2001, according to the U.S. Missile Defense Agency (MDA). An Aegis interceptor launched at sea released a kill vehicle that maneuvered toward a medium-range missile target in space and destroyed it. For Aegis, this was the fifth success in a row and the 28th overall. The system, guided by its own radar, is now destined for expanded use ashore in Romania and Poland, where it will serve as a forward sentinel and potential line of defense against missiles fired at Europe. THAAD, for its part, has had 11 hits in 11 trials since 2006, according to the military, and its radars are or will be deployed in Hawaii, Guam, Turkey, Israel, and Japan. In contrast, the big Alaska–California system—the "ground-based midcourse ballistic missile defense"—has failed as often as it has succeeded in tests.

    At an appropriations subcommittee hearing on 17 July, Senator Richard Durbin (D–IL) praised Aegis as "reliable" and "proven," and MDA's director Navy Vice Admiral James Syring called it "fantastic." Asked to explain why the Navy system performs so well, Syring said that it was developed slowly with a "systematic, system-engineered approach," whereas the large land-based interceptors were "fielded very quickly for a growing threat" on the assumption that they would be improved later. Syring defended both approaches. But the Government Accountability Office (GAO), Congress's audit agency, has warned many times that to meet a 2004 deadline set by the White House, builders bypassed quality control checkpoints for the big system—allowing "concurrencies" in development and deployment. These, according to an April report by GAO official Cristina Chaplain, caused delays, late failures, and expensive redesigns that have yet to be fully tested.

    An enthusiastic supporter of the Aegis system, which is now aboard 27 ships, asked Syring at the July hearing if the Navy system could be stationed off the U.S. East Coast to beef up the ailing land-based system. Syring responded that the answer is classified. Independent experts say that while more Aegis ships would be useful, the system wasn't designed to hit long-range missiles, although it may be modified in an effort to enable it to do so in the future. For the next 5 years, at a minimum, U.S. national defense will rely heavily on the troubled interceptors based in Alaska and California.