April 2001

From University of Washington

U.S. needs major steps to overtake European climate research, UW scientist says

The United States seriously lags behind England and Germany when it comes to computer-driven climate research, and a University of Washington scientist says it is time to take dramatic steps toward leadership in the field.

"I find it extraordinary that England does more focused and more extensive climate modeling than the United States does. That shouldn't be," said Edward Sarachik, a UW atmospheric sciences professor who headed a National Research Council panel that recently examined the issue.

Although the United States spends $1.8 billion a year on climate research, only 6 percent goes to modeling that can provide a much clearer picture of long-term changes, Sarachik said. England, on the other hand, has focused its spending, with $50 million for the European Center for Medium Range Weather Forecasting and another $25 million for Hadley Centre for Climate Prediction and Research. The centers use data and small-scale research generated in the United States to synthesize large-scale climate information.

The lack of coordinated climate research and observations in the United States could have serious ripple effects nationally. Farmers and power generators, for instance, won't have the information to make long-range plans and protect themselves and consumers from the whims of climate variability and change, Sarachik said.

Besides its own recommendations, the NRC Panel on Improving the Effectiveness of U.S. Climate Modeling endorsed previous NRC reports that called for creation of a national climate service to conduct climate observations and provide some forecasts, much as the National Weather Service does for weather. On April 30, Sarachik, an expert on the interaction between atmosphere and ocean, will brief U.S. Senate Energy Committee staffers on the NRC panel's recommendations.

Among problems that must be solved, the panel found, is an exodus of skilled computer programmers and technicians from the research community to the private sector and an inability, because of government regulations, for research groups to buy the best available computer technology for climate research. Many climate scientists believe that, because of superior processor speeds, the best supercomputers for such work are Japanese models, widely used in England and Germany, that cannot be purchased by U.S. agencies because of anti-dumping tariffs. "We should be able to buy the best equipment in the world that suits our needs. If we can't, we're a third-world country," Sarachik said.

Private companies, he added, get around the problem by maintaining offices in foreign lands where they can buy those computers.

Other recommendations by the NRC panel include:
* Focusing adequate resources on centralized activities aimed at short-term climate forecasting, long-term climate change, the human role in climate change, global assessments of atmospheric ozone, and assessing regional impacts of climate change.
* Improving research access to high-end computing facilities that are linked to the centralized operations of collecting climate data.
* Interlinking existing climate modeling efforts, and using a common modeling and data framework for easy communications among the modelers and the research community.
* Conducting research on social and economic aspects of climate and climate modeling, so institutions and governments get the information they need to provide effective climate services.

Computing technology and expertise are critical for climate modeling research, largely because of the mountains of research data and the complex calculations that are required. In Europe, those resources have been parlayed into large, centralized climate centers in Germany and England.

"This work is extraordinarily computer intensive," Sarachik said. "In the United States, our top two centers together don't amount to one-fifth of the European effort. We have a lot of small research centers, but we don't have the critical mass."

As part of its recommendations, the panel wants experts on government organization to find the best way to design and implement a government climate agency with both research and observation functions. The federal government was last reorganized in 1970, Sarachik said, a time when little thought was given to long-term climate research. Now there should be a centralized system that collects climate data, produces climate information and coordinates research efforts so general information about climate is more accurate, Sarachik said. That would lend to greater trust of scientific conclusions about climate variability and change.

"We should do for future generations what we would have liked previous generations to have done for us," Sarachik said.

For more information, contact Sarachik at (206) 543-6720 or sarachik@atmos.washington.edu

The report is available on the Internet at http://www.nap.edu/books/0309072573/html

This article comes from Science Blog. Copyright 2004

Archives 2001 E