The National Science Foundation's push for a budget boost could help the United States pull ahead in the race for breakthroughs in research.

Marking its 50th anniversary this year, NSF supports about half of all nonmedical basic research at universities and colleges across the country, as well as extensive science and mathematics education programs. The research efforts range from manipulating individual atoms and molecules to probing the origins of the universe; from analyzing cellular metabolism to computer modeling of complex ecosystems; from campus laboratories to field studies in jungles, deserts, the oceans and the vast frozen expanses of Antarctica.

NSF, an independent agency that has been a budgetary favorite of the Clinton administration, sought $4.6 billion for fiscal 2001-a 17 percent increase over the previous year-to pay for several major initiatives as well as expanded funding of core disciplinary research. After congressional action, the agency ended up with an appropriation of $4.4 billion for fiscal 2001, an increase of $529 million or 13.6 percent. In August 1998, Colwell, a marine microbiologist, became the first woman to lead NSF, after physicist Neal Lane went to the White House Office of Science and Technology Policy to serve as President Clinton's science adviser. One of Colwell's chief goals is to at least double the agency's budget within the next five years. "NSF is only about 3.5 percent of the total government budget, which is much, much too small," Colwell says. "The NSF budget ought to be somewhere between $10 billion, $12 billion, $13 billion for a nation that has a federal budget of $1.8 trillion and an economy of $9.6 trillion." Substantially increased investment in research, Colwell contends, is "absolutely critical" if the United States is to maintain world leadership and preserve its free society, economic prosperity and national security.

Colwell's advocacy of significantly expanded government spending on fundamental research is echoed by Eamon M. Kelly, an economist and president emeritus of Tulane University who chairs the National Science Board, a 24-member group overseeing NSF. Kelly says his "overriding concern" is U.S. under-investment in this area. "We now spend substantially less than 3 percent of our gross domestic product on research and development, and an infinitesimal percentage in terms of basic research," he says, adding that other countries are now starting to devote a much higher proportion of their GDP to research.

NSF, whose headquarters and 1,150 employees are based in Arlington, Va., is generally held in high regard within the U.S. research community and among Washington policymakers and interest groups. The agency usually receives broad bipartisan support on Capitol Hill. However, NSF's operations have been marred by difficulties with its "FastLane" system for computerized processing of grant proposals and other research project documentation.

The NSF's origins date back to the closing days of World War II. In July 1945, Vannevar Bush, director of the wartime Office of Scientific Research and Development, sent President Harry S. Truman a report, "Science-the Endless Frontier," stressing the importance of government support for science. He wrote that "without scientific progress no amount of achievement in other directions can ensure our health, prosperity, and security as a nation in the modern world." Focusing on the vital role of fundamental research in leading to new applications of science, Bush observed, "Basic research leads to new knowledge. It provides scientific capital. It creates the fund from which the practical applications of knowledge must be drawn. New products and new processes do not appear full-blown. They are founded on new principles and new conceptions, which in turn are painstakingly developed by research in the purest realms of science." Truman signed legislation creating the National Science Foundation in May 1950.

NSF started out dealing almost entirely with the physical sciences, and most of its directors have come from those disciplines. But the agency's portfolio has gradually expanded to include the biological sciences, education, engineering, and the social and behavioral sciences. In fiscal 2001, NSF plans to devote $837 million to four initiatives, ramping up substantially from this year's efforts. These areas are information technology research, nanoscale science and engineering, biocomplexity in the environment and education of the 21st century workforce.

Computers on the Head of a Pin

NSF is the lead agency for a federal IT research and development program that involves nine other agencies, including the Defense Advanced Research Projects Agency, National Security Agency, Energy Department, NASA, National Institutes of Health and National Institute of Standards and Technology. The NSF initiative is budgeted at $327 million for fiscal 2001, a 160 percent increase from this year. Ruzena Bajcsy, NSF assistant director for computer and information science and engineering, says the initiative emphasizes basic research because in computing, the technology is far ahead of scientific understanding-just as the practical craft of metallurgy, for example, long preceded the development of modern materials science. "We build systems-networking systems, software systems, embedded systems. And by the seat of the pants we put them together and tune them and make them work. And they are making a revolution in our society," she explains. "But the scary thing is that we don't fully understand why these systems work. We cannot predict when they will fail. We cannot guarantee their safety. We cannot guarantee their security.

"Here we are talking about systems where the components are in hundreds of thousands and soon to be millions," Bajcsy says. "They are interacting, and we don't understand what the consequences of that interaction will be. In addition to that, we have people as components of these systems. These people are affected by these systems, and they affect the systems. And understanding, modeling people, as we all know, is very difficult. People can be rather unpredictable." Bajcsy says that during the current fiscal year, "we are going to be able to support 10 to 12 percent of proposals that we receive. This is terribly low. We will have to reject another 10 to 12 percent of proposals that were ranked excellent and very good." NSF's initiative in nanoscale science and engineering involves design and construction of ultra-small structures, down to the scale of individual atoms and molecules. NSF is budgeting this initiative at $217 million for fiscal 2001, a 123 percent increase from this year. "Nanoscale science and engineering promise to yield a dominant technology for the 21st century because the control of matter at the nanoscale underpins innovation in critical areas from information technology and medicine to manufacturing and the environment," according to the agency's budget proposal. "Possible future uses of nanotechnology include artificial photosynthesis for clean energy and computer chips capable of storing trillions of bits of information on an area the size of a pinhead." When asked how realistic such forecasts are, NSF Deputy Director Joseph Bordogna-an engineer whose career has included work with laser communications systems and electro-optic recording materials-notes that some parts of computer circuits already have dimensions of about 10 nanometers (a nanometer is one-billionth of a meter). Soon, "we might have DNA molecules doing the computing," he says. "Certainly within the next decade these things are going to happen."

Putting It All Together

NSF's initiative on biocomplexity-budgeted at $136 million in fiscal 2001, an increase of 173 percent from this year-bears the pronounced stamp of Director Colwell, who herself continues to pursue research on cholera, a major scourge in developing countries. Biocomplexity is Colwell's term for an interdisciplinary approach to biodiversity, sustainability and ecosystems studies that puts emphasis on quantitative modeling. The initiative is aimed at understanding the many complex systems that are structured or influenced by living organisms or biological processes. Mary E. Clutter, NSF assistant director for biological sciences, notes that molecular biology generally has used a reductionist approach, while traditional ecology was observational. By contrast, she says, biocomplexity takes an interdisciplinary approach to understand the big questions about living things and systems. It's a systems approach. Clutter, who has been at NSF since 1989, says Colwell is the first director she has known who is "actually a bench scientist." That gives Colwell an appreciation for what it is like to be in the trenches, fighting for funding. "I think Rita is thinking along the right lines in many, many ways-certainly in biocomplexity, which many people professed in the beginning not to understand," Clutter says. "People are understanding [now], because the kind of [research grant] proposals we are getting are terrific."

The 21st century workforce initiative-budgeted at $157 million for fiscal 2001, an increase of 113 percent from this year-seeks to build upon NSF's long-standing programs to support science and math education. Those programs involve about 82,900 K-12 teachers, 11,300 K-12 pupils, 30,000 undergraduates, and 21,400 graduate students. "Many times we embark on activities in the education arena without really having a solid research base for knowing how to move ahead. And that's something that this initiative is in part trying to remedy," observes Judith S. Sunley, NSF assistant director for education and human resources. "So it's emphasizing the importance of research on the science of learning, on understanding how children and adults learn, and how we can facilitate that learning process through education." Another key aspect of the workforce initiative is building partnerships between higher education and K-12 education. "You can have the research-level scientists, mathematicians, engineers involved in content-rich parts of K-12 education-developing materials, helping to train teachers, and things of that sort," Sunley says. Colwell strongly advocates another workforce initiative program that involves placing science and engineering graduate students in K-through-12 classrooms, where they gain a better understanding of teaching and serve as resources for the schoolteachers and pupils.

NSF draws nationwide and global attention for its central role in running U.S. research programs and facilities in Antarctica and the Arctic. In Antarctica, the agency operates three year-round research bases-McMurdo Station on Ross Island, Palmer Station on Anvers Island and Amundsen-Scott Station at the South Pole. Ski-equipped planes, helicopters and ships, including a specially constructed ice-breaking research vessel, support the bases. Arctic facilities include camps and sites for studies of greenhouse gases, monitoring stations for research on ultraviolet radiation, ice coring sites for studies of global climate history, high latitude radar observatories and magnetometers for upper atmospheric research.

In Antarctica, a major effort is under way to rebuild and modernize the South Pole Station. According to Karl A. Erb, director of NSF's Office of Polar Programs, the $150 million effort is on budget and scheduled for completion in 2005. The reconstructed station is designed to accommodate up to 150 researchers and support personnel. Erb notes that Antarctica has become popular for astronomers and astrophysicists in addition to researchers in ecology and geophysics. Astronomers love it because of its extremely dry air, virtually free of image-distorting water vapor, he says. "For certain kinds of astronomy and certain wavelengths, it's the best place anywhere in the world."

FastLane Jammed

Every year NSF receives approximately 30,000 new research grant proposals and funds about 10,000 of them. It also receives more than 85,000 proposal reviews. To help cope with the processing work, the agency started an electronic administration program in 1994 called FastLane. NSF officials contend that the computerized system is being implemented quite successfully.

But some university research administrators say FastLane has been giving them headaches.

"There have been a whole lot of both technical and procedural problems," says Jack W. Lowe, executive vice provost for research at Cornell University in Ithaca, N.Y. "The thing is that FastLane was put up as an experiment originally. The objectives are admirable and we support them completely. The trouble is that before the system was really ready to handle the load . . . they informally made it mandatory for many big deadline proposals. We've had many cases where the system gets swamped." Lowe says that the ongoing problems with FastLane have caused "an enormous amount of grief. There's a feeling out in the [research] community that, yeah, this is nice, but it's a hell of a lot of trouble to deal with it." "We've had a variety of different challenges, and we are attacking each one," responds Linda P. Massaro, NSF's chief information officer. "We've certainly added to our server capacity. And I think we have that particular problem licked. We had earlier this year some Internet service provider problems. . . . What we've done to solve that problem is we're adding a second or backup Internet service provider." Massaro views the FastLane telephone help desk's overly large workload as the biggest remaining challenge to the system. She says the problem is being dealt with by assigning additional personnel during peak periods. "There are still challenges ahead, but I think that we're meeting the tough ones head on and we're in good shape," Massaro says.

Leading at the Edge

Within the research community, Colwell's performance as NSF director draws generally good reviews. D. Allan Bromley, dean of engineering at Yale University and White House science adviser during the Bush administration, says the agency has been doing extremely well under Colwell. "Rita has been very effective in representing the activities of the National Science Foundation in congressional hearings and in general presentations to the scientific and technological community," Bromley says. Marcel C. LaFollette, a research professor at the George Washington University's Center for International Science and Technology Policy, adds, "From the standpoint of not only her symbolic but her real activities, she's actually taken us light years ahead."

A veteran NSF official, who spoke on condition of anonymity, observes that Colwell's management style is much different than that of her low-key, consensus-oriented predecessor, Neal Lane. "Neal was the kind of leader who liked to get a lot of people around the table and discuss where the agency should go and what should be its policies and priorities, thoroughly, with all of the senior managers given an opportunity to lay out what they think and why," he says. "Rita, on the other hand, is much more decisive and more quick to act. She does want advice from people, but on the other hand she's not against making a decision on her own without what seems like a whole lot of discussion amongst the senior folks at NSF, and then moving ahead on it."

The challenge of effectively managing NSF is accentuated by the constant need to keep the agency at the leading edge of science and technology. "The management challenge is to be agilely responsive to change, because the frontier is constantly changing. And we all know it, and it's uncomfortable," says Deputy Director Bordogna, NSF's chief operating officer. "Change is uncomfortable, and it's disruptive at times because a new theory kills ana old theory. But that's what we're all about. The vision is very simple: enable America's future . . . through discovery, learning and innovation," he adds.

Catching a Gravity Wave

NSF's willingness to take risks on major research projects is exemplified the Laser Interferometer Gravitational-Wave Observatory (LIGO). To date no one has ever directly detected and measured gravitational waves, although Albert Einstein predicted their existence in 1918 in his general theory of relativity. Gravitational waves are believed to be ripples in the fabric of space and time produced by violent events in the universe, such as the collision of two black holes or in the cores of supernova explosions. Such waves are emitted by accelerating masses much as accelerating charges produces electromagnetic waves. In 1994, Joseph Taylor and Russell Hulse were awarded the Nobel Prize in physics for their observations on the apparent influence of gravitational waves on a binary pulsar, two neutron stars orbiting each other.

In an effort to directly detect and measure gravitational waves, NSF is spending about $300 million on LIGO, the largest single enterprise ever undertaken by the agency. LIGO consists of two identical installations, one at Livingston, La., and the other at Hanford, Wash. Each installation has a 4-foot diameter vacuum pipe arranged in the shape of an L with 2.5-mile arms. At the vertex of the L and at the end of each of the arms, test masses hang from wires and are fitted with mirrors. Ultrastable laser beams traversing the vacuum pipes measure the effect of gravitational waves on the test masses.

Basic construction of the facilities-designed by a team of scientists from the Massachusetts Institute of Technology and the California Institute of Technology-has been completed, and extremely high-precision laser interferometers are being installed. According to Robert A. Eisenstein, NSF assistant director for mathematics and physical sciences, the plan is to start taking measurement data in the last quarter of 2001. "We're quite pleased," Eisenstein says. "But it's a risk. We don't know what LIGO will see."

Eventually LIGO will be part of an international network of gravitational-wave observatories, with other installations to be constructed near Pisa, Italy, and Hanover, Germany, and in Japan. An NSF fact sheet on LIGO notes that "investing in leading-edge research and education is a future-oriented endeavor, which involves taking risks. Increasingly, it requires international collaborations and integrating knowledge across traditional disciplinary boundaries. And, as science reaches the brink of what is considered nearly impossible measurement, it requires technological innovations that were barely conceived of even a few decades ago."