Washington

The National Science Foundation (NSF) should build up environmental research in the United States by backing more multidisciplinary work and putting more resources into infrastructure and the training of young scientists, says an advisory committee.

The panel's 10-year-plan for environmental science at the NSF echoes earlier reports from both the National Academy of Sciences and the National Science Board (NSB) that governs the foundation. But the committee, which was chaired by Stephanie Pfirman, an environmental scientist at Barnard College in New York, failed to endorse a call by the NSB for a billion-dollar increase in annual environmental research spending by the NSF.

In 1999, the science board recommended tripling the NSF's $600-million environmental research budget within five years. But since then, the programme has expanded only gradually, to $829 million last year.

An academy report in 2000 identified eight “grand challenges” for environmental science, from understanding biodiversity to improving hydrologic forecasting. The advisory panel review, explains Margaret Leinen, an oceanographer who heads the NSF's geosciences directorate and coordinates its environmental research, was commissioned to recommend what part of that portfolio the NSF should take on between now and 2012.

The review says that the NSF should focus on three areas: coupled human and natural systems, coupled biological and physical systems, and people and technology. Leinen says that the breadth of these categories reflects the need for a multidisciplinary approach — and that the NSF's support for environmental research will cut sharply across the agency's traditional, discipline-based organization. Study of the environmental impact of urban development, for example, may well involve both the geosciences and the social and behavioural sciences directorates.

Such collaboration is feasible in the short term in some areas, such as building computer systems to mine environmental data, Leinen says. But other areas, she adds, including the interaction of environmental and social sciences, may take longer to develop.