agenda indicate that this program may follow the precedent set by NAPAP. It is likely to produce a lot of good science but to remain largely irrelevant to the policy decisions that the United States and other nations face over the next decade.

How can a multibillion-dollar research program involving some of the best scientific minds be so predictably irrelevant? The lessons learned from NAPAP and the current plans for R&D on global climate change offer some answers.

What went wrong with NAPAP?

In the early 1970s, a number of U.S. scientists gained national attention by asserting that lakes in Adirondack Park were dying from “acid rain." Pollutants such as sulfur dioxide and nitrogen dioxide were mixing with water vapor in the atmosphere and returning to earth in the form of acidic rain, snow, or fog. A growing number of environmental groups echoed the warning that acid deposition was damaging forests and lakes across the northeastern United States and Canada. Congressional leaders reacted to the public outcry by demanding large reductions in emissions of sulfur oxides and nitrogen oxides from midwestern coalfired power plants. However, the cost of reducing emissions was predicted to be high, leading to steep increases in utility bills for some consumers and severe unemployment in the high-sulfur coal industry. Pointing to the scientific and technical uncertainties surrounding the acid rain debate, the Reagan administration endorsed the Acid Precipitation Act of 1980, creating a 10-year research program-NAPAP.

Congress delayed acid rain legislation for nearly 10 years, in part waiting for NAPAP's results. In 1990, the program's 27 technical reports and three-volume Integrated Assessment were released. By that time. however, Congress had nearly completed action on the 1990 Clean Air Act Amendments, including acid rain emission controls similar to those called for in the early 1980s. Guided primarily by pre-NAPAP studies, Congress decided on a 40 percent reduction in total sulfur dioxide emissions-a decrease of about 10 million tons per year.

Part of what NAPAP reported, however, was that many of the early scientific claims about aquatic damage were exaggerated, that fewer lakes were being acidified than originally thought, and that reducing sulfur dioxide emissions by about 30 percent yielded

48

essentially the same long-term environmental benefits to aquatic systems as the more costly reductions of 40 to 50 percent that had been proposed. Although NAPAP was silent or equivocal about many other effects of acid rain, the tone of the final assessment suggested that acid rain was not the environmental catastrophe widely portrayed a decade earlier. Yet these results never made it into the political debate.

One reason that NAPAP's findings were ignored is that they were not reported in a timely, credible, or understandable manner. NAPAP's first interim assessment, due in 1985, was delayed a full two years, largely because of a change in directors. This delay, combined with accusations by NAPAP scientists that the new director had attempted to downplay the significance of acid rain effects in the interim assessment, significantly damaged NAPAP's credibility. NAPAP's original plan to publish additional interim assessments in 1987 and 1989 gave way to a hurried attempt to redirect the effort and finish on time.

Further, NAPAP's reports were largely unintelligible to Congress. The interim assessment was essentially a four-volume compilation of scientific findings, not the long-awaited, policy-relevant assessment of acid rain damages. The final assessment, which attempted to address some of the earlier shortcomings, received only a cursory congressional committee hearing, shortly before the new Clean Air Act Amendments were passed into law. Again, NAPAP simply failed to communicate what Congress wanted to know.

Nothing was said, for example, about the impacts of acid rain controls on regional coal mining employment or customers' electric bills. Nor were the benefits of acid rain controls—of major interest to Congressreported in ways that Members of Congress could understand. NAPAP's final Integrated Assessment, which was supposed to be policy-oriented, summarized the impact of acid rain on soils by writing that a "reduction of sulfur deposition by 50 percent over 10 years would cause a slight increase in base saturation of some shallow sensitive soils with low cation exchange capacity, but most soils would not be affected." Oh.

A major failure of NAPAP is that no serious effort was made to define policy-related research priorities and to then shape an appropriate set of projects and timetables to answer these questions in a meaningful way. Instead, independent government agencies.

ISSUES IN SCIENCE AND TECHNOLOGY

CLIMATE RESEARCH

driven by different missions, motivations, and expertise, largely pushed their own scientific research agendas. A small policy and assessment program begun during the early years of NAPAP was dis

gram in a move to keep NAPAP science-oriented. Thus, despite the recognition by some NAPAP participants that policy needs were important, the general quest for scientific knowledge rather than the specific policy needs of Congress dominated the thinking and programs of NAPAP.

Major resources, for example, were committed to developing a complex atmospheric transport model that could allow someone to identify the sources of an acid aerosol falling on a particular region of the country at a particular hour of the day. Although this research clearly advanced the state of the science of atmospheric transport modeling and chemistry and resolved some longstanding technical debates among researchers, the substantial resources devoted to this endeavor did little to refine the existing policy options available to Congress.

On the other hand, studies of the effects of nitrogen oxides on forests and soil systems, for instance, were not begun until halfway through the program, with findings only now beginning to emerge. As a result, some scientists contend that NAPAP may have significantly underestimated the emission reductions needed to protect against some of the long-term effects of acid deposition. Similar criticisms have been leveled at NAPAP's failure to adequately study other areas such as effects on materials and human health.

The NAPAP effort was poorly focused for several reasons. First, the Reagan administration put little new money into NAPAP in the early years. Instead, it directed the Environmental Protection Agency and other NAPAP agencies to reprogram existing funds. Many program managers and agencies responded by altering descriptions of existing research projects to emphasize the links to acid deposition, even where the link was flimsy. NAPAP, with no independent budget authority, no strong control over agency agendas, and no real willingness to make waves, turned a blind eye

to this strategy. Thus, few of NAPAP's resources were new or newly targeted on policy-related acid rain research. Not until roughly halfway through the program did NAPAP's funding increase to

address some of the key policy

relevant issues. But that effort

proved too little, too late.

A second key reason for NAPAP's lack of policy relevance is that research was managed and reviewed largely by researchers interested in pursuing scientific research, not applied, policy-related studies. NAPAP scientists

produced a fundamentally different approach to research than was needed for a policy-oriented program. For example, aquatic scientists initially focused on the detailed mechanisms of lake acidification, not on how many lakes were actually acidified or how many would be improved by acid rain controls—issues critical to the policy process.

Third, NAPAP reflected the political ambivalence of its creators. Congress could not agree on what action to take and, so, compromised on a program of research. The Reagan administration wanted to delay action. If policy-relevant research had been performed early in the process, the results might have been used to force legislation that the administration didn't want. The obvious solution was to emphasize scientific uncertainties, call for more scientific research, and criticize findings that did not reflect the best possible science. And who can be opposed to good science?

Thus, despite Congress' explicit intent that NAPAP be policy-oriented, NAPAP grew increasingly detached from the policy process. By the late 1980s, when acid rain again attracted public concern, Congress and environmentalists were determined to resist any further delaying tactics, such as hand-wringing over scientific uncertainties or appeals for better science. An acid rain bill became likely when a new president, having campaigned as a friend of the environment, endorsed acid rain controls and sent a proposal to Congress as part of a new Clean Air Act. Most of the quibbling was over who should pay, how to reduce costs through “market mechanisms," and how quickly to implement controls. The question of the

WINTER 1991-92

extent of needed controls—a quintessentially scientific question was decided with virtually no input from NAPAP.

Following in NAPAP's footsteps

The Bush administration's global climate change research program is “déjà vu all over again." The president has rejected environmentalists' call for immediate action to abate greenhouse-gas emissions on the grounds that we don't yet know enough. Instead, he has initiated a multiyear, multibillion-dollar research program. Although the federal government is expected to fund $1 billion in climate-change research for fiscal year 1992, no more than half of this represents new money for research. To a large extent, program managers have once again repackaged their existing projects to emphasize their relevance to global climate change.

Agencies throughout the government have created plans and research objectives to define the scope of their research on global climate change. Boasts of significant R&D activities on global climate can now be heard from nearly a dozen federal agencies, including the National Aeronautics and Space Administration, the Environmental Protection Agency, the National Science Foundation, and the Departments of Agriculture, Commerce, Energy, and the Interior. Most of these agencies participated in NAPAP. In addition, new interagency groups have been created for global climate research. The Committee on Earth and Environmental Sciences, involving representatives of 18 federal departments and agencies, coordinates the effort.

Most of this research, however, is motivated by scientific inquiry or existing programs, not policy issues. NASA. for example, with two-thirds of the global change research budget, is using space probes to measure properties of the earth's atmosphere to better understand cloud physics and other phenomena affecting the global energy balance. NOAA oceanographers are pursuing studies of deep ocean currents that affect global energy transport. And in several U.S. agencies, atmospheric scientists are developing more refined mathematical models of atmospheric transport and chemistry for application in the general circulation models used to predict the world's climate.

Much of this research will ultimately help to gain a better understanding of how the climate system

works and how human activity affects it. But, as with acid rain, there has been no attempt to ensure that the overall research agenda is addressing the most important problems, sequencing research programs appropriately, using the most cost-effective methods, and avoiding unimportant issues. Most important, there is no individual or agency responsible for seeing that the program produces timely results that are relevant to the most pressing policy issues confronting Congress and the president. A research program that begins to deliver answers 10 or 20 years from now will be of little value to the policy decisions facing us in the

near term.

Over the next several years. for example, key policy choices will require judgments about how serious the impacts of climate change might be, how costly it might be for society to adapt, and how much "insurance" society should buy through reduced emissions. Many of the most pressing issues before Congress require a better understanding of the economic and social costs of taking actions now to avoid the possible future consequences. But so far, few resources have been devoted to studying the impact of global warming on humans or the implications of strategies to mitigate or adapt to possible changes.

It is easy to predict the program's outcome 5 to 10 years from now. Some good research will have been completed, but the program will not have adequately informed the important policy decisions that will have been made, nor will it be producing results in a form to support pending policy decisions. At some point, public pressure is likely to build to the point where either Congress or the White House takes action. But by then, Congress will have found that the climatechange research program could not answer its most pressing questions and will have learned to ignore it. Even if the program's results have significant implications for public policy, they will be found too late to influence the process or will not be communicated effectively or credibly enough to make a difference. In short, after having spent billions of dollars to study global climate change, Congress will make its decisions with little more scientific input than is available today.

Implementing integrated assessments

A billion-dollar research program scattered among a dozen government agencies and hundreds of in

29

ivity affects it. But, as with ■o attempt to ensure that the addressing the most imporng research programs apost cost-effective methods, mt issues. Most important, ency responsible for seeing ces timely results that are ng policy issues confronting mt. A research program that 0 or 20 years from now will cy decisions facing us in the

1

years, for example, key
ire judgments about how
mate change might be, how
ety to adapt, and how much
buy through reduced emis-
pressing issues before Con-
erstanding of the economic
ctions now to avoid the pos-
s. But so far, few resources
dying the impact of global
e implications of strategies
ssible changes.

program's outcome 5 to 10
od research will have been
m will not have adequately
licy decisions that will have
roducing results in a form to
decisions. At some point,
co build to the point where
ite House takes action. But
ave found that the climate-
could not answer its most
I have learned to ignore it.
Its have significant implica-
ey will be found too late to
will not be communicated
ugh to make a difference. In
billions of dollars to study
Congress will make its
cientific input than is avail-

ted assessments program scattered among a cies and hundreds of in

SUES IN SCIENCE AND TECHNOLOGY

The first priority is to articulate the program's objectives and adopt a management plan for achieving them. What precisely is the federal government's global climate change research program to accomplish? And by when? Is the current level of funding adequate to support these goals? Does the current program address all of the issues identified as important? Who is responsible for what research and by what dates? Will the inputs that different research groups need from one another be provided on time and in a usable form?

Once these questions have been answered, a mechanism must be established for reviewing results and reevaluating research priorities. We believe that a series of comprehensive, integrated assessments offers the most effective means of achieving this end. Integrated assessments are needed because the issues are multidisciplinary and highly interactive; multiple assessments are needed because the issues are too contentious and too important to entrust to a single group. Ideally, such assessments should have been started already, before major funding was initiated. But late is still better than never.

The primary functions of an integrated assessment are to survey the state of current knowledge regarding climate change and to reach scientifically informed judgments about what we know and don't know, what the key uncertainties are, and where new research could aid the policy process most effectively. Special attention must be paid to the links among research projects in various disciplines and to the magnitude and relative importance of uncertainties that characterize aspects of the problem. A key task is to determine the value of new information. Thus, the assess

sities, or industry.

ments strive to keep cused upon identifying ing key uncertainties. they change over time.

As they evolve, ments may also focus

lated to research meth

ample, should groun

space-based platform

survey changes in temp concentrations of gases strata in the atmosphe sessments may even compare the relative m ducting certain types through government l

Integrated assessments must also serv between the scientific community and the munity. Policymakers need to understand tions of what science can determine, an must understand what the policy comm needs. By evolving in steps, the assessm can facilitate convergence between what is what is possible, within various time and re straints. The goals at the outset are to a questions, influence the research agenda, sure that appropriate priorities are establi assessments proceed, the goal is to ens technical and often heavily qualified 1 science is translated into terms that policy lay audiences can understand and use to m formed decisions. The need for more eff munication will require concerted attentio change research is to fulfill its goal of infor policy.

Finally, it is reasonable to ask: If integ ments are such a good idea, why hasn started already? Can it really be true that r federal program has done this yet? Cer have been a number of important efforts few years to synthesize current knowl global warming. Recent studies by the Off nology Assessment, the Intergovernmen Climate Change, and the Panel on Policy of Greenhouse Warming of the National A Sciences and Engineering and the Instit icine have been the most comprehensive. ]