The climate change and ozone transport issues share more than a potential physical and enforcement relationship: They also share in being a poor translation of science into the public debate.
Tension between science and public policy is inevitable, especially in a democracy. Fear demands action, not reflection. Too often the two proceed in parallel, so that when science finally finds the answer, the wrong solution is already in place.
Today, public policy is pushing two fear-driven mega-projects that will affect the electric power industry profoundly, with little science to justify them. The fears are climate change and ozone transport. The science is sweeping, from atmospheric chemistry to terrestrial biology, nonlinear dynamics to computer science. But in both cases the science is still vague at best, and quite often contradictory.
The parallels are striking. Does a power plant in Tennessee affect ozone levels in Philadelphia? Does it affect global temperature? There are two basic problems, or one problem at two levels. First, both issues are scientifically intensive, but the public and policy makers cannot possibly grasp the science. They must depend on translators. Enter the press. Second, the science is so broad that no one can be expert in all of it, so the scientists themselves depend on translators. In both cases the translation is failing.
Take climate change. Story after story says the U.S. is the largest source of greenhouse gases in the world. This is misleading, because it ignores natural sources which are far greater than man-made ones. The mass of the atmosphere is around 5,000 trillion tons. Water vapor, by far the most common greenhouse gas, averages 50 trillion tons, carbon dioxide 2 trillion tons. Water vapor is emitted by evaporation in vast quantities, roughly the sum of all the world's rainfall. About 500 billion tons of carbon dioxide are emitted naturally each year, from the ocean and from decay and respiration of plants and animals, especially plants.
Man-made emissions of water vapor are negligible compared to nature's. Emissions of carbon dioxide from all human activities are about 25 billion tons per year, or 0.05 of what nature puts out. Of this the United States puts out about 5 billion tons, or one one-hundredth of natural emissions. If we cut U.S. emissions by 10 percent, which would be very difficult, this would be just 0.001 of nature's emissions, not counting the water vapor.
The fact that human activity influences the climate is often said to be proved by computer models. But the climate change models do not directly model human influence on climate. Rather, they assume a doubling of overall carbon dioxide concentrations in the atmosphere. Today's dire climate change predictions are based precisely on assuming a doubling of carbon dioxide concentrations, not our emissions. More recently, models have gone to tripling concentrations and more, because improvements in the models have reduced the effects of doubling. Human emissions are not factored into the models because they are too small.
For man-made emissions to cause such a result would require a Herculean effort: We would have to add trillions of tons of carbon dioxide to the atmosphere. Since the natural greenhouse gas numbers are never reported, people think that carbon dioxide in the atmosphere is a man-made pollutant, so stories often mistakenly say the models look at a doubling of man-made emissions. Why are these natural emission numbers always ignored?
At the scientific level there is a deeper mystery. After $10 billion in climate research, climate feedback mechanisms are known to be chaotic, that is, their equations are extremely sensitive to initial conditions, the so-called butterfly effect. Chaotic systems are by definition unpredictable, so how can the computer modelers make predictions? For scientists to ignore chaos is fraud; to fail to understand it is a huge blunder.
But people -- even scientists -- shun chaos theory because at bottom it interferes with their desire and deep belief that all should be knowable, that blame can be measured and penalties assessed. And perhaps, just perhaps, the climate modelers avoid the chaotic aspects of climate because it renders their models useless.
Why is the new science of chaos being ignored?
What about ozone transport? In the 1980s, research and computer modeling in the Northeast found plumes of ozone that crossed state lines. In the 1990 Clean Air Act amendments, Congress created the Ozone Transport Commission, comprising the 13 Northeastern states from Virginia to Maine, to solve the problem.
But by 1994 the Northeast states faced sanctions when computer models could not demonstrate that controls would yield attainment. The problem, they said, was ozone transport from outside the OTC region, so EPA organized the Ozone Transport Assessment Group, comprising the 37 states from North Dakota to Texas and eastward, to solve the problem. After two years of deliberation OTAG ended in stalemate: Some members claimed that the Midwest contributes 30 percent or more to ozone nonattainment in the Northeast. Others insisted the effect is nonexistent.
After millions of dollars in research, how can there be so little agreement? The answer is that human activity does not emit ozone, so we can't track it directly. OTAG's transport research used modeling and statistical science, two very different approaches. Modelers take the worst ozone episodes and play "what if". Statistical scientists take data for all episodes and look for patterns. The scientists say the modeling episodes are rare; there is no reason to expect them to repeat, so the results are irrelevant. The modelers point out that attainment is a matter of extremes, not averages, so the statistical work is irrelevant.
Modeling shows different episodes having different causes, so a given set of controls averaged over the modeled episodes tend to wash out. Modeling has produced a sea of numbers, but most modelers insist that only qualitative conclusions can be drawn. Advocates have taken the numbers they like and run with them publicly.
Statistical analysis does not show cause and effect, as modeling sometimes does. If high ozone levels extend for 600 miles, is that transport, or just the size of the episode, or something in between? Air tends to flow from Midwest to Northeast, but often it does not. What is the effect of organic compounds released by millions of acres of trees in between, the so-called 'sea of isoprene'? Why would ozone aloft come down to the surface? If ozone does flow in, doesn't it just displace local ozone that flows out? The scientists advise caution, but their findings are widely, and not always responsibly, used in the public debate.
With ozone, as with climate change, the fact of chaos has been ignored. Both the weather and ozone-producing chemistry are known to be chaotic systems, making prediction impossible. Yet the modeling results are accorded great weight in the public debate.
OTAG policy makers struggled to synthesize the work of the modelers and the scientists, but the research simply has not converged on a consensus. This fundamental fact has never been reported in the popular press nor, incredibly, has it been noted by EPA in the preamble to its proposed ozone transport NOx control rule. On the contrary, EPA claims that its proposal flows from OTAG findings, which is utter nonsense. Here too the translation from science to public policy has failed.
Why?
David E. Wojick, Ph.D., P.E. may be contacted at dwojick@shentel.net
Material presented on this home page constitutes opinion of the author.
Copyright © 1997 Steven J. Milloy. All rights reserved. Site developed and hosted by WestLake Solutions, Inc.