Assessing Science's Part In a Good 'Earth' Policy

AS part of the Contract With America, House Republicans have passed legislation that would require the federal government to assess the risk of ''major'' environmental rules before they are applied. This bill, which has yet to pass in the Senate, is controversial. Why should risk assessment, ostensibly a science-based process, generate so much debate?

The main reason is that risk assessment is not a purely scientific exercise, because it blends science with policy choices. A recent study funded by the Department of Energy, ''Choices in Risk Assessment,'' concluded that ''risk assessments can be designed and biased to achieve predetermined regulatory outcomes and objectives.'' Politics will always play a role in environmental policy. But science should lay the groundwork for policy choices; it should not be distorted to justify political decisions.

Another problem is that most risk assessments are conducted with incomplete and inconclusive information. Policymakers must use partial data, create exposure scenarios, and rely on overly simplified models to determine risk. When it comes to ecological issues, such as threats to wetlands, the scientific data tends to be even more difficult to obtain and evaluate.

To base risk assessments more on science than on politics, we need an objective, reliable, accessible source of quality environmental science. We should shield this source from political influence. That's the challenge facing Congress, the executive branch, and the scientific community -- and it goes a lot deeper than risk assessment.

The simplest solution might be to leave environmental research where it is, and try to establish a ''firewall'' to protect it from politics. That may be difficult to do inside regulatory and resource management agencies like the Environmental Protection Agency (EPA) and the United States Forest Service, where science is largely a response to past congressional directives. Science needs to be in the vanguard of environmental policy, not the rear guard.

Three former administrators of the EPA have backed a proposal to create a National Institute for the Environment (NIE), an independent, nonregulatory institute charged with improving the scientific basis for environmental decisions. The NIE would integrate scientific assessments of environmental issues with research, information services, and education and training, while staying clear of political and regulatory agendas. Rather than conducting risk assessments, a nonregulatory NIE would review risk-assessment methods, improve them, communicate them clearly to the public, and provide much-needed peer-reviewed data on which to base risk assessments. The result is not policy-free risk assessment, but far less room for those who would disguise policy as science.

Some have suggested combining all or parts of existing agencies like EPA, NASA, the National Institute for Environmental Health Sciences, and the National Oceanographic and Atmospheric Administration, to form a new environmental science agency. This would ostensibly bring cohesion and independence to environ- mental science while remaining budget-neutral.

Others are concerned that, under this approach, the credibility of the science would be undermined by inherited regulatory agendas and fragmented missions. Trimming budgets and reallocating the resources to a new independent institute with an exclusively scientific mission may ultimately be easier and more effective than reshuffling existing programs.

As we approach the 25th anniversary of Earth Day, America's environmental and political leaders stand at a crossroads. Few will argue with basing environmental policy on science. But, simply mandating more risk assessment will not solve our problems if it continues to be based on inadequate, politicized science.

Given the Republicans' legislative drive and the Clinton administration's effort to reinventing government, the nation's environmental science agenda might be the perfect prototype for bipartisan progress.

Congress and the administration have been telling us they can cooperate when they share common objectives. Here's a chance to prove it.