National Academies Press: OpenBook

Energy: Production, Consumption, and Consequences (1990)

Chapter: 2. Environmental Issues

« Previous: 1. Supply, Demand, and Reappraisal
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 73
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 74
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 75
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 76
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 77
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 78
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 79
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 80
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 81
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 82
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 83
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 84
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 85
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 86
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 87
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 88
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 89
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 90
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 91
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 92
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 93
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 94
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 95
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 96
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 97
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 98
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 99
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 100
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 101
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 102
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 103
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 104
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 105
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 106
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 107
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 108
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 109
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 110
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 111
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 112
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 113
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 114
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 115
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 116
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 117
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 118
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 119
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 120
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 121
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 122
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 123
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 124
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 125
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 126
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 127
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 128
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 129
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 130
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 131
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 132
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 133
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 134
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 135
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 136
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 137
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 138
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 139
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 140
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 141
Suggested Citation:"2. Environmental Issues." National Academy of Engineering. 1990. Energy: Production, Consumption, and Consequences. Washington, DC: The National Academies Press. doi: 10.17226/1442.
×
Page 142

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Environmental Issues

Energy: Production, Consumption, and Consequences. 1990. Pp. 75 84. Washington, D.C.: National Academy Press. Global Environmental Forces THOMAS C. SCHELLING Greenhouse warming is global in at least two respects. First, carbon dioxide (CO2) and the other gases released or withheld anywhere on earth disperse rapidly into the global inventory. The location of origin makes no difference. Second, the effect will be a change in global circulation of air and water. Although the mean rise in atmospheric temperature is commonly used as an index of climate change, the change in temperature differential between equatorial and polar regions may be a better measure of global environmental forces. The standard point estimate of global warming for a doubling of the concentration of CO2 in the atmosphere is 3° C (National Research Council, 1982~. But it is usually estimated that the warming in the polar regions associated with this 3-degree average change might be 8 or 10 degrees, whereas the change in atmospheric temperature near the equator might be closer to 1 degree (National Research Council, 1982~. Offhand this sounds like a welcome dispersion of temperature change: it will mainly get warmer where it is already very cold and warm up the least where it is already hot. But more significant is that it is the temperature gradients between equatorial and polar regions that drive the winds, which in turn drive the oceans, and a change of 7 or 8 degrees in the mean temperature difference will change the atmospheric and oceanic circulation much more than would a uniform global rise in atmospheric temperature. Most climates may get warmer, some will undoubtedly become cooler. But the observed changes will include not only temperature and temperature variation from season to season and year to year but also, probably more importantly, the amounts, . 7s

76 llI0~4S C SCHELLING the seasonal distribution, and the year-toyear variation in rainfall, snow, wind, fog, sunlight, humidity, and storms. For the purpose of comparing forthcoming changes in climate with changes experienced in the past, the mean global atmospheric tempera- ture is probably not only a reliable index but something of a measure of magnitude. Using the commonly accepted 3-degree rise from a doubling of the atmospheric concentration as an approximation to what may be forthcoming, the ensuing temperature will not only be well outside the range of atmospheric temperatures experienced in the past 10,000 years but may be several times the range of temperature variation experienced in that time. This observation is frequently expressed, and correctly, as a change in climate greater than any that mankind has experienced since the dawn of history. It is expressed more accurately as changes in cli- mates—plural not singular because different climates around the globe will change differently. Without belittling the unprecedented nature of such climate changes or the prospect of some change that is not gradual but catastrophic, it is fair to point out that most people will not undergo in the next 100 years changes in their local climates more drastic than the changes in climate that people have undergone during the past 100 years. No climate changes are forecast that compare with moving from Boston to Irvine, California, or even perhaps from Irvine to Los Angeles. The Goths and the Vandals, the Romans and the Vikings, the Tartars and the Huns migrated through more drastic changes than any currently anticipated; Europeans who migrated to North and South America similarly underwent drastic climate changes. In this country in 1860 barely 2 percent of the population lived outside the humid continental or subtropical climates; in 1980 the percentages outside these zones had increased from 2 percent to 22 percent. Furthermore the microclimates of urbanized lblyo, Mexico City, and Los Angeles have not deterred their population growth; the microclimates of London and Pittsburgh changed dramatically during the century before 1950 and have changed again almost as dramatically since then. Even urbanization itself, without the associated air pollution, changes the condi- tions created by climate. Most Americans, Europeans, and Japanese never experience muddy roads anymore. The expectation is that climates will change gradually, both over time and over space (National Research Council, 1983:Ch. 1-3~. The climate of Nebraska may gradually change into the current climate of Kansas, not into the climate of Massachusetts or Oregon. Climates will "migrate." This expectation is on the whole reassuring, but it could be mistaken. The models used in the computer simulation of climate may be incapable of producing discontinuities because the current state of meteorological knowledge is confined to continuous processes. There may be no reason to expect

GLOBAL ENVIRONMENTAL FORCES 77 discontinuities, but the fact that the models produce no discontinuities may reflect an inability to design models based on the state of the art that can discover such phenomena. Aside from a possible rise in ocean level, which I shall discuss presently, the most predictable physical and economic consequences of climate change will be in agriculture. By "predictable" I mean not that the actual changes can be predicted but that it can be reliably predicted that there will be changes. These will be changes in rainfall, winter snow for summer irrigation, humidity, daylight and cloud cover, and perhaps the health and comfort of livestock There is no reason to believe that the revolutionary improvements in agricultural productivity that have developed over the past 75 years and that in many cases have spread worldwide will not continue. Depletion of soils may continue, but control over plant and animal genetics and the possible production of new proteins may drastically change for the better what crops people will grow and what foods they will eat 50 or 100 years from now. An increase in the cost of food production by 5 or 10 percent, even 20 percent, which would be a somewhat extravagant estimate, may easily be offset many times over by another century's improvements in agricultural productivity. There will undoubtedly continue to be parts of the world that are intractably poor and dependent for a livelihood largely on local production of food or other climatically dependent crops. These countries may have little of the capacity to adapt that the more advanced countries can afford. So even if the damage to food production may not average enough on a global scale to be cause for alarm may not even be noticeable—there may be particular areas in which the damage to agriculture coupled with population growth could severely retard progress. (Population growth may be the more serious.) This situation may demand foreign aid to the poorest countries. I would neither expect nor recommend foreign aid directly related to hardships induced by climate change, but rather aid to the poorest. What I have said so far will sound to many readers as insufficiently alarmist. "Optimistic" it may appear. One reason for the unexcited tone, which I shall elaborate shortly, is pessimism, not optimism. I do not believe that serious measures will be taken over the next quarter century to curtail the emissions of carbon into the atmosphere. I do not believe that even an alarmist appraisal will lead to a substantial policy response. I therefore do not believe that exaggerating the dangers will serge a useful purpose. But there is, I acknowledge, another reason why my assessment is so mild. As I mentioned earlier, I am attempting to assess predicted changes, and it may be that our climate models predict only what we understand well enough to include in the models. Maybe we are also good at adapting to

78 THOA{4S C SCHELLING phenomena we understand, as well as good at predicting them; and the ones we do not understand well enough to predict will cause difficulty because we do not understand them well enough to adapt. In other words, there is bias in our assessment of dangers: those we understand well enough to perceive we understand well enough to overcome, those that we have no hints of may be the dangers we would least know how to meet and overcome. Reduced rainfall in Kansas 25 or 50 years from now we may adapt to with moisture-conserving agricultural techniques, genetically altered crops that require less moisture, or the acquisition and transport of water. The phenomenon is familiar, the adaptations are familiar, and the predictions are based on familiar principles of meteorology. The "collapse" of the West Antarctic Ice Sheet would be an altogether different phenomenon. As recently as 15 or 20 years ago, the accepted estimates were that the grounded ice ice resting on the sea bottom and rising a kilometer or more above sea level might, with a warming of the oceans attendant upon a warming of the atmosphere, slide or glaciate into the ocean within 75 years, causing a 20-foot rise in sea level. Like seismology in response to the test-ban controversy of the l950s, glaciology has advanced in the past decade or two, assisted by satellite sensing, and the currently accepted estimates are that if that grounded ice should be added to the ocean level it is likely to be gradual and to take several hundred years. The urgency of that particular danger is thus reduced by an order of magnitude (unless further rapid advances in the relevant glaciology bring comparable changes in estimates back in the opposite direction). What is worrisome is that there may be other phenomena, perhaps, like the ocean level, not being perceived as "climatic," that could be as devastating as a 20-foot rise in sea level and that will not, upon further inspection, yield to more benign estimates. When asked for an example, I can of course protect myself by pointing out that predicting the unpredictable, foreseeing the unforeseen, especially as an amateur, cannot be demanded of me. But when I am in a mood to worry I think about possible changes in the Gulf Stream and the Japanese current. The current global circulation models, as I understand it, do not include changes in the direction and velocity of ocean currents, and I am not sure that enough is known about the response of ocean currents to changes in wind patterns to predict whether there may be catastrophes, that is, flipflops from one equilibrium to another, rather than gradual change. Thus, there may be a missing feedback loop from warming to winds to currents to climate that, when added to the current models, will produce something more worrisome. than the migration of the climate of Kansas to South Dakota. As I said at the outset, the problem is global; and that is why it is exceedingly unlikely that anything substantial will be done to curtail fossil

GLOBAL ENVIRONMENTAL FORCES 79 fuel emissions. Any nation that attempts to mitigate changes in climate through a unilateral program of energy conservation or fuel switching (or expensively scrubbing CO2 from smokestacks) in the absence of some international rationing or compensation arrangement, pays alone the cost of its program while sharing the benefits with the rest of the world. Consider the Federal Republic of Germany, which accounts for about 4 percent of world's energy consumption and just about 4 percent of each of the three fossil fuels, coal, oil, and natural gas. If that country took the drastic step of reducing by one-third its consumption of fossil fuels, the cost in lost productivity and consumer welfare, even if it were done gradually over a period of two decades, could be equivalent to 3 - percent of its gross national product while the concentration of CO2 in the atmosphere would be reduced by barely 1 percent. Even for the United States, the largest energy consumer of all, phasing in a one-third cutback in fossil fuel consumption over the next 20 years at a cost perhaps equivalent to $150 billion or $200 billion per year at today's prices and income levels, would reduce emissions worldwide by less than 10 percent. The time to a doubling of CO2 in the atmosphere might be reduced from something like 85 years to 80 years. I think it is a fair estimate that for no individual country, with the arguable exception of the United States, is it economical to curtail CO2 emissions unilaterally in the interest of retarding climate change. Any significant effort to curtail emissions would require an inter- national rationing regime, covering the larger fraction of world energy consumption, to ration the consumption of energy, or the consumption of fossil fuels, or the consumption of carbon, in some manner that could con- fidently be expected to remain in force long enough to be effective, say 50 years or more. It would have to include the Soviet Union, it would have to include the People's Republic of China, and it may well have to include the Organization of Petroleum Exporting Countries (OPEC). It would require mandating compliance on the part of scores of nations that would greatly prefer to be outside the regime. And it would require for many nations trading urgently needed economic growth now for the dubious future ben- efits of a rationing scheme that depended on a more disparate membership than even that of OPEC. Eventually, because most of the world's known coal resources are in the Soviet Union, China, and the United States, the scheme would require those three nations to collaborate effectively and indefinitely as a cartel. The political likelihood of solid and confidently expected collaboration of that kind would be approximately zero if energy were a homogeneous commodity consumed uniformly worldwide. But to put in effect a rationing scheme the impact of which will begin to hurt and be effective only after several decades of energy growth would require dealing with economic growth itself, and that in turn requires attention to things like population

80 lots C SCHEMING growth (Ausubel and Nordhaus, 1983; Nordhaus and Yohe, 1983~. Do the Chinese claim that a policy of zero population growth is more than sufficient as a curtailment of energy use and that their country should therefore be exempt? Do the countries in the Organization for Economic Cooperation and Development participate as a unit, negotiating long-term shares in energy growth? Is there any chance they could be more successful than they have been with defense budgets, oil imports, or agricultural trade? My pessimistic conclusion is that nothing of the sort is going to hap- pen. I do not believe the Montreal Protocol on Substances that Deplete the Ozone Layer, signed in September 1987, is any harbinger for sup- pression of CO2. Economically what is at stake is two or three orders of magnitude greater for fossil fuels than for chlorofluorocarbons (CFCs) and the prospects for technological replacement of CFCs are much brighter. (The Ozone Protocol does illustrate the need for worldwide collaboration to make restrictions worthwhile: the treaty takes effect only when ratified by nations representing two-thirds of world consumption.) If world politics change as much in the next 75 years as in the past 75, a global fuel regime of some kind may become possible, but none is now foreseeable. If I am wrong, and world rationing of fossil fuels becomes economically and politically feasible, we shall still face the prospects for climate change. There is absolutely no possibility that fossil fuel emissions can cease altogether in the foreseeable future, and even the most optimistic could hardly hope that fuel emissions would stop growing within the fore- seeable future. A most ambitious goal might be to reduce by half the growth rate in fossil fuel emissions. (As the fraction of fossil fuels represented by petroleum and natural gas declines over the coming century, fossil fuel consumption will have to increase at less than half the unrestricted growth rate in order that carbon emissions be only half what they might otherwise be.) A not unreasonable estimate, for purposes of illustration only, of growth in fossil fuel consumption over the next half century might be 2 per- cent per year, a rate at which the atmospheric concentration of CO2 might double in about 85 years, reaching 50 percent elevation in about 50 years. Holding emissions to 1 percent growth would carry us beyond the middle of the next century before we reached concentrations half again as great as today's. The implied curtailment in emissions, at 1 percent compared with 2 percent, would be 10 percent at the end of the first decade, 25 percent at the end of three decades, and 40 percent by the end of five decades. That seems to me to be the outside limit to what might be economically acceptable worldwide. (How that 40 percent aggregate curtailment would be shared among consuming nations I hesitate even to conjecture.) National programs to phase in nuclear power to replace fossil fuels for electricity, even for the production of hydrogen fuels, may again become popular. But it is still hard to measure the half-life of anxiety resulting from

GLOBAL ENVIRONMENTAL FORCES 81 the accidents at Three Mile Island and Chernobyl. Any new reactors will have to be economical as well as clean. Cutting the growth of emissions from 2 percent to 1 percent may well require all electric power capacity in the future to be nuclear. Energy conservation measures deserve emphatic attention, but invest- ments in conservation will mainly be limited to what the private economy finds economical. National or international policy will probably be limited to research, development, demonstration, and technology transmission. Energy-efficient investments may yet get a boost from another doubling or more of the price of crude oil, but that is probably not a boost to be hoped for. What else may be done to cope with the greenhouse problem? CO2 can be removed from the atmosphere by increasing the mass of living vegetation or by "refossilizing" timber, burying it underground or in the ocean or coating it so that it cannot oxidize. And CO2 can be scrubbed from smokestacks at very substantial expense. Probably at enormous expense, some attenuation could be achieved in this fashion. (Some small increase in the carbon density of forests may result naturally from the enhancement of CO2 in the atmosphere.) The concentration of CO2 will therefore certainly increase, and at an increasing rate, and I consider it unlikely that we shall be rescued much before the concentration has nearly doubled. The main response will be adaptation, and most of that by ordinary people and businesses. Some of the adaptation will be by governments, but local and regional governments as much as national governments. There will be changing climates to cope with, changing urbanization, changing population densities, and in most countries probably drastic changes in the ways that people live and work and transport themselves, perhaps s~gn~ncant changes In what they eat. Much of the adaptation will seem generally "environmental" rather than specifically climate oriented. And, of course, there is continuous adaptation to climate even when it is not changing: we change the technology and the efficacy with which we heat ourselves and cool ourselves and clean our air and protect ourselves from storms and cope with droughts and floods and dispose of snow. The pace of change may be such that people will find themselves adapting to climate rather than to changing climate. Just as businesses shift to take advantage of better productive climates, they will keep shifting to better climates with perhaps small regard for the prospects of changing climates in given locations. ~ ~ ~ . . . .. There remains to be discussed a response to climate change that re- ceives so little attention that it deserves emphasis here—direct intervention in weather and climate. When Thomas F. Malone was chairman of the Committee on Atmospheric Sciences of the National Research Council, he wrote, 20 years ago, "The possibility that large effects may be produced

82 lots C SCHELLING from relatively modest but highly selective human interventions opens up the possibility that weather and climate modification may some day be operationally feasible" (Malone, 1968:1136~. And of the modification of hurricanes he said, "If five years are allowed for the development of an ad- equate mathematical model, five more years for assessing the consequences of interventions of various kinds, and then ten years of field experimenta- tion for validation, it seems unreasonable to expect much before 1990, with the probabilities fair to good that a proven technology will exist by the year 2000." He added, "The probability of success in broad climate modification is likely to exceed 50 percent by the year 2018" (1968:1138), that being the 50-year mark from the time he wrote. Most experiments with weather modification or with changing geo- graphical features that may lead to climate change have been local and regional. That has been true of cloud seeding and would be true of the manipulation of hurricanes. In a discussion of greenhouse warming, the possibility of global intervention has to be considered. An important kind of human intervention in global climate may be efforts to change the ra- diation balance itself. We know it can be done: we are doing it. That is what the greenhouse discussion is all about. The fact that we are doing it unintentionally, and the fact that the consequences may not be welcome, do not contradict that we know how, at some expense if necessary, to change the world's climate more than it has changed in the last 10,000 years. Warming the atmosphere currently is more economical than cooling it because it happens as a by-product of energy consumption that would be costly to reduce or terminate. If we were faced with a "little Ice Age" over the next century, we might be glad to get some of that CO2 in the atmosphere at no cost and without having to negotiate climate change diplomatically. But we know that, in principle, cooling could be arranged. Volcanic eruptions have done it. Discussions of "nuclear winter" took seriously the possibility that human activity might lower global temperatures cataclysmi- cally. Considering the development of nuclear energy in both its explosive and its controlled uses and the feat of landing a team on the moon and returning it safely, and that we now know how to warm the earth's atmo- sphere and possibly to cool it (though through unacceptable means), we should not rule out that technologies for global cooling, perhaps by inject- ing the right particulates into the stratosphere, perhaps by subtler means, will become economical during coming decades. A more benign example, compared with nuclear winter or induced volcanic eruptions, may be the manipulation of cloud cover. Let me again quote Thomas Malone (1968:1135~. A characteristic of the atmosphere that frustrates the weather forecaster while providing a basis for optimism on the part of the weather modifier is a tendency

GLOBAL ENVIRONMENTAL FORCES for the processes in the atmosphere to demonstrate certain traits of instability.... For example, a small pulty-type cloud may grow to a towering thunderstorm in a matter of hours; a gentle zephyr in tropical latitudes may develop into a "killer" hurricane in a matter of days; and a small low-pressure center may grow to a vigorous extratropical cyclone within a single day.... An avenue may be opened up by which great erects may be produced from relatively modest but highly selective human interventions. 83 If somebody learned in the next 50 years how to affect the extent and global distribution of certain kinds of cloud cover, incoming radiation may become manipulable by nations, international agencies, or even interested private organizations, depending on the nature of the technology, its expense, and perhaps geographical considerations. It is difficult to mention such a possibility without appearing to rec- ommend it, or to use it as a "technological fix" in the future to divert attention from some need for immediate policy intervention. I am not rec- ommending, I am predicting. Independently of CO2, we have to consider that weather and climate modification may become feasible in a period of time no longer than the elapsed time since electronics, genetics, antibiotics, and nuclear fission were unimagined. The greenhouse warming may gen- erate an interest among most nations in moderating the changed radiation balance, and if it proves more expensive to facilitate outgoing radiation than to obstruct incoming, there may be powerful motives for considering it. And if the technique for moderating incoming radiation were globally uniform or nearly so, an international agreement would have only to decide how to share the costs, a unidimensional problem compared with sharing the reduction of emissions. If intervention is more regional than global, or global but not uniform in its distribution, intervention could become exceedingly controversial. Mexico and China are counting on those hurricanes they are an essential source of rainfall for crops—whereas the Cubans, Filipinos, Japanese, and residents of the Texas coast would suppress them if they knew how. In closing I must say a word about sea level. I believe the current wisdom is that we may be in for rising sea level that could be on the order of a meter per century for several centuries (Robin, 1986~. Anything upwards of a meter, perhaps even half a meter, would primarily be due to the collapse of the West Antarctic Ice Sheet. The full hotfoot rise corresponding to the complete disappearance of that body of ice would put the White House rose garden under water, make Beacon Hill in Boston an island, and isolate the southern third of Florida by making the middle third disappear under water. A country like the United States should be able to adapt (eventually by doing, perhaps, what the Dutch have been doing for centuries~onstructing dikes). No such "easy" solution is available to a country like Bangladesh, which is densely populated in large areas that would be inundated by the

84 THOR 45 C SCHELLING full sea level rise, and which could not be protected with dikes. (If dikes were erected along the coastline to protect against seawater flooding, the area would simply be flooded with fresh water that could not Dow out to sea.) If current estimates hold up, the potential devastation of rising sea level will mainly be 100 years away, and the government of Bangladesh should worry much more about population and productivity than climate change. If the more prosperous nations were prepared to help Bangladesh at great expense to themselves, aid now would probably appeal more to Bangladesh than heroic efforts to forestall floods a century hence. (That country already has floods to cope with in this century!) Estimates of rising sea levels depend not only on thermal warming of the oceans, melting of glaciers, and what happens to the West Antarctic Ice Sheet; they can also depend on what happens to the Antarctic climate. There has been some conjecture that a warming of the South Polar air may lead to greater snowfall on Antarctica. The area of Antarctica is about one-fortieth the area of the oceans; a 1-centimeter rise in ocean level would be offset by a 40-centimeter rise in the water content of the snowfall on Antarctica, or an average snowfall of 4 meters per year. Storing water as ice on Antarctica might be the ideal solution to the water-level problem. Even the people most offended at the thought of deliberately tampering with our climate to offset the greenhouse gases may agree that learning to make it snow on Antarctica is a worthwhile project for the next century. REFERENCES Ausubel, J. H., and W. D. Nordhaus. 1983. A review of estimates of future carbon dioxide emissions. Pp. 153-185 in Changing Climate: Report of the Carbon Dioxide Assessment Committee. Washington, D.C.: National Academy Press. Malone, T. F. 1968. New dimensions of international cooperation in weather analysis and prediction. Bulletin of the American Meteorological Society 49113~1140. National Research Council. 1982. Carbon Dioxide and Climate: A Second Assessment. Carbon Dioxide Review Panel. Washington, D.C.: National Academy Press. National Research Council. 1983. Changing Climate: Report of the Carbon Dioxide Assessment Committee. Washington, D.C.: National Academy Press. Nordhaus, W. D., and G. W. Yohe. 1983. Future carbon dioxide emissions from fossil fuels. Pp. 87-152 in Changing Climate: Report of the Carbon Dioxide Assessment Committee. Washington, D.C.: National Academy Press. Robin, G. deQ. 1986. Changing the sea level: Projecting the rise in sea level caused by warming of the atmosphere. Pp. 323 359 in SCOPE 29: The Greenhouse Effect, Climatic Change, and Ecosystems, B. Bolin, B. R. Doos, J. Jager, and R. A. Warrick, eds. New York: John Wiley & Sons.

Energy: Production, Consumption, and Consequences. 1990. Pp. 85-110. Washington, D.C.: National Academy Press. Regional Environmental Forces: A Methodology for Assessment and Prediction THOMAS E. GRAEDEL The ongoing impact of human development on the biosphere is a fact of life. Development assumes many different forms, including better housing, improved communications, and advancing technology, ultimately directed at upgrading the quality of human existence. However, human development Is biospheric development and can ultimately be beneficial only if it is sustainable: that is, if the ways in which development is accomplished do not result in biospheric impacts sufficiently detrimental that they outweigh the benefits pursued. Among the most significant of the impacts are those resulting from the generation and use of energy. The preponderance of energy generation involves the combustion of fossil fuels, with resulting effects on air, water, vegetation, and so forth. Alternative technologies have detrimental aspects as well, of course. The issues involved are complex, both in the scientific and in the societal sense, and it is this complexity that has led to restricted or simplistic analyses and has hindered development of rational overall plans to optimize energy provisioning. Throughout most of history, the interactions between human devel- opment and the environment have been relatively simple and local. More recently, the interrelatedness and scale of these interactions, as well as our awareness of them, have increased. This increased complexity presents a considerable challenge to those attempting to assess the conditions of the future. It has become clear that environmental impacts and their causes cannot be envisioned in isolation as has traditionally been done, because one impact may have many causes, and one cause may have many effects. .85

86 THOMAS E. GRAEDEL In partial response to these concerns, this chapter presents a method- ology for the assessment and prediction of environmental effects of human development. It draws on this author's research experience as well as that of others (especially Clark and Munn, 1986; Darmstadter et al., 1987; Meyer and Turner, 1988~. The assessment is limited to the atmosphere and to some of its important interactions with the rest of the planetary system. Regional effects are emphasized, but global effects are considered as well, because the ensemble perspective that is desired must encompass all spatial and temporal scales of interest. A typical regional dimension is defined here as 1,000 kilometers, because most of the world's nation states tend to have spatial dimensions of that order. The assessments reported here are not intended to be definitive sci- entific assessments. As Victor Hugo said with considerable discernment, `'Science has the first word on everything and the last word on nothing." The research is designed, rather, to suggest a path for the practice of what James Lovelock (1986) has called planetary medicine, or geophysiology. Lovelock's goal is to understand the planetary system well enough to an- swer such questions as: How stable is it? What will perturb it? How much will it be perturbed? Can the effects of perturbation be reversed? This paper is addressed to providing a framework for answering these questions and thus moving toward developing the biosphere in a rational and knowl- edgeable manner. To say it another way, the work is intended to assist those desiring to become better planetary physicians. TOWARD A SYNOPTIC FRAMEWORK Noteworthy Ionospheric Properties The goal of a synoptic framework is to establish the causal relation- ships between atmospheric properties that exert significant influence on an ecosystem, and potential sources of change to these properties (and the processes behind them). The importance of changes in these properties and their consequences for individuals depends on their social, political, and current environmental circumstances. This discussion cannot embrace the full diversity of human interests that could be affected by continued alteration of the atmosphere. However, one of the clearest lessons from the assessment experience of the past decade is that unless some short and clearly defined set of atmospheric "benchmarks" is established, a meaning- ful analysis is impossible. Therefore, for the purposes of this discussion, the set of significant atmospheric properties presented in Table 1 is used as a point of departure (Crutzen and Graedel, 1986~. The goal is to understand the relation- ships between these influential properties of the atmosphere, their natural

REGIONAL ENVIRONMENTAL FORCES TABLE 1 Atmospheric Properties and Processes 87 Property/Proeess Desertion Ultraviolet energy absorption Radiation balance alteration Photoehemieal smog formation P. . . reclpltatlon aclalty Visibility Corrosion potential The ability of ozone in the stratosphere to absorb ultraviolet solar radiation, thus shielding the earth's surface fran its effects. Me eomplieated processes through which the atmosphere transmits much of the energy arriving from the sun at visible wavelengths while absorbing much of the energy radiated from the earth at infrared wavelengths. The balance of these fluxes, interacting with the hydrological eyele, exerts considerable influence on the earth's temperature. This process is commonly addressed in diseussials of the "greenhouse" problem. The oxidizing eharaeteristies of the atmosphere are due to a variety of highly reactive gases. In this paper, the emphasis is on local-scale oxidants that are often implicated in problems of smog, such as asthma, crop damage, and degradation of works of art Ihe acid-base balance of the atmosphere in rain, snow, and fog. Visibility is reduced when visible light is absorbed or scattered by gases, moisture, or particles in the atmosphere. Me ability of the atmosphere to corrode materials exposed to it, often through the ehlondation or sulfurization of marble, masonry, iron, aluminum, copper, and other matenals. fluctuations, and how they might be affected by human activities. Recent advances in our understanding of atmospheric chemistry and its interactions with the biosphere now allow specification of such relationships in terms of fundamental biological, chemical, and physical processes. Interactions Current knowledge about the atmospheric properties affected by changes in specific atmospheric chemicals is given qualitative expression in Figure 1. The convention that is used is to indicate only direct effects. For example, changes in ozone concentrations affect ultraviolet energy ab- sorption because the ozone molecules themselves capture the photons, or rays of light. Halocarbons and nitrous oxide, though surely relevant to ultraviolet energy absorption, are not shown to influence this atmospheric

88 THOA{4S E. GRAEDEL property because their effect is indirect. It is important to note from Figure 1 that a significant number of chemicals have multiple impacts. Sources of atmospheric chemicals or policies that affect the introduction of these chemicals into the environment must therefore be assessed in the light of these multiple atmospheric impacts, and the "environmental-problem- of-the-month" approach that has been so common in the past must be avoided. Figure 2 shows the current state of knowledge about the sources of atmospheric chemicals (see Able 2), again indicating only direct effects. It is again important to note that many atmospheric chemicals have more than one source. The pervasive influence of the biosphere~cean life. plants, soils, and animals on atmospheric chemistry- is also evident in the figure. To complete the connection between the sources of atmospheric chem- icals and their influence on the important properties of the atmosphere, it is necessary to account for indirect effects the fact that changes in chemical species A may affect a given atmospheric property through an intermediate influence on chemical species B. Understanding the indirect effects of chemical interactions is one of the central tasks of contempo- rary atmospheric science. The immense complexity of even the relatively well understood interactions precludes their discussion here. Conceptually, however, the substance of such a discussion can be captured in a matrix constructed along the lines of Figure 3. The chemical compounds of Figures 1 and 2 thus provide the common basis for an analysis of biogeochemical processes that link changes in the contribution of one source of a chemical to all the important atmospheric properties affected. The three figures can be combined to provide a synoptic framework for atmospheric assessment. One can begin with a property such as '~precipi- tation acidity" and its direct chemical causes (Figure 1), trace those back through their interactions with other atmospheric chemicals (Figure 3), and finally identify the sources of those chemicals that influence precipitation acidity (Figure 2). The synoptic matrix relates the change of each potential source of atmospheric chemical to the affected atmospheric component and can be regarded as being generated by a matrix operation: [source x prop- erty] = [source x chemical] [chemical x chemical] [chemical x property], or [Figure 4] = refigure 2] [Figure 3] [Figure 1~. The initial result of an analysis based on such a matrix, shown in Figure 4, is qualitative, as befits the present state of knowledge. It also includes estimates of the reliability of that knowledge, an important component of such an assessment effort (Ravetz, 1986).

REGIONAL ENVIRONMENTAL FORCES ATMOSPHERIC PROPERTY ft~ ~ ~ I ~ cop CO NOy A SOx H2S g Cat - I In C (soot) o3 COS . CH4 CxHy N2O NH3/NH 4 Organic S . Halocarbons Other Halogens Trace Elements 89 FIGURE 1 Direct effects of atmospheric chemistry on important atmospheric properties. The squares indicate that the listed chemical is expected to have a significant direct impact on the listed property. Definitions of the atmospheric properties are given in Table 1. Data are from sources listed by Clark (1986~.

9o SlUG~013 a~eJ1 · ~ SU060leH J6410 suoqJecOleH S o!ue6~o + HN/0HN oZN AHX~ o tHO ~: :E SOO cn ~0 (~oos) SZH - xos XON - 00 Zoo 0 CO E =, ~5 Ct U) U' , a O ~ .° Ct a) ~— > ~n o ._ ~n o C: U) cn CO o ._ m 3~nos ~o o E E c ~n a) cn CO a' o tL .m - ~n CD ~ — (,, ~ .> o ~ C~ Ct a~ c) . _ ~ C) C~2 C.) Ct C~ o a' ~ _ C: C: o ._ . _ _ ~ s: .= ~ a Ct _ CQ '— ~V . S.= ._ _ ~ o." =: ~ " C) O '_ . - ~ ~ CQ ~: O C) CC ~ O _ cc ." s~ ~ O Ct Ct X ~ O .L) — C)_ LI4 ~ - Ct CQ s" o C~

REGIONAL ENVIRONMENTAL FORCES TABLE 2 Sources of Atmospheric Chemicals 91 Source Descnpnon Oceans and estuaries Includes coastal waters and biological activity of the oceans. Vegetation and soils Includes activities of soil microorganisms; does not include wetlands or agricultural systems. Wild animals Includes microbes, except for those of soils; does not include domestic and marine animals. Wizards An important subcomponent of vegetation and soils; does not include nce Biomass burning Includes both natural and anth~ogenic bumping. Cop production Includes rice, fenili7~ion, and irrigation, but not forestry. Domestic animals Includes grazing systems and the microbial fauna of the guts of domestic animals. Petroleum combustion Includes impacts of refining and waste disposal. Coal combustion Includes impacts of mining, processing, and waste disposal. Industrial processes Includes cement production and die processing of nanfuel minerals and chemicals. NOTE: See also, Figure 2. ASSESSMENTS The simplest atmospheric impact assessments involve only a single cell of the matrix. A typical example is the study of the impacts of a single source, such as a new coal-fired power station, on a single noteworthy atmospheric property, such as precipitation acidity (location a in Figure 4~. More complex atmospheric assessments have addressed the question of aggregate impacts across different kinds of sources. A contemporary example is the study of the net impact on the earth's thermal radiation budget caused by chemical perturbations due to fossil fuel combustion, biomass burning, land-use changes, and industrialization (e.g., location d in Figure 4~. The assessment then becomes a column total (location e) in the synoptic framework An example of a "column" assessment is Bolin et al. (1986~. Even more useful for the purposes of policy and management are assessments of the impacts of a single source on several noteworthy at- mospheric properties. The simple study noted above would fall into this category if it assessed the impacts of coal combustion not only on acidifi- cation but also on photochemical oxidant production, materials corrosion, visibility, the radiation balance, and stratospheric ozone (i.e., locations b in Figure 4~. The impact assessment then becomes a row total of the

92 Impact of THOMAS E. GRAEDEL Impact on Ox Ox ~ _ 0 I o CO o OF O In 8 ~ U)570 ~ In - a) a) $ cop CO NOy A Son H2S _ ~ _ C (soot) o3 COS CH4 CxHy N2O NH3/NH Organic S _ Halocarbons * Other Halogens Trace Elements FIGURE 3 Atmosphenc chemical interaction matrix for the compounds listed in Figures 1 and 2. As an example of the matrix element entries needed to fill this matnx, the element indicated lay the asterisk represents an assessment of how changes in halocarbons affect ozone concentrations. synoptic assessment matrix (location c). Examples of"row" assessments are National Research Council (1979, 1981). Figure 4 shows that the sources of most general concern, as indicated by their impact ratings, are almost wholly anthropogenic: fossil fuel com- bustion, biomass combustion, and industrial processes. Emissions from crop production, especially methane from rice paddies, and from estuaries near heavily populated areas may have future impacts. Some sources (the animal kingdom and vegetation) have sufficiently small effects on atmo- spheric processes that they need not cause much concern, even if their emissions should increase.

REGIONAL ENVIRONMENTAL FORCES Atmospheric ~ Sources ° ~° ' ~ 8 0 Oceans, Estuaries (3 ~ Vegetation, Soils =~ O =3 === Wild Animals =3 Wetlands =, O Biomass Burning C' - ) ODD ~~ ~ G) ~ Domestic An ~ ~ ~ O _ Petroleum Combustion - ' ~ _ _ Coal Combustion O - ~ _; ~b ~ c Industry ~ ~ ~ ~ ~ All = e == = = = Net Effect of All Sources on Each Property Potential Importance (ca. 1985) Some I ~ Moderate Major I_ Controlling 93 CO a' ._ a) Q o o a) Cal o CO C) LU o C) a) LL 4— a) A Assessment Reliability (ca. 1985) Moderate FIGURE 4 A synoptic assessment of impacts on the atmosphere. In this figure, adapted from Clark (1986), the atmospheric properties defined in Table 1 are listed as the column headings of the matrix. The sources of disturbances to these properties as defined in Able 2 are listed as row headings. Cell entries assess the relative impact of each source on each component and the relative scientific certainty of the assessment. "Column totals" would, in principle, represent the net effect of all sources on each noteworthy atmospheric property. "Row totals" would indicate the net erect of each source on all noteworthy atmospheric properties. These totals are envisioned as judgmental, qualitative assessments rather than as literal, quantitative summations. The cells labeled a through e are discussed in the text.

94 THOMAS E. GRAEDEL Figure 4 also suggests areas in which useful interdisciplinary studies could be performed. These are areas that have considerable potential im- portance but whose assessment reliability is low to moderate. The two best candidates for such studies are biomass burning (thought to play an im- portant role in tropical photochemistry, the atmospheric radiation balance, and visibility) and vegetative emissions (potentially linked to photochemical oxidant formation, the radiation balance, and precipitation acidity). The diversity of primary anthropogenic sources of causative chemical species shown in Figure 4 is surprisingly small: fossil and biofuel combus- tion and industrial processes being dominant. There is enough evidence to state that a significant decrease in the rate of fossil fuel combustion would tend to stabilize the atmospheric radiation balance, improve visi- bility, hinder smog formation, and minimize acidic precipitation and its effects. Stratospheric modification is best constrained by devising alterna- tives to the use of chlorofluorocarbons (CFCs) as aerosol propellants and refrigerants. The involvement of transition metal chemistry in droplets and aerosol particles is not yet understood, but better control of metal emission from combustion and smelting operations may be desirable to alleviate acid deposition problems. The problem of corrosion would have to be handled differently, because natural sources are responsible for many of the corrosive agents. Thus, it might be most effective to reduce corrosion by improved selection and treatment of materials rather than by control of emission sources. EXTENDING THE FRAMEWORK IN SPACE AND TIME Concepts and Goals Thus far, the synoptic framework developed in the previous section has only considered interactions occurring at a given instant in time. However, this is insufficient because many of the impacts of human development on the environment are cumulative. This arises because many sources of atmospheric disturbance operate over space and time scales such that the disturbing species can accumulate faster than it is removed. Although the interplay is complex, to a first approximation, the longer a species remains in the atmosphere, the more likely it is to accumulate and the greater its spatial impact can be. For example, heavy hydrocarbons and coarse particles are short lived, dropping out of the atmosphere in a matter of hours. Hence, they have little opportunity to travel more than a few hundred kilometers from their sources. On the other extreme, the residence time of carbon dioxide (CO2) in the atmosphere is so long that production of this gas anywhere in the world contributes to its effects everywhere in

REGIONAL ENYIRONMENTAL FORCES 95 the world. This is why the greenhouse effect has its long-term, global- scale character. The species with moderate atmospheric lifetimes include a group of chemicals associated with the acidification of precipitation. Because these substances last a few days, they have the opportunity to affect regions as far away as a 1,000 kilometers or more from the source of the emission. These concerns, together with the evidence that the concentrations of many chemical compounds in the biosphere are increasing, indicate that additional extensions to the conceptual framework are needed to provide it with spatial and temporal dimensions. The effort described earlier focused on present-day impacts across various local, regional, and global conditions. One can envision preparing separate versions of the framework shown in Figure 5 for global, regional, and local interactions, and for different epochs in time. The assessment should eventually consist of a single, global- scale analysis, plus several analyses for specific, large-scale regions (e.g., Europe) selected to reflect interesting interactions between development and environment. For each spatially defined regime, a sequence of figures would be needed to show how the relations between sources and noteworthy env~- ronmental properties change through time. As suggested in Figure 5, this sequence might consist of separate versions of Figure 4 created to reflect "slices in time" through the evolving conditions at 25- or 50-year intervals. In the version being explored, this sequence will extend several hundred years into the past and a century into the future. The result will help to put the changing character of interactions between human activities and the environment into a truly synoptic historical and geographical perspective. This historical dimension requires assessment of the kind of human activity being carried out at specific times in the past, and how much and what kind of activity might occur in the future. Air Quality Assessment and Scenario Design The degree of development of a country or geographic region has a major influence on its air quality, and the intensity of technology and of land use in a geographical region define many of that region's impacts on the atmosphere and therefore must be taken into account. It is often pos- sible to locate information about the most frequently monitored air-quality parameters in the major cities of the world. Information is sometimes available also for less widely reported atmospheric constituents or for other locations, although such information is generally unearthed only through informal contacts and negotiated exchanges. Where no measurements have

96 17:IOA~S E. GRAEDEL Oh Let i: o in L \ \ ATMOSPHERIC PROPERTIES \ 1~. ANAL L' \ \ \ \ loop ]Ol! Impact \ \ \ \ \ \ FIGURE 5 A history of disturbances to the atmosphere, as might be expressed through a time series of source-impact matrices such as that shown in Figure 4. This display suggests how one of the matrix elements would be evaluated at each of the time slices, perhaps becoming more significant with time as indicated here lay the shading of the symbol. The full assessment would include such an evaluation for each of the individual matrix elements at each time slice.

REGIONAL ENVIRONMENTAL FORCES INTENSITY OF LAND USE (people per km2) High Low Hi y Q o — o o llJ IL o cn Z ~ ~ O Z ~ 97 Western Europe, North Japan America Southeast Humid Asia Tropics FIGURE 6 A development matrix for the interaction between intensity of land use and intensity of technology. Regions Apical of each matrix element are listed. SOURCE: After Crosson (1986~. been made, information on emissions sources such as power plants, indus- trial processes, and transportation facilities can often suffice to give initial estimates of air quality. As noted, human response to atmospheric impacts depends, in large part, on the current developmental and environmental situations that pre- vail in a given region. For this reason, it is useful to categorize regions on the basis of land use and energy intensity. Figure 6 shows simple represen- tation of divisions into which the earth may be divided by these parameters (Crosson, 1986~. Darmstadter et al. (1987) have shown that it is possible to project future air quality for regions typical of each cell in the Figure 6 matrix: Western Europe (high intensity of both land use and technology) Northeastern North America (high technology, moderate land use) India's Gangetic Plain (high land use, low technology) Amazonia (low land use, low technology) This straightforward classification scheme embraces some obvious re- lationships: that a high intensity of land use implies impacts on soil and

98 THOMAS E. GRAEDEL groundwater chemistry. Perhaps less obvious are the spatial scales of the impacts, which tend to be large for high-technology regions and small for low-technology regions. It is beyond the scope of this chapter to discuss examples of assessment in each of the regions of Figure 6. Europe is an appropriate region to show the process, however, because the high density of technology and population in the region has resulted in readily identifiable impacts. It is an appropriate region also because it may serve as a predictor of trends in regions at lower levels of environmental loading, if development in those regions proceeds along lines similar to those of Europe. Do development scenarios were used in the Darmstadter study; they can be summarized briefly as follows. The first (referred to as scenario C in reference to its constant emission coefficients) is a conventional growth scenario for the world. It predicts global growth rates of 1.5, 3.2, and 2.1 percent per year for population, gross national product, and total energy use, respectively, for the period 1980-2030. For the period 2030-2080 the comparable growth rates predicted are 0.3, 2.2, and 1.8 percent per year. For each period, appropriate growth rates are assigned also to particular geographical regions of interest. The emission factors (i.e., the mass of emittants per unit mass of feedstock in a given source process) are assumed to remain constant at 1980 values. Scenario C is regarded as a conservative but not particularly restrictive projection for world development. The second growth scenario (termed scenario D in reference to its decreasing emission coefficients) is identical to the first in its global and regional growth rates. It differs from scenario C by assuming emission factors for the sources will decrease over the century 1980 2080 at a rate of 1 percent per year. For this scenario, therefore, increases in the total use of feedstocks are offset somewhat by improvements in technology. Scenario D is more conservative than scenario C in its potential atmospheric impacts. Given these scenarios, it is possible to use a variety of computer models and evaluation techniques to derive predicted concentrations of selected atmospheric parameters. In the study by Darmstadter et al. (1987), four noteworthy atmospheric properties, photochemical smog, acid precipitation, corrosion of metals, and stratospheric ozone were selected for analysis for the epochs 1890, 1920, 1950, 1980, 2030, and 2080. The results for Europe are reported in liable 3. Assessment for Europe The "Europe" study region used by Darmstadter et al. (1987) comprises all of Western Europe. Heavy industry occurs throughout the region, with several localities of intense industrialization. The density of motor vehicles is high, particularly in and near the major urban centers. Two urban areas

REGIONAL ENVIRONMENTAL FORCES TABLE 3 Environmental Quality Assessments for Europe 99 Epocl1 1890 1920 1950 1980 2(80C 21)80C 20SOD 2()80D O3, pph 10.0 20.0 35.0 55.0 180.0 280.0 130.0 130.0 Smog seventy L L I^l M H H H H Precipi~on pH 5.4 5.4 5.3 4.3 3.5 2.8 3.8 3.3 P~ec~pitanon acidity L L L M H H H H SO2, ppb 7.0 1 1.0 12~0 14.0 40.0 160.0 25.0 60.0 a-, pe~ 30.0 37.0 39.0 4~0 62~0 120.0 54.0 7~0 Co''osion seventy I^I M M M bSJH H M ~JH KEY: C~onstant~missions scenano; D~eclining emissions scenano; L - ow unpact; M~noderate impact; and H~igh unpact. SOURCE: Darmstadteretal.(1987). serve to characterize the region. One is Brussels, a city of 2.2 million people, generally upwind of the principal industrial centers of the region; the other is Stockholm, a city of 1.6 million people, which is generally downwind of emission sources. Although variations in topography exert influences on local weather conditions, a principal feature of the weather pattern is its consistency for all of Europe. The pattern characterized by unsettled weather throughout the year as a consequence of the migration of weather systems from west to east. Sometimes, however, a region of low or high pressure can be stationary over a wide area, resulting in uniform weather there for several days or even weeks. Such conditions promote the buildup of atmospheric contaminants. Air quality measurements in Europe have been taken for about two decades, but extensive data are available only from about 1975. The 1980 epoch can thus be readily defined, and it can be related in some detail to current emissions inventories, especially those for sulfur dioxide (SO2~. Emissions data for nitrogen dioxide (~02) are less complete, although the need for such data is minimized somewhat by the extensive ozone (03) measurements that can be drawn upon. Selected measurements of other airborne trace species are available as well. In the absence of

100 THOMAS E. GRAEDEL historic measurement data, estimates of the atmospheric properties can generally be made by comparing human development activities at earlier times with those of the present day and inferring emission fluxes from those comparisons. To project atmospheric properties over future epochs, plausible models for development must be formulated to estimate the impacts of biospheric development. These models are initialized and validated by the historical information that is available. The development scenarios selected for Europe call for increases of 0.5, 2.4, and 1.8 percent per year, respectively, in population, gross national product, and total energy use for the period 1980 2030. For the period 203~2080, the corresponding rates are 0.3, 2.0, and 1.5 percent per year. In nearly every case, these numbers are significantly lower than the estimated global average values; for example, Europe is expected to undergo development during the next century at rates lower than those for most of the rest of the world. Regional Impacts Photochemical Smog Although photochemical smog consists of many trace species, ground-level ozone is its best single indicator. Bojkov (1986) and Volz and Kley (1988) have reevaluated and summarized the surprisingly extensive ozone data from a number of stations in Europe during the interval 1850-1900. The ozone concentrations during that period are about one-fourth of the mean of the daily maximum values of precise surface ozone measurements taken in the same geographical regions during the past 10-15 years, indicating that significant increases have occurred. For 1980, measured concentrations establish the average European ozone value. The concentrations during the epochs between 1900 and 1980 are established primarily on the basis of the trends in emissions of the oxides of nitrogen (NOT. Table 3 indicates that for scenario C, the predicted ozone concentrations are quite high, the values being comparable to those of the Los Angeles basin on a smoggy day. A more moderate increase in smog concentrations is predicted for scenario D. Acid Precipitation Data on precipitation chemistry take many forms, but the acidity, or pH, is perhaps the most meaningful. A few historical measurements of precipitation chemistry in Europe are useful. For the mid-19SOs the concentrations of sulfate and nitrate in precipitation were determined in Stockholm (Environment '82 Committee, 1982~. Some his- torical data on acidity of precipitation also exist for several locations in Europe, although data on sulfate (SO4-) and nitrate concentrations are thought to be more reliable. Contour maps of precipitation acidity in Eu- rope for the past several decades are presented by Likens et al. (1979~.

REGIONAL ENVIRONMENTAL FORCES 101 Precipitation chemistry in earlier epochs can be estimated by relating its characteristics to that of known levels of emission of precursor species. The data described above establish the values of acidity or related parameters at several epochs. Emission fluxes and relational evaluation techniques can then be used to estimate acidities for the development scenarios. As shown in Able 3, acidity of precipitation was generally of little concern before 1950, when it began to increase rapidly. The projections for the future, by either scenario C or scenario D, indicate that high acidity of precipitation is envisioned for the twenty-first century. Corrosion of Metals Because the corrosion of metals exposed to the atmosphere nearly always involves chlorine and sulfur, it is appropriate to study common atmospheric forms of those elements. Given 1980 data, one can derive historical SO4- and C1- concentrations and proceed to make corrosion assessments. Precipitation data for sulfate in the early 1950s allow a good assessment for that epoch as well. An assessment of corrosion potential in Europe is then derived by relating the precipitation SO4- and Cl- concentrations to the European fluxes of SO2 to the atmosphere. The results, shown in Table 3, indicate that the corrosion impact has been low to moderate since the midpoint of the current century. The projection for development scenario C (constant emissions) is that atmospheric corrosion during the twenty-first century will be very severe. Under scenario D (declining emissions), the corrosion impact is slightly less severe but still worthy of concern. Global Impacts Ultraviolet Absorption Europe is a major producer of CFCs. As such, it bears a substantial responsibility for any stratospheric ozone depletion that occurs, and hence for decreases in the ultraviolet absorption capacity of the atmosphere. Under scenario C, the impact attributed to Europe is severe. Under scenario D, improvements in technology and reductions in manufacturing levels modify that impact somewhat, but it is still moderate to moderately severe throughout the period included in the present study. The study did not include the emission controls that may result from the 1987 multinational Montreal Protocol on Substances That Deplete the Ozone Layer, which, if implemented, would reduce the impact of the assessments. Atmospheric Heat Retention The heat retention propensity of the atmosphere, that is, the potential for the greenhouse effect, is directly related to the atmospheric concentration of carbon dioxide and other gases capable of absorbing the outgoing radiation from the earth's surface.

102 THOMAS E. GRAEDEL Changing carbon dioxide scenarios were not studied by Darmstadter et al. (1987) but have been widely investigated elsewhere. A common calculation for the study of atmospheric heat retention is one in which the concentration of atmospheric carbon dioxide is set to twice its current value. (It is predicted that such concentrations will be reached at some point in the twenty-first century.) The results of calculations by several groups of researchers have been presented and discussed by Luther (1985) and Grotch (1988), and some of their comments are abstracted here. For doubling of the CO2 concentration, the change in global mean surface air temperature ranges in different models from 1.5 to 4.5°C. The increase is not uniform but is generally 1-3°C near the equator and 16°C at high northern latitudes during winter months. This increase in temperature will be sufficient to reduce snow cover, melt some sea ice, and strongly influence the global water cycle. The models predict warming in the troposphere and cooling in the stratosphere, with somewhat unpredictable impacts on the global circulation patterns. Cloud cover and precipitation will change as well, though the ability of researchers to model these changes is somewhat uncertain. All of the models predict an increase in global mean precipitation (note that a warmer atmosphere can hold and thus cycle more water than a cooler one). Precipitation decreases will probably occur in some areas of the globe, however. It is difficult to predict the environmental impacts of increased atmo- spheric heat retention in specific regions, especially small ones, since model simulations do not agree well on length scales less than a few thousand kilometers. Assigning responsibility for the global impacts is easier, because the causative agents are identified: carbon dioxide and, to a lesser extent, methane, nitrous oxide, ozone, and a variety of chlorofluorocarbons. Fossil fuel combustion is the dominant source of CO2 and nitrous oxide as well as the principal source of the precursors of tropospheric ozone and a minor source of methane (Wuebbles and Edmonds, 1988~. Allocation of the global effects of atmospheric heat retention to specific regions can thus be made on the basis of each region's use of fossil fuels and chorofluorocarbons. Summary As shown in Able 3, most air quality indicators that were satisfac- tory near the end of the nineteenth century are noticeably degraded today and can be expected to deteriorate markedly from present levels over the century to come. Given the recognition of a problem, the next informa- tion requirement is to identify the factor or factors causing that problem. Identification of this type is the purpose of Figures 7 and 8. These figures follow the intent of the "time slice" diagrams of Figure 5, but use clustered slice rectangles to condense six Figure 5 style diagrams into a single display.

103 , ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~, 1~ ~ ~ ~ ~ ~ii~ ~1 o' _ _ _ ~ o 1~-.~1 1~1 ·_ , ~ ~. ~ ~ ~ ~ ~ ~ :~: T1 ~ ~ ~ ~ ~: ~ ~ ~ ~ ~, o ,~ ~ ~ ~ ~, , ~ ~ ~ ~ ~ ~ ~ ~ ~ ~, ~~:~:~:~l L~-~:-~-~l o ... ' ~ ~ ~:~ ~ ~ ~,~:~ _ _ ,~: ~: ~ ~ _ ~~ T ~ :~: T~ 72 , ~ ~ ~ ~ ~ ~ :~ ~ ~ ~ ~ ~ o ~ ._ O.~ [~;~1 _ ~ ,~ ~ ~ ~ ~, ·_ ~ ~2~: iT ~ ~ ~ ~ ~ ~ ~ ~ '~ ~ 1~-~:~1 Q 1~:~:~;~l ~ ~ ~ ST ? ~ g ~ 1~:~.~ ~! 1~:~ ~:~ ~ ~ ~1 C~ ._ ~ O S O cn - Q r ~ I~-~-~l I 1 1~ ~: ~ ~:~ ~ ~:~ 1 ~1 ~0 . o cn o . ~5 C~ I :~:~ ~ ~ ~ ~ ~1 t ~ ~ ~ ~ 7: T251 _ ~:~: ~ ~ ~ .~..~ ~' ~ ~ ~, r ~ ~ :~' l ~: ~ ~ :~ ~ ~ ~ ~ :l o ._ 0e 3 o cn U. o ._ m 1~ ~ ~ ~ T 1~::. ~: ~ ~ ~ :~ _ ~ ~:~ ~ ~ _ ~:~:~ ~ ~ l ~:~ ~ ~ ~: : ~, ::~:~ ~ ~ ~ ~:~ ~:~ :~:~::~:~ o ._ _e ~n 3 8 o 3 o Q o ._ - ~n 8 o - o 1 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ; ~ ~ ~ ~ I I? ~ ~: ~ ~ ~:~:~:~1 1 ~-~-~.~d 1~ ~ 72~ :~~ ~: ~ 1 I ~ ~ ~.~.~. [~.~ ~:~ _# 3 - :^ - C~ 3 - CD a _ — t ~ C,8 O ~ ._ ~ ~n o' ~: o~oe o~oa oo~ °~a, o~ ~66 o~ ~ C) Q - S o Q O s C~ ._ ~ ._ ~a ·c C~ o ~: cq s~ o Q) - t ~: ~ - ct CQ ~ o . - o . - - - ~ o ~ wo ~ _ ca ~ e o ~ C) Ct5 ·E s~ ~ ~ - ~ . - - cD ~ ~ cs) ~ oo

104 THOMAS E. GRAEDEL On these figures, a qualitative source impact assessment is shown for each epoch for the several environmental impacts produced by each source type. It is instructive first to examine Figure 7 in detail, which treats impacts for the past century in the context of scenario C. The bottom row in Figure 7 is a coded version of the environmental quality assessments given in Table 3 (e.g., it contains the column summations of the impacts). The display demonstrates that few measurable impacts occurred before 1980 and that only the combustion of coal and petroleum cause major environmental impacts at present. For the future, the same sources are expected to be those primarily responsible for the deterioration in regional air quality, with industrial emission of CFCs contributing to stratospheric ozone reduction. Biomass combustion and emission from vegetation and soils will play small roles, especially in photochemical smog production. The source impact assessment for Europe under scenario D is shown in Figure 8. It resembles that of Scenario C except for modest decreases in the severity of the impacts, reflecting the assumption of declining emission coefficients. Even with this assumption, however, the impact on precipita- tion acidity remains very high in both 2030 and 2080; the impact on smog and corrosion improves slightly but remains high. The ensemble impact is rated high in 2030 and very high in 2080. Thus, even the unremarkable development postulated in scenario D and the assumed significant decline in emission coefficients (more than 60 percent) leave the region with major atmospheric perturbations a century from now. As in scenario C, coal and petroleum combustion and CFCs are the sources primarily responsible for this deteriorating condition. Since Figures 7 and 8 (and similar figures can be produced for other geographical regions) serve as summaries of the assessment efforts, it is appropriate to call attention to certain aspects of the diagrams. One point is that smog, acidity, and corrosion are local or regional in scale and are largely a consequence of emissions from within the region. Stratospheric ozone depletion, however, has been assessed on a global basis and the indication in the diagram refers to a regional contribution to that global impact. Strict parallelism is thus sacrificed to the desire to communicate a more comprehensive view of the effects of emissions. A second point with respect to Figures 7 and 8 is that the four atmospheric perturbations are assigned equal value in determining the total impact from each of the sources. Whether different weights should be assigned to the perturbations is a policy question, not an analytic one; Darmstadter et al. (1987) and the extensions to it considered here adopted the simplest of the possible choices.

105 In — t) _ CO O Q Cal ._ S o NO ~ O An lo ._ CtS ·— — ~ ~ ~ ~21 JO 4$ L~-~-~-~! ~ ~1 1. ~,~.~1 ,~ ~~ ~~-~ a. At, - ~5 ._ ~ o s ~ — do En 1 ~~ ~~ ~~ I _ ~~:~ ~~ ~~ ~~ ~~ - o - o ._ CR >__ E - W<o~ odor odor odd Obey 0~e odd o .c Ct a: CQ o no - CQ - s: CQ au ~2 CO Ct Cal ._ o U. Go ~4

106 THOMAS E. GRAEDEL EXTENDING THE FRAMEWORK TO DIPPE;RENT REGIMES It is important to recall at this point that the previously presented assessments have been limited to the atmospheric effects of development. In fact, other regimes in the biospheric system are important as well. For example, concern about the cumulative erects of acidic precipitation is based not on the accumulation of certain chemicals in the atmosphere, but rather on the accumulation in other media, such as soil and water, and on the increasing fluxes of precursors of acidic precipitation as a consequence of the growing use of fossil fuels. Expanding the synoptic framework to contend with these additional environmental and developmental dimensions is a major part of research on sustainable development. Beginnings have been made for water systems (Douglas, 1976) and soil systems (Harnoz, 1988~. A relatively easy addition to the framework depicted in Figures 1 - would be one or more noteworthy environmental properties that reflect the role of atmospheric chemicals as direct fertilizers or toxins for plants. Such a modification would allow the integrated treatment of such phenomena as the stimulation of plant growth by carbon dioxide and its inhibition by sulfur oxides both products of fossil fuel combustion. Somewhat more ambitiously, the approach could be expanded beyond its present chemical focus to include the appropriate physical and biological processes and the sources of disturbance to them. Dickinson's (1986) sketch of a compre- hensive framework for understanding the impact of human activities on climate shows the potential of such an integrated approach. Ultimately, the need is for a qualitative framework that puts in perspective the impacts of human activities and natural fluctuations, not just on the atmospheric environment, but also on soils, water, and the biosphere as a whole. Although a comprehensive treatment along the lines of Figures 8 and 9 has yet to be made for other regimes, it is possible to envision one way in which such an assessment might be accomplished. The first step would be to select regimes of interest other than the atmosphere, such as oceans or soils. The second step would be to use the same sequence used for the atmosphere in Figures 1 - to determine the matrix elements on a source-impact diagram similar to that of Figure 4. A possible example for soils is shown in Figure 9. The matrix elements are then summed to give a total regime impact for each source; these appear in the right column of Figure 9. The same sequence of operations is repeated for each regime of interest, not only for the present and perhaps for the past, but for future times as well. In the final ensemble display, the total impact columns from each regime display are extracted and combined, producing a display of the

107 os — I, _ So - ~ .~ ~-~ ~ ~' .= ~~ ~~ ~~ - if-- - ~ , ~ ~-~.~.~. o o ~~-~.~.-~.~--~ U. ·O In A: ~ 1 ~ a, ~ m ~ ~ EE ~ ~ ~ C.) 0 0 0 0 0 0 In In In In ~ Ct 5~; ~ at 9~z . ~e,,,j,^ ~°6? So> So> 1 _ E ~ au Q S - C) Ct ._ - o CQ o C A C) - o C ;` — 4) ._ S ~ O C O S , CQ C) ~ Ct

108 l In — ,, ~ as o s a, o s In o ~ m ~ ~ in, ~ 0 0 0 In In In ~°6, odor odor °~156~ USE' odd odd C) Ct 0 ._ C) o - ._ "C 0 ¢ I: ._ ~ 0 Ins ·— CD ~ ._ _ ~ _ . _ = ~~ 0 O ~ ~ 0 ~ 0 i_ ~

REGIONAL ENVIRONMENTAL FORCES 109 type shown in Figure 10. A sequence of such diagrams would be used to illustrate progress or retrogression over time. DISCUSSION AND CONCLUSIONS Human activities are now major forces affecting the evolution of our planet. These forces have impacts on the environment on many different time and space scales. On a local spatial scale, concern about urban carbon monoxide concentrations is important. On a regional scale, the health of a major river may be of concern. On a global scale, the stability of the climate cycle is a central question. Current assessments of all of these impacts are important, as is prediction of future impacts. Few environmental impacts have a single cause, and few causes have a single effect. If problems are examined one at a time, a remedy for one problem may exacerbate another. The planetary system is complex but not incapable of being studied in ensemble form if useful techniques for doing so are developed and applied to the problem. It is, of course, much easier to provide a framework of unfilled matrix elements for environmental as- sessment than it is to fill the elements with assessments or projections made by highly trained and highly motivated scientists, but only by accomplishing such assessments and projections can environmental problems be viewed from a rational perspective. The methodology presented in this chapter for assessment and pre- diction of the consequences of environmental forces is primarily applicable to atmospheric impacts at the regional scale, although the approach can be extended to other environmental regimes and time and space scales. It is obvious that global demand for supplies of energr is increasing, and it is equally obvious that the environmental performance of the technologies used to provide that energy will be crucial to the sustainability of the planet itself. Informed decisions regarding impacts and their causative technolo- gies are crucial; they can be well informed only if the information on which they are based is comprehensive, expertly evaluated, clearly presented, and fully utilized. ACKNOWLEDGMENT ~ thank W. C. Clark for insightful comment and continuing encourage- ment. REFERENCES Bojkov, R. D. 1986. Surface ozone during the second half of the nineteenth century. Journal of Climate and Applied Meteorology 25:343-352.

110 THOMAS E. GRAEDEL Bolin, B., B. R. Doos, J. Jager, and R.A Warrick. 1986. The Greenhouse Effect, Climatic Change, and Ecosystems, SCOPE 29. New York: John Wiley and Sons. Clark, W. C. 1986. Sustainable development of the biosphere: Themes for a research program. Pp. 5~8 in Sustainable Development of the Biosphere, W. C. Clark and R. E. Munn, eds. Cambridge, England: Cambridge University Press. Clark, W. C., and R. E. Munn, eds. 1986. Sustainable Development of the Biosphere. Cambridge, England: Cambridge University Press. Crosson, P. 1986. Agricultural development Looking to the future. Pp. 104 136 in Sustainable Development of the Biosphere, W. C. Clark and R. E. Munn, eds. Cambridge, England: Cambridge University Press. Crutzen, P. J., and 1: E. Graedel. 1986. Ike role of atmospheric chemistry in environment- development interactions. Pp. 21~250 in Sustainable Development of the Biosphere, W. C. Clark and R. E. Munn, eds. Cambridge, England: Cambridge University Press. Darmstadter, J., L. W. Ayres, W. C. Clark, P. Crosson, P. J. Crutzen, T. E. Graedel, R. McGill, J. F. Richards, and J. A. Tarr. 1987. Impacts of World Development on Selected Characteristics of the Atmosphere: An Integrative Approach. ORNL/Sub/86-22033tl. Oak Ridge, Tenn.: Oak Ridge National Laboratory. Dickinson, R. E. 1986. Impact of human activities on climate- A framework. Pp. 252-289 in Sustainable Development of the Biosphere, Vat C. Clark and R. E. Munn, eds. Cambridge, England: Cambridge University Press. Douglas, I. 1976. Urban hydrology. Geographical Journal 14265-72. Environment '82 Committee. 1982. Acidification today and tomorrow. Translated by S. Harper. Swedish Ministry of Agriculture, Stockholm. Grotch, S. ~ 1988. Regional Intercomparisons of General Circulation Model Predictions and Historical Climate Data. DOE/NBB-0084. Washington, D.C.: Office of Energy Research, U.S. Department of Energy. Harnoz, ~ 1988. The Role of Soils in Sustainable Development of the Biosphere. Laxenburg, Austria: International Institute for Applied Systems Analysis. Likens, G. E., ~ J. Butler, J. N. Galloway, and R. F. Wright. 1979. Acid rain. Scientific American 241~4~:43-51. Lovelock, J. E. 1986. Geophysiology: A new look at earth science. Bulletin of the American Meteorological Society 67: 39~397. Luther, F. M. 1985. Projecting the climatic effects of increasing carbon dioxide: Volume summary. In Projecting the Climatic Effects of Increasing Carbon Dioxide. DOE/ER- 0237. Washington, D.C.: Office of Energy Research, U.S. Department of Energy. Meyer, W., and B. L" lUrner, eds. 1989. The Earth as Transformed by Human Action. Cambridge, England: Cambridge University Press. National Research Council. 1979. Alternative Energy Demand Futures. Committee on Nuclear and Alternative Energy Systems. Washington, D.C.: National-Academy of Sciences. National Research Council. 1981. Atmosphere-Biosphere Interactions: Towards a Better Understanding of Ecological Consequences of Fossil Fuel Combustion. Washington, D.C.: National Academy Press. Ravetz, J. R. 1986. Usable knowledge, usable ignorance: Incomplete science with policy implications. Pp. 415~32 in Sustainable Development of the Biosphere, W. C. Clark and R. E. Munn, eds. Cambridge, England: Cambridge University Press. Volz, A, and D. Kley. 1988. Evaluation of the Montsouris series of ozone measurements made in the nineteenth century. Nature 332: 24~242. Wuebbles, D. J., and J. Edmonds. 1988. A Primer on Greenhouse Gases. DOE/NBB-0083. Washington, D.C.: Office of Energy Research, U.S. Department of Energy.

Energy: Production, Consumption, and Consequences. 1990. Pp. 111-142. Washington, D.C: National Academy Press. The Automobile and the Atmosphere JOHN W. SHILLER Many of the necessities and comforts of our daily life are made possible by motor vehicles, which contribute not only to the creation and delivery of the essentials of life but also to the privilege of expanded personal mobility. In urban areas both the favorable and the unfavorable environmental aspects of the energy sources selected to operate motor vehicles are important determinants of our quality of life. The urban environmental factors associated with motor vehicle use are the subject of this chapter. As with any business, the motor vehicle industry must be responsive to both customer and societal needs. For example, the balance of customer needs and expectations, resource availability, and governmental and soci- etal requirements of today's transportation system represents the dynamic culmination of many competing innovations and options that have been selected and tested with the passage of time. From many points of view, our current motor vehicle transportation need represents a good balance of values, including cost efficiency, safety, environmental needs, and the direct requirements and desires of vehicle owners. Any energy strategy designed to turn away from petroleum, which supplies more than 97 percent of the energy needed to operate our transportation system, will require a new balance. If this is to occur, the challenge will be to find a balance that continues to provide value to the consumer as well as to society as a whole. 111

112 JOHN ~ SHILLER ALTERNATIVE MOTOR VEHICLE ENERGY STRATEGIES Consideration of a wide variety of energy strategies started at the very beginning of the development of motor vehicles. The early days of the motor vehicle era were characterized by no firm ground rules or guidelines on which to base vehicle design and associated energy strategy. The fathers of the automobile nudged knowledge forward in tentative bits and pieces. Many of their early ideas are recorded in the automotive magazine The Horseless Age, a monthly journal that was devoted to motor vehicle interests during their early development (ca. 1895-1918~. It is interesting to note the rich variety of energy strategies reported, including motor vehicles built to operate on gunpowder, calcium carbide-acetylene, compressed air, ether, compressed springs, carbonic acid-sestalit, electric battery, alcohols, coal gas, crude petroleum, gasoline, and other energy sources (Bolt, 1980; Ha- gen, 1977; Ingersoll, 1895, 1897a-e; Staner, 1905~. During the development of these ideas and with advancements in both engine technology and fuels, the advantages of liquid fuels slowly became apparent in such important practical considerations as driving range, convenience, safety, and cost. The Energy Storage Qualities of Liquid Fuels As an illustration of one of the advantages of liquid fuels, Figure 1 shows that the specific energy of liquid fuels far exceeds that of alternative gas, solid, or electric fuel systems. Specific energy is defined as the energy available to drive a motor vehicle per unit mass of fuel, fuel container, and fuel transfer equipment. The figure is approximate because the relative ranking of energy strategies, in terms of specific energy, depends not only on the state of technological development but also on other considerations such as safety and cost. A low value for specific energy means a more limited driving range before the need to refuel, which can be offset to some extent if one is willing to use a more massive on-board energy storage system (limited by vehicle design, function, and size) or if less conservative safety factors are used (such as operating closer to the rated bursting pressure of compressed gas tanks). In Figure 1, gasoline is used as the basis for comparison by normalizing its specific energy to 1.0. For current gasoline vehicles, the specific energy is the amount of energy available to drive the vehicle per unit mass of the fuel, fuel container, and transfer equipment, such as a common fuel tank and pump. For electric vehicles, it is the amount of energy per unit mass of the battery and electrical cables. Thus, the storage mass effectively is constant across all scenarios in the figure. Because specific energy relates to available driving energy, typical engine and power train conversion efficiencies (useful mechanical work output divided by energy input) have

THE AUTOMOBILE AND THE ATMOSPHERE Diesel Gasoline Liquefied Petroleum Gas Ethanol Methanol H2 (Hydride) Compressed Natural Gas NaS Battery H2 (compressed) Coal/Steam Lead-Acid Battery Compressed Air 113 .................. -.-.- -.-.- -.- - ' ' ' ' ' ' - - ' ' ' ' ' ' - - '' ' "' ' ""' 1 -. - ......... ..... ........ . ... .... ............. . 1 , , Liquid Gas · Solid Electric I I 1 1 0 0.2 0.4 0.6 0.8 1 1.2 SPECIFIC ENERGY FIGURE 1 Specific energy for venous alternative energy systems. In this comparison the relative value for gasoline is set to 1.0. been included. The available energy of the stored fuel is based on published heat of combustion data (not including the heat of vaporization of water) or its equivalent (Chemical Rubber Company, 1983~. The energy required to manufacture each fuel or energy source is not included. This figure shows that the amount of useful mechanical energy available to operate current-technology motor vehicles, at a combined fixed mass of fuel and its on-board storage system (relative to gasoline), varies widely for these different energy strategies and is greatest for liquid fuels. The alternative energy form closest in performance to petroleum-based fuels is the alcohol fuels: ethanol and methanol. If fluctuations in crude oil prices or availability and changes in U.S. willingness to depend on imported oil require a move away from petroleum, alcohol fuels are an alternative worth considering. Another reason for interest in alcohol fuels is their possible (but not proven) potential for emission reductions that may help attain Clean Air Act air quality standards. The promotion of alcohol fuels is not new; and with one exception (Brazil), alcohol has failed to gain significant market share. In 1925, Henry Ford predicted that the fuel of the future would come from agricultural products. Consistent with that prediction, in 1927 the first Ford Model A vehicles that were manufactured came equipped with a manually adjustable

114 JOHN ~ SHILLER carburetor specifically intended to give the vehicle owner a choice of using either gasoline or ethanol. Idday, about 60 years later, the adjustment to accommodate alcohol fuels, gasoline, or any mixture of the two is accomplished automatically with Ford's experimental flexible-fuel vehicle (Nichols, 1986~. Fuel flexibility reduces the risk of investing in a vehicle capable of running on alcohol because a switch to gasoline can be made easily at any time. Beginning in 1926 and continuing throughout the 1930s, William Jay Hale, a chemist at Dow Chemical Company, lauded the benefits of alcohol fuel and started the "power alcohol" promotional movement in the United States. Other countries were also involved in alcohol motor fuels. Brazil passed a law in 1931 requiring gasoline importers to mix their fuel with domestic ethanol made from sugar cane and in 1979 launched an extensive program for vehicles fueled by neat (100 percent) ethanol in addition to its continuing program on alcohol-gasoline blends (Bernton, 1979; Szwarc and Branco, 1987~. Other countries, including Chile, China, Czechoslovakia, Germany, Hungary, Sweden, Austria, France, Italy, and Poland, also had some form of alcohol-gasoline blending program after 1930 (Bernton, 1979~. History suggests that the optimum energy strategy for motor vehicle operation will continue to evolve as long as customer and societal values, resource availability or cost, and technological knowledge are subject to change. To date, the relatively low cost of gasoline and diesel fuel and their desirable inherent advantages have preempted the use of alternative energy systems. DOMINANCE OF GASOLINE AND DIESEL FUEL Virtually all the oil produced in the 1800s was used to meet the demand for kerosene (a good substitute for whale lamp oil), lubricants, and wax. Gasoline was considered a nuisance and was run off into rivers or flared because no significant market for the product existed until the development of the internal combustion engine in the early 1900s. With the growth of the nation, total U.S. energy consumption reached nearly 77 quads (quadrillion Btu) in 1987 (U.S. Energy Information Ad- ministration, 1988~. Figure 2 shows that from 1955 to 1986, petroleum was the largest and the fastest growing segment of total energy consump- tion. In 1986, 43 percent of total U.S. energy consumed was supplied by petroleum. Highway vehicles consumed nearly 50 percent of the petroleum or 21 percent of the total energy used in the United States, as shown in Figure 3. U.S. domestic production of petroleum has not kept pace with consumption, as shown in Figure 4. There are many reasons for the dominance of oil in the U.S. trans- portation sector. For most of our history, oil has been a relatively cheap

THE AUTOMOBILE AND THE ATMOSPHERE 1 986 1 985 1 980 1 975 1 970 1 965 1 960 1955 ~~ Hi/////// ] ~ _. 0 10 20 30 40 50 60 70 80 ENERGY CONSUMPTION (quads) Petroleum ~ Natural Gas ~ Nuclear Coal ~ Hydroelectric FIGURE 2 U.S. total energy consumption by source. 115 energy resource of widespread availability. Oil has a high energy density, which provides good vehicle driving range (see Figure 1~. In addition, significantly improved combustion properties of petroleum-based fuels re- sulted from the advancement of fuel engine technology. The handling, storage, and use of petroleum-based fuels also have been optimized to the point that these procedures are both convenient and relatively safe. The internal combustion engine became the standard because of its inherent energy efficiency. DEVELOPMENT OF EMISSION CONTROL FOR CONVENTIONAL VEHICLES Because motor vehicles burn hydrocarbons using air as the source of oxygen, they emit a variety of combustion products, including carbon dioxide (CO2), carbon monoxide (CO), unburned hydrocarbons (denoted HC), oxides of nitrogen (mainly NO with small amounts of NO2, denoted NO=), and other possible trace gases. Further, HC emissions can originate from fuel refining, transportation, and distribution. The emission of some of these gases (e.g., H2O) has been considered to be of no consequence, while others (e.g., CO) have been recognized as requiring control when concentrations exceed public health-based air quality standards. Until recently the production of CO2 by fossil fuel combustion was thought to

116 JOHN YE SHILLER 1 986 1975 1 970 I 1 1 1 1 1 1 1 1 1 0 10 20 30 % of Total Energy % of Petroleum 40 50 60 70 80 90 100 ENERGY USE (percent) FIGURE 3 Percent of petroleum and total energy use by U.S. highway vehicles. 1 1 960 ~ , _-, :.::.:: Consumed :~:~:~:. Produced 0 10 20 30 40 PETROLEUM (quads) FIGURE 4 Total U.S. petroleum consumption and domestic production.

THE AUTOMOBILE AND THE ATMOSPHERE 117 be environmentally innocuous, but fossil-based energy use from all sources over many decades has noticeably elevated the global concentration of this gas and raised the issue of the greenhouse effect (Helm and Schneider, in this volume; Schelling, in this volume). As vehicle use and density have increased in recent decades, the concentration of motor vehicle combustion gases increased to the point that the need for emission controls has been recognized and corrective action has been taken. Ozone Formation Before discussing ozone control strategies, it is important to under- stand the complexity of ozone generation. Ozone (03) is not emitted directly as a pollutant. Rather, it is formed by a variety of photochemical reactions among precursor pollutants as they mix in the troposphere and are irradiated by sunlight. In the troposphere, ozone is formed through the dissociation of NO2 by sunlight to yield an oxygen atom, which then reacts with molecular oxygen (02) to form an O3 molecule. If it is present, NO can react rapidly with O3 to form NO2 and an O2 molecule. A steady-state, or equilibrium, concentration is soon established between 03, NO2, and NO, which in the absence of competing reactions exhibits low O3 levels, which are of no concern. The injection of HC into the atmosphere upsets this equilibrium and allows ozone to accumulate at higher than steady-state concentrations by offering an alternate route for NO oxidation to NO2. NO is emitted from high-temperature combustion sources and is converted to NO2 in the atmosphere. The presence of HC, NOR, and sunlight does not mean that the photochemical reactions will continue indefinitely. Terminal reactions gradually remove NO2 and HC from the reaction mixtures, such that the photochemical cycles slowly come to an end unless fresh NO and HC emissons are injected into the atmosphere. The Development of Emission Controls From the early 1950s when Arie J. Haagen-Smit (1952) discovered the role of emissions in smog formation, a basis was developed upon which many steps have been taken to improve air quality. Haagen-Smit also observed that different organic materials give rise to different levels of photochemical smog, of which ozone is a principal component. These observations were verified in numerous smog chamber experiments, which eventually led the Los Angeles Air Pollution Control District to promulgate Rule 66 (Brunelle et al., 1966), thus making California the first state to legislate emission controls based on the reactivity of hydrocarbons to form smog. The emphasis of Rule 66 was to limit the emissions of highly reactive organic compounds more severely than emissions having relatively

118 JOHN ~ SHILLER low reactivity. Rule 66 was the first effort to apply a practical hydrocarbon reactivity scale to a smog control program. Among the first of the emission control measures taken by the auto- mobile industry was the installation of crankcase blowby control devices to eliminate the emission of hydrocarbons from the crankcase by circulating them back into the engine to be burned. After the introduction of this system in California in 1961, it was incorporated into all the cars distributed nationwide in 1963. From 1968 through model year 1974, tailpipe exhaust emission control was implemented through engine calibration and modifi- cation to achieve more complete combustion along with the enhancement of desirable emission-reducing thermal reactions in the exhaust system. The trend in engine design up to model year 1970 had been toward engines with high compression ratios that required higher octane gasoline, but in 1971, anticipating the introduction of catalytic converters to reduce exhaust emis- sions further, manufacturers began to phase in the hardened valve seats and lower compression ratios needed for engines to burn the unleaded fuel required for proper operation of catalysts (Faith and Atkisson, 1972~. In the 1975 model year, catalytic converters were introduced on motor vehicles, and today more than 80 percent of the in-use passenger car fleet has them (Motor Vehicle Manufacturers Association, 1987~. Regulation and control of hydrocarbon evaporative emissions from motor vehicle carburetors and fuel tanks were first applied in California to 1970 model year vehicles, and federal regulation followed in the next model year. In addition to vehicle evaporative control measures, refueling emission control equipment has been installed at motor fuel service stations in California, Washington, D.C., and parts of a few other states. Motor vehicle inspection and maintenance programs exist in more than half of the states for the identification and repair of vehicles with excessively high emissions. Transportation control measures, vehicle surveillance and recall programs, and environmental impact planning and permitting requirements also have been used to limit emissions. Some of the more recent technological developments that aid in complying with the most stringent emission control requirements include computerized engine control, three-way catalysts (for HC, CO, and NOR control), fuel injection systems (including optimized sequential injection), diagnostics (quicker and better repair reduces emissions), airflow mass sensors, platinum-tip spark plugs, and distributorless ignition. Current passenger car exhaust emission standards require at least a 96 percent reduction in the emission of HC and CO and a 76 percent reduction in NOX emissions compared with precontrol levels, as shown by the first three bars to the left in Figure 5. Light trucks (next three bars) must achieve a 90 percent reduction in the emission of HC and CO and a 67 percent reduction in NOR emissions. Heavy gasoline trucks must also meet !

THE AUTOMOBILE AND THE ATMOSPHERE 100 _ 90 _ 80 _ 70 _ a) a, 60 _ Q - O 50 _ 60 50 40 30 20 10 o a) ·;-:-.-: ........ HC CO NOX HC CO NOX Heavy Trucks Heavy Trucks < 14,000 lb > 14,000 lb <0 ~—~ i ~ ~ ·) l 1 l 1 1 1 ::::. .. . ~ _ _ l l 1 1 1 ,....... :-:-:-:- :::::::: ::::: ::::: ::::: ::::::::: ::::::::: ·:-:-:-:- ::::: ::::: :-:-:-:- ,, , HC CO NOX HC CO NOX Cars Light Trucks < 8,500 lb ~ Emissions L~J Reduction 119 [~ Emissions [~ i Remaining FIGURE 5 Mandated exhaust emission control reductions from precontrol levels (1991 gasoline motor vehicles). stringent emission requirements, although there is more uncertainty in the exact magnitude of their reductions because of fewer available precontrol measurements. ENVIRONMENTAL RESPONSE Ambient Concentration Data Success of pollution abatement activities for both mobile and stationary emission sources is best indicated by the response of the environment through the observed trends in air quality. Based on data from the U.S. Environmental Protection Agency (U.S. EPA, 1988a), the top half of 1bble 1 lists measured trend improvement in the ambient concentration of several atmospheric pollutants. The bottom part of Able 1 lists the results of measurements taken in the Lincoln Tunnel in New York City (Lonneman et al., 1986~. Because the atmospheric pollution added to the atmosphere inside the Lincoln Tunnel is exclusively from motor vehicles, these data indicate that progress has been made in reducing emissions from motor vehicles. In contrast, with the exception of lead, the EPA ambient data

120 TABLE 1 Measured Air Quality Improvements JOHN YK SHINER Podut~t (mfermce) Base Year linprovement lateral I) Improvement (percent) Ambient dam (U.S. EPA, 1988a) Carbon mono~ude (U.S. EPA, 1988a) Estrogen dioxide (U.S. EPA, 1988a) Ozone (U.S. EPA, 1988a) 1977 9 87 197~7 9 32 1977 9 14 1979 7 13 Iincoln Tlmnel air quality data Hydrocarbons odder than meditate 1970 12 74 (Lonneman et al., 1986) Carbon monoxide (Lonneman et al., 1986) Oxides of nitrogen (Lonneman et al., 1986) 1970 12 76 1970 12 62 measures the aggregate trend in reducing emissions from all sources, which makes it impossible to assess the progress made by motor vehicles alone. Lead is an exception because its emission is almost exclusively due to the combustion of leaded gasoline, and therefore the measurement is not affected much by other sources. Lead reductions are seen to be the largest. According to EPA air quality data for 1985-1987 (U.S. EPA, 1988b), a total of 68 areas were reported to have exceeded the ozone air quality standard in at least 1 year of the 3-year period, 6 more areas than previously reported. A total of 62 areas (11 in California) exceeded the ozone air quality standard in the 1984 1986 period (U.S. EPA, 1987c), 14 fewer areas than EPA's earlier compilation of 76 areas for the period 1983-1985. For many areas where ozone is measured, the summer of 1987 was hotter and more conducive to ozone formation than the previous period and is believed to be mainly responsible for the reported increase. A similar statement can be made for 1988. On the other hand, ambient levels of CO continued to decrease as expected, leaving 59 reported areas above the standard for the 2-year period 1986 1987, 6 fewer areas than in the previous 2-year period. Moreover, 23 of the 59 areas were in compliance with the CO air quality standard in 1987. Thus, even with the important

THE AUTOMOBILE AND THE ATMOSPHERE 121 progress achieved, a number of areas of the country still did not attain all air quality standards by the August 1988 congressional deadline. Nai'onal Ambient Air Quality Standards The EPA currently uses 3-year time base data to determine if an area meets the air quality standard for ozone of 0.12 part per million (ppm). That is, the ozone standard is attained `'when the expected number of days per calendar year with maximum hourly concentrations above 0.12 part per million (235 ~g/m3) is equal to or less than 1" (40 CF~ 50.9~. In general, most ozone monitor readings are below the standard. For example, Houston, considered by some to be the second worst area of the country for ozone pollution, had ozone levels below the standard for 99.47 percent of the hours monitored between 1981 and 1985 (American Petroleum Institute, 1987~. The EPA uses the fourth highest daily maximum 1-hour average ozone value per three consecutive years of data to determine compliance with the standard, because the standard allows for an average of one incident above 0.12 ppm each year. The National Ambient Air Quality Standard (NAAQS) for CO is 9 ppm for an 8-hour nonoverlapping average and 35 ppm for a 1-hour average, neither of which is to be exceeded more than once a year. The 9 ppm 8-hour CO standard is the more stringent of the two and is the controlling standard. The nitrogen dioxide NAAQS is based on an annual mean of 100 ~g/m3 (0.053 ppm), not to be exceeded. It is important to note that both emissions and ambient concentrations are subject to considerable variability. For example, the rate of automobile CO emissions has a characteristic pattern in time for most areas; it is highest during rush hours and lowest in the early morning. However, ambient CO concentrations can be highest during periods of lowest emissions if a very low inversion exists (low ambient mixing height), thus limiting dilution. As the sun rises, the mixing height increases. An additional source of variability exists for ozone because it is not emitted directly as a pollutant. As described above, ozone is formed by photochemical reactions of precursor pollutants as they mix in the troposphere. Thus, the concentration of ozone in time and space is a complex function of precursor emissions, weather, topography, and demographics. Location and Sevens of Violations The Los Angeles area, where the air quality standard for ozone is exceeded on more than 140 days a year (for at least 1 hour per day), leads the nation in the frequency of such violations. The next most frequent occurrences of the standard being exceeded, all less than on 40 days a year, were recorded in a few metropolitan areas in Texas and the metropolitan

122 JOHN ~ SHIl J OR areas along the northeast Atlantic coast (New York City and surrounding areas). Los Angeles is also the only area in the nation exceeding the air quality standard for NO2. A combination of poor atmospheric ventilation, frequent sunshine, and bordering mountain barriers increases the potential for higher pollution levels in the Los Angeles area. Forty years of regulatory effort and research and development resulted in tough emission control programs for both stationary and mobile sources. These efforts produced significant, but insufficient, air quality gains toward attaining established, but difficult, goals. Because of the special air quality situation in Los Angeles and the difficulty experienced in achieving air quality standards, the potential environmental benefits of both advanced conventional vehicles as well as alternatively fueled vehicles are being actively studied and will be described below. ALTERNATIVE FUELS AND THEIR EPPL;CTS ON AIR QUALITY Atmospheric emissions from alternatively fueled vehicles typically have a different composition from those produced by gasoline vehicles, which results in different reactivities for ozone generation as well. Many countries are experimenting with alternative power sources for transportation, as shown in Figure 6. Largely as a result of special circumstances (including heavy national subsidies in certain countries), the number of alternatively powered vehicles of all types may be as many as 7 million worldwide (U.S. Department of Energy, 1988~. Countries that have achieved more than 10 percent use of alternative fuels for transportation are Brazil (ethanol, 23 percent) and the Netherlands (liquefied petroleum gas, 11 percent). Although this represents a small percentage (about 1 percent) of total world motor vehicles, much is being learned about the effects of alternative fuel strategies on air quality. The study of air quality effects due to the use of alternative fuels, as well as conventional fuels, has been based on several methods, including atmospheric modeling and smog chamber data using emission rates obtained from vehicles in use, prototype vehicles, and engineering judgment. The results from several studies suggest that under certain conditions, conversion of light-duty vehicles to methanol fuel has some potential to reduce ozone levels (Carter et al., 1986; Chang et al., 1989; Nichols and Norbeck 1985; O'Toole et al., 1983a,b; Pefley et al., 1984; U.S. EPA, 1985, 1988c; Whitten and Hogo, 1983; Whitten et al., 1986~. Much work needs to be done to determine the validity of the assumptions that underlie these conclusions and to understand better vehicle emission performance. The EPA has issued some preliminary estimates of air quality benefit from the use of alternative fuels based on expected emissions and their associated ozone generation potential (U.S. EPA, 1987b). The alternatives

THE AUTOMOBILE AND THE ATMOSPHERE United States Spain Japan Italy Holland Canada Brazil Australia Compressed Natural Gas Ethanol Liquefied Petroleum Gas 1 l l l l l l 0 500 1000 1500 VEHICLES (thousands) FIGURE 6 Number of vehicles fueled by alternative energy systems. 123 2000 2500 3000 addressed include gasoline blends with 10 percent ethanol (gasohol), 5 percent methanol, and 11 percent methyl tertiary butyl ether (MTBE), as well as compressed natural gas (CNG) and various methanol fuels (100 percent methanol, referred to as M100, and a mixture of 85 percent methanol and 15 percent gasoline, known as Mew. Because gasoline blends contain more oxygen than gasoline alone, they contribute to a leaner combustion process for vehicles with limited or no self-compensating ability to adjust to changes in fuel oxygen. Obviously, leaner combustion also could be accomplished with engine calibration. Therefore, blends are a method to lean-out the vehicle fleet after the vehicles are built. Blends might be used at high-altitude locations where the vehicle fleet may be running richer than at low-altitude locations. Thus, under certain circumstances, the use of blends leads to lower emissions of hydrocarbons (1-9 percent) and carbon monoxide (1~30 percent) at the expense of some increase in NOR emissions (4~6 percent). Based on this fact, Colorado started a special blend program to reduce wintertime carbon monoxide levels in the mile-high city of Denver in December 1987 (Miron et al., 1986; Regulation No. 13, 1987) and is in the process of evaluating its effectiveness. However, blends are much less effective for the newer- technology vehicles having adaptive fuel metering systems with nonvolatile computer memory and exhaust gas oxygen feedback control. These systems automatically adjust to the amount of oxygen in the exhaust gas in order to achieve optimal performance from the three-way catalyst. In the future, blends are likely to be used to reduce emissions only in special situations

124 JOHN ~ SHILLER as long as they continue to be effective (depending on the number of older vehicles still in service). Vehicles operating on compressed natural gas, and emitting mostly methane, would also be expected to have low ozone generation potential because methane has very low reactivity and generates essentially no ozone. However, CNG vehicles do produce some reactive hydrocarbons during the combustion process and therefore emit some hydrocarbons other than methane. In addition, these vehicles can have higher NOx emissions because of the relatively high name temperature (U.S. EPA, 1987a). Methanol as a Molor Fuel The potential air quality benefits of neat (M100) or near neat (M85) methanol are related to the fact that the organic emissions from vehicles designed to operate on methanol primarily consist of unburned methanol, which has significantly lower photochemical reactivity for ozone formation than typical hydrocarbon emissions from gasoline-fueled vehicles. livo requirements are necessary before a beneficial effect on air quality can be realized: (1) emissions of other coemitted substances must not significantly offset the lower reactivity of primary emissions and (2) the total mass of emissions compared with conventional vehicles should not be high enough to offset the benefit of lower reactivity. Methanol-fueled vehicles emit formaldehyde, produced by the partial oxidation of methanol, as well as nonoxygenated hydrocarbons that are more chemically reactive in the atmosphere than methanol. Catalysts can be used to remove most of the formaldehyde before it is emitted. However, vehicle data suggest that it does not appear possible, with current technology, to control continuously the more reactive formaldehyde emissions from methanol vehicles to the low level typical of gasoline vehicles over their expected lifetime (Nichols et al., 1988~. It is simply expecting too much for a catalyst to maintain nearly 100 percent efficiency over vehicle lifetime. Another issue of particular importance is the failure of catalysts and the occasions on which catalysts are removed, because formaldehyde is both an eye and nose irritant. It is also not yet clear what the total mass of emissions will be under typical in- use conditions during the life of a methanol vehicle. More developmental work is essential to understand better the potential of methanol fuel and to define vehicle emission limits accurately. The recommended methanol fuel specification of Ford Motor Com- pany is M85. Several safety-related considerations prompt this specification: the 15 percent gasoline provides flame luminosity (a pure methanol flame is invisible in daylight); it makes the vapor above the liquid in the fuel tank too rich to ignite at normal operating temperatures; and it acts as

THE AUTOMOBIl:-F AND THE ATMOSPHERE . .E 7o F Formaldehyde , ~ ,.:.;::. _ ,... 4,000 50,000 Catalyst Age (equivalent miles) 2000 - 1 600 - - c~ ' 1 200 of o co cn LL 800 _ 400 _ - O- Catalyst Age (equivalent miles) 125 FUEL BLEND O M85 ~ M100 :-:-:-:-:-:- :-:-:-:-:-:-: ............. _ , .......... ............. .. ...... :-:-:-:-:-:-: :-:-:-:-:-:- ·:-:-:-:-:-: :-:-:-:-:-:- ·:-:-:-:-:-: :-:-:-:-:-: ............ .....~:. 1 rut ...2....... :-:-:-:-:-:-: ·:-:-:-:-:-:- :-:-:-:-:-:-: ·:-:-:-:-:-:- , . ........ ·:~:.:.:~:.2: ...~......... .:::.::... ............... ....... - - ............... ·.2...~.. ....... . NONOXYGENATED TO rAL HYDROCARBON HYDROCARBON NONOXYGENATED TOTAL HYDROCARBON HYDROCARBON 4,000 50,000 FIGURE 7 Organic emissions, from a flexibly fueled Ford Escort using M85 and M100, as a function of catalyst age. Even though M100 contains no nonoxygenated hydrocarbons, its emissions are comparable with M85. an odorant that discourages ingestion. In addition, the 15 percent gaso- line provides the vapor pressure necessary to cold start an engine at low ambient temperatures and may improve engine durability. Theoretically, M100 should be a better fuel than M85 from an environmental stand- point because M100 does not contain any nonoxygenated hydrocarbons to emit. However, preliminary emissions testing of current-technology flexible- fuel vehicles equipped with laboratory-aged exhaust catalysts (simulating 4,000 and 50,000 miles of driving) indicates somewhat higher emissions of formaldehyde and methanol with M100 than with M85. In addition, as shown in Figure 7, nonoxygenated hydrocarbon emissions with M100 fuel were comparable with those of M85 (Nichols et al., 1988), even though M100 contains no nonoxygenated hydrocarbons. This is because complicat- ing factors such as hydrocarbon formation in the combustion chamber from fuel and oil (with both M85 and M100) and harder starting with M100, cause nonoxygenated hydrocarbons to be emitted out of proportion to their presence in the fuel (Gold and Moulis, 1988), at least for vehicles using current technology. Thus, it is not clear yet whether technological advances will make one fuel specification better than the other from an environmental standpoint without compromising important vehicle operating and safety requirements. For example, in one advanced-concept methanol prototype vehicle, a Toyota lean-burn system with swirl control, the average low mileage exhaust hy- drocarbons and formaldehyde were nearly equivalent with M85 and M100

126 JOHN ~ SHILLER (Murrell and Piotrowski, 1987~. However, higher hydrocarbon levels in some tests with M100 may have resulted from difficulties experienced in starting. Evaporative emissions, although below gasoline vehicle evapo- rative standards, were higher on M85 than M100. Until additional data indicate the contrary, M85 and M100 can be considered nearly equivalent from an environmental standpoint. In terms of other important aspects listed earlier, M85 is preferred. Computer Simulation Saldies (Single Day) Over the past few years, numerous computer simulation studies of atmospheric chemistry have been undertaken to model the impact of mo- tor vehicle methanol fuel substitution (e.g., Chang et al., 1989; Nichols and Norbeck, 1985; O'Toole et al., 1983a,b; Pefley et al., 1984; Whitten and Hogo, 1983; Whitten et al., 1986~. Results from these studies are summarized in Table 2. These simulations were based mainly on "trajec- tory models," such as the city-specific empirical kinetic modeling approach (U.S. EPA, 1977), and thus share the characteristic that a time-dependent emissions profile, called a trajectory, is specified as input. A typical simulation might be undertaken to seek the upper limit for ozone reduction, assuming that gasoline vehicles have been completely replaced by methanol vehicles. Because actual motor fleet turnover (i.e., new vehicles completely replacing old) would occur over some characteristic period of time (up to several decades), such "all-methanol" scenarios are somewhat academic. Details on the substitution assumptions are listed in Table 2. Another important input parameter that must be specified for a model is the total mass of the emissions. As noted, this quantity is not well understood, and therefore an equal-carbon basis is usually assumed. That is, because a methanol molecule contains one carbon atom, a number of methanol molecules equal to the number of carbon atoms contained in the average exhaust and evaporative molecule is used (see Figure 8~. Therefore, an all-methanol trajectory would be one in which each vehicle releases an amount of methanol exhaust and evaporative mixture that contains the same number of carbon atoms as would be emitted if gasoline vehicles were used instead. In so doing, the total mass of emitted carbon presumably is representative of current transportation load, and the effects of the differing exhaust chemicals can be assessed. Three all-methanol simulations are listed in Table 2. One-day modeling simulations are highly sensitive to input variables such as the ratio of reactive hydrocarbons to NO=, ozone concentration aloft, initial pollutant concentrations, and hydrocarbon and NOX emissions along the trajectory. Because of these sensitivities, and because of the

127 ~ ~: o ~ o CQ ;Z - , £ . ~ ~ a, gl) ~: 4.) C ~ ~o 'cn U. '6 ~ ~ _ C~ ._ - ~Q bO ._ - o ; - C~ bO ._ U) C~ o :t a, _ '_ 4, ° o C: C~ ~ — ~ o~ c~ ~ ~ ~ ~ O e~ ~ — ~ ~ c~ ~ ~ C~ _ _ _ ~ _ _ C~ _ — o o Q Q O O O O O O O 0 0 0 0 0 0 ~n ~ _ _ _ _ _ _ 88 °888 _ _ _ _ _ o o o o o o o o ~ o o o o o o C~ C~ _ o o _ o o o o o o o o o o o o O O O O O O d ~ ~ O O O O ~ ~ _ _ C~ _ _ _ _ _ Q O =` O O o `0 ~ oo o O o o o l~ o C~ oo ~ ~ ~ ~ C~ O o~ O — O _ _ _ _ _ 4D O ~ _` ~ 1~^~ ~, :10^~ ,~,, 1- ~ ~ Z ~ e^~ 6 ~ == ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ g ~ aq U~ ~ ;- ~ ° 8 E _ ~ ~ - ~ - ~ ~ ~ ~ ~ . 3 .~ o

128 ~ _ o :^ ~ ~ o U' z Zen :~: ., ~ 4.) o so ~ ·~ ~ ·~: ~ _ _ _% 4) ~ _ ._ ~: o m - :: ~ C) ~ ~ C~ O C~ ~ ~ t_ _ ~ 88888 888 _ _ _ _ _ _ _ ~4 C~ o o ~ C~ o o _ ~ _ o o o o o o o o CO ~ o o C~ _ o o _ _ ~o o o ~ ~ o o C~ ~ ~ ~ - 1 _ ~ ~ ,.c ~ o ~ ~ 04 ~ o = ~ .= ~ o O _ ~, o C~ o~ oo o~ ~_ o o o o - oo ~: ° 6 '= o 8'§~.= ~ 73 — ~ o 3 E ' E ~ ~ ~ D C o c, ~ ~ 3 ~ C -.° C ~ ~ ~ i ~ ~ ~ ~ ~ _ ~ E~ ~ D °o ~ ~ &~ 0" ~ !oo~ _ ,_ ~ ~ o 't, O ~ ~ ~ E E-- Y E .o, ~ ' 8 ) o _' o ~ ' - ' ~ ~ _ ~ ~ O >` e . ~ ~ O ~ ~ ~ .0 ~ ~ ~ ~ ~ o ~ ' 8 z @]

THE AUTOMOBILE AND THE ATMOSPHERE Representative emissions molecule, gasoline vehicle H H C H H—C H H C H carbon basis / H- C H H—C—H H 5 carbon atoms in one molecule 129 H H C—OH H H H C—OH > H H H C—OH H H H C—OH H H H C—OH H 5 one-carbon methanol molecules FIGURE 8 Equivalent carbon-for carbon basis for one typical emission molecule from a gasoline vehicle. Note that both evaporative and exhaust emissions are composed of many different types of molecules; typically, equal carbon basis calculations are camed out for each chemical species. modeling assumptions discussed here and listed in Table 2, one must be cautious when interpreting simulation results. The results of trajectory model simulations such as those found in Table 2 have been used to estimate the amount of ozone produced relative to gasoline hydrocarbon emissions, assuming an equal-carbon emission basis (Chang et al., 1989; EPA, 1987a; Gold and Moulis, 1988; Shiller, 1987~. For most cases, the effect of a switch to methanol, including the effect of emitting some formaldehyde, was investigated under the assumption that the NOX emission rate from methanol vehicles is the same as that from gasoline vehicles. The most recent calculated average reactivity factors for 20 cities were about 0.58 for methanol and 2.15 for formaldehyde, compared with 1.0 for conventional hydrocarbon ozone generation potential at equal carbon emissions. These factors suggest that formaldehyde (carbon atom for carbon atom) is more than twice as efficient as hydrocarbons in generating ozone, whereas methanol is less than about 60 percent as efficient as hydrocarbons. Previously, the difference between the calculated reactivity factors for methanol and formaldehyde were larger (U.S. EPA,

130 JOHN W: SHIl I OR 1986a, 0.02 and 2.95; Nichols and Norbeck, 1985, 0.38 and 4.8; and U.S. EPA, 1987a, 0.43 and 4.83~. For a few of these simulations, reductions in NOR emissions were also evaluated, which caused either an increase or a decrease in projected ozone levels. This, for example, is illustrated by comparing different impacts of NOR reductions on ozone projections for Los Angeles (cases 2 and 3 of the Jet Propulsion Laboratory study in Table 2) and Philadelphia (cases 1 and 2~. The Philadelphia study suggests that NOR control can be counterproductive to ozone control, whereas results for Los Angeles suggest that the opposite can be true. In general, the greater the impact of NOR control on reducing ozone (greater impact with increasing ambient HC/NO~ ratio), the less important reactivity differences become (Dodge, 1984) and the smaller is the projected impact of methanol-fueled vehicles. Consistent with this interpretation, the Philadelphia simulation seems to be more sensitive to reactivity changes than the Los Angeles simulation. Of course, projected results depend on the accuracy of the chemical mechanism and emission inventory assumptions used in the model (which determines the local HC/NOX ratio) and the particular wind flow field selected for the simulation. Further study is necessary to determine how general the results are for each city and their implications for ozone control planning. In addition, because ozone builds up more slowly with methanol, it is necessary to determine the impact of methanol vehicles under multiday episodes of stagnant air, which provide more time for ozone formation. Smog chamber studies of simulated urban air (Carter et al., 1986) confirm the results of single-day simulations that lower peak ozone levels occur for the first day for the case in which one-third of the total vehicle emissions are from methanol-fueled engines. However, the ozone levels in the chamber rose over the course of the second and third days, becoming similar to those observed in the control case of 100 percent gasoline engine emissions. This suggests that even if one-third of the emissions from gasoline vehicles is replaced with methanol emissions, little or no benefit will be achieved for multiday ozone episodes. Because of the computer model sensitivities noted above, and because smog chamber simulations may be affected by the greater surface-to-volume effects inherent in a chamber, further study is needed to strengthen these findings. At this stage, it is prudent to interpret these results qualitatively, rather than quantitatively, when discussing this matter. Multiday Simulation Shoddies The scope of ozone modeling is being extended to study the impact of methanol-fueled vehicles during multiday episodes. The most noteworthy multiday study, employing a Cray X-MP/48 supercomputer, was conducted

THE AUTOMOBILE AND THE ATMOSPHERE 131 at Carnegie Mellon University under the sponsorship of the California Air Resources Board (CARE). Some of the results have been released (California Air Resources Board, 1988; Computers in Science, 1988; Harris et al., 1988~. A sequential modeling approach was used to guide the study effort, starting with chemical kinetic modeling, then moving to trajectory analysis, and finishing with airshed modeling of the entire California South Coast Air Basin. For the final airshed modeling analyses, 13 control strategy scenarios reflecting projected emission assumptions and expected conditions in the basin for the year 2000 have been run. Vehicle emission rate assumptions were supplied by CARE (1988) for six cases (three with conventional fuels and three with methanol). This matrix is shown in Table 3. Emission rates for the first case of conventionally fueled vehicles were based on CARB's EMFAC7C data (CARB, 1986) and reflect all post-1987 regulations that have been adopted or implemented. The second case reflects additional conventional control measures that are being considered for adoption in the l990s (e.g., 0.25 g/mi nonmethane hydrocarbons [NMHC] beginning in 1995 and 0.2 g/mi NOD beginning in 1997) if they can be shown to be feasible and desirable. This was simulated by adjusting the EMFAC7C data by the ratio of proposed standards to existing standards. The third case for advanced conventional vehicles assumes that all motor vehicles will be designed to meet the lowest foreseeable emission standards and will perform at the lowest end of the in-use emission range that might be possible with further technical development. The next three cases are methanol scenarios. Each assumes that 100 percent of the 1990 2000 model year highway vehicles are built to run on methanol, with a similar percent conversion for off-road mobile sources and retail gasoline distribution. The scenario labeled case four explores the effect of using M85 by assuming a 15 mg/mi emission rate of formaldehyde for cars and light trucks; a contribution for nonoxygenated hydrocarbon emissions was also included. Case 5 assumes the same (case 2) vehicle fueled with M100, carbon for carbon, a comparatively high formaldehyde emission rate of 55 mg/mi, and zero nonoxygenated hydrocarbon emissions. Case 6 is identical to case 5 except that a lower value for formaldehyde emission, 15 mg/mi, is assumed. Note that both of these M100 cases assume no nonoxygenated hydrocarbon emissions, which is contradictory to the M100 data reported in Figure 7. In general, some of these emission rate assumptions are optimistic and have not been shown to be feasible; the results of these simulations appear in liable 4. Note that a seventh case also appears in this table. Case 1 was selected as the base case by which we can compare the reductions predicted by the other simulations. Comparisons are reported in terms of the percent of the unachieved ozone reduction required to bring

132 TABLE 3 Muliiday Study Simulation Manx JOHN ~ SHINER FUEL TYPE .Gasoline Methanol Case 1: V Current conventional vehicles E (nodal fleet turnover) . H Case 4: I M85 @ 100 percent substitutions C 15 mg/mi formaldehyde, ? HC Case 2: L 199~2000 model year vehicles Case S: E (additional regulations, M100 @ 100 percent subsiituiiona nonnal fleet turnover to 2000) 55 mg/mi fonnaldebyde, no HC T Case 6: M100 @ 100 percent substitution Y 15 mg/mi formaldehyde, no HC P Case 3: E Advanced vehicles (technological break- through) and 100 percent substitution aComplete replacement of methanol vehicles for both highway and off-road mobile vehicles and re- placement of methanol for gasoline in die retail fuel distribution system. the base case into compliance (100 percent) and appear in Figure 9. For the base case scenario, the maximum peak ozone level was calculated to be 0.27 ppm. Because 0.12 ppm is the value specified by the ozone standard, a 57 percent reduction in ozone level would be required to achieve compliance; this is represented by the top bar of Figure 9 plotted at 100 percent and 0.27 ppm. From Table 4, we note that cases 2 and 3 were found to reduce ozone levels by 3 and 9 percent, respectively; this corresponds, respectively, to 5 and 16 percent of the reduction required to reach the 0.12 ppm level (greater by the ratio of 100/574. These two cases are indicated by bars 2 and 3 of Figure 9. The next three bars indicate percent of achievement for the three methanol cases. Note that the best methanol case, M100 with low formaldehyde, achieves only 29 percent of the reduction required to reach 0.12 ppm. These preliminary results of this study are consistent with previous research; however, the reader is again reminded that these

133 ~ _ ~ As_ "o ~ o Vat :Z U. o ~ C ID ~ — U. .~ _ To - to · - m As 43 _ _ 4) ~ ~ ~ TIC ~ - 9D C) Us 0 cot ~ cry 0 lo ~ ~ _ ~ ~ 7; ' C} ~ C ~ i ~ 8 - '0; 0 3 ~ 3 o ~ 8 8 ~ _ _ ~ ~ 2 o - U, oo o ~ C? ~ — Cat cot 'S c`4 as Q DO .~ .E ~ _ _ ~ ~ _ —4 — 2 m ~ A ~ 6 ·Uc ~ ° ~ o _ c, 3 o s ~ ~, o ;~. o ~ ~ U. o. o ·V} ~ ·6 =.' ~ .~ =~.S _ ~ .O ~, 6 oo `: s ~o ~ ~ ~_ a, ~ . o _ P" ~ ~ 6 o ~ ~ ~ aq U~ o 6.8 3 o, ~ ~ _ ~ C ~ ~ o '-# ~ ¢,' ~ ~ ~ ~ ~ 3 ~ e =~ ,= O C ~ ~ o ~ O ~ ~ 0, C

134 Conventional Vehicle Base Case 1900-2000 Model Year Vehicles Maximum Advanced Conventional Vehicle MB5, Low F M100, High F M100, Low F No Mobile HC No Mobile NOX No Mobile HC and NO 0.12 x No Man-Made Sources JOHN ~ SHILLER Peak O3 Level (ppm) 0.27 ~ 1 ;;;;.;;;;;;;; _- .. 2 . ~ ,,.,., ,,,, ......... , ..................... ~ _ 1 1 1 1 1 1 1 1 1 1 2 3 4 ~ m 5 ~ of 6 'L 7 0 20 40 60 80 100 UNACHIEVED O3 REDUCTION (% of 0.15 ppm) FIGURE 9 Companson of ozone reduction strategies for the South Coast Air Basin for the year 2000. The solid bars show the percent of unachieved reduction of ozone levels from a base case value of 0.27 ppm to the 0.12 ppm level required by the ozone standard. Case numbers are the same as in Cables 3 and 4 and are described further in the text. These results suggest that no strategy limited to the motor vehicle transportation system alone will be sufflaent to achieve the ozone standard in the Los Angeles area. SOURCE: California Air Resources Board, 1988. findings depend on the plausibility of the emission rate assumptions and initial conditions. The bars below case 6 in Figure 9 show the reductions achieved by eliminating all ozone precursor emissions due to the motor vehicle trans- portation system, that is, emissions from all retail gasoline and diesel distribution centers, all petroleum refining operations, and all motor vehi- cles (both highway and off-road vehicles) except motorcycles. The three cases studied were as follows: the removal of all HC with reduced NOR achieved a reduction of 16 percent, the removal of all NOR with reduced HC achieved 20 percent, and the removal of all HC and NOR (case 7) achieved 24 percent, far short of the 57 percent required. Further, a peak ozone level of 0.139 ppm was estimated for the extreme case of removing all in-basin emissions (all sources) and assuming low-level initial conditions (low pollutant concentrations entering the basin). Even for this case, only 87 percent of the reduction required to achieve the ozone standard was attained, as shown by the bottom bar of the figure. These results suggest that achievement of the ozone standard in the

THE AUTOMOBILE AND THE ATMOSPHERE 135 Los Angeles area will not be possible with any strategy that targets mobile sources alone. It may not be possible even with the removal of all man- made sources. The results further indicate that methanol-fueled vehicles do have some potential to reduce ozone in the Los Angeles area but that the magnitude of the reduction during multiday episodes is likely to be smaller than that of earlier single-day modeling results (see Able 2 for Los Angeles). The emission assumptions for the M85 case are considered to be more realistic (for either the M85 or the M100 case). This suggests a 6 percent ozone benefit (9 percent versus 3 percent, respectively) for methanol-fueled vehicles over conventionally fueled vehicles (with additional controls) may be possible if the assumed optimistic emission rates can be achieved. More work is needed to examine the accuracy of the emission assumptions and the general applicability of these results to other modeling conditions. Public Health and Direct Formaldehyde Emissions For any ozone control strategy, it is vital to consider the possibility of undesirable effects that may arise as a consequence of the control strategy itself. In other words, we must make sure that the "cure is not worse than the disease.'' Thus, it is necessary to evaluate the effect of direct emissions of formaldehyde on public health, particularly in confined spaces such as parking garages. Work is being done to gain a better understanding of this concern. For example, the Health Effects Institute (HEI) has under- taken a study of formaldehyde to complement their earlier study on the health effects of direct methanol emissions (HEI, 1985~. The earlier study indicated that adverse human health effects due to methanol at projected exposure levels associated with a move to methanol fuel are unlikely. In another study (Gold and Moulis, 1988), EPAs proposed methanol vehicle emission standards (U.S. EPA, 1986b) are shown to prevent concentrations of methanol and formaldehyde from reaching toxic levels in most com- monly encountered driving scenarios. However, further analysis of parking garages and other areas with restricted air dilution is needed before all concerns can be fully resolved. As for open and unrestricted spaces, the Carnegie Mellon study (Computers in Science, 1988) showed that 90 percent of atmospheric formaldehyde is created by photooxidation of hydrocarbons and that only 10 percent comes from direct emissions. Thus, a significant increase in average ambient formaldehyde levels with methanol vehicles is unlikely.

136 JOHN ~ SHILLER PUBLIC POLICY STRATEGIES In recognition of the difficult task of achieving standards in the Los An- geles Basin, the South Coast Air Quality Management District (SCAQMD) and the Southern California Association of Governments (SCAG) have announced a three-tier strategy to attain air quality standards over the next 20 years (SCAQMD and SCAG, 1987, 1988~. The plan, which was proposed in June 1988, acknowledges that full-scale implementation and advanced development of known technology will be inadequate to achieve the standards in Los Angeles. Significant technological breakthroughs are required. To illustrate the magnitude of the situation, Figure 10 shows the 1985 emission inventory for both mobile and stationary sources of reactive hydro- carbons and oxides of nitrogen in the South Coast Air Basin (SCAQMD and SCAG, 1987~. Based on estimates by SCAQMD and SCAG, am percent reduction in reactive hydrocarbons from the 1985 inventory would be necessary to attain the air quality standard for ozone. Elimination of highway vehicle hydrocarbon emissions would only provide a 43 percent reduction, far short of the required 79 percent. Therefore, the demands of attaining the ozone standard in Los Angeles would exceed even the most aggressive application of current motor vehicle technology. Note that this conclusion is consistent with the multiday computer simulations described above. The three-tier strategy proposed by SCAQ to reduce emissions to the point where all air quality standards are achieved by 2007-2010 requires a full-scale implementation of known technology (tier 1), significant ad- vancement of known technology (tier 2), and technological breakthroughs (tier 3~. In the second phase of the plan, it is assumed that 40 percent of passenger cars, 70 percent of freight vehicles, and 100 percent of buses will use clean-fuel technologies (Acurex, 1986~. In January 1988, SCAQMD adopted a five-year $30-million Clean Fuels Program, which includes 19 mobile source related demonstration projects for technologies, such as electric vehicles, methanol, and compressed natural gas. Tier 3 assumes full electrification of all motor vehicles and stationary combustion sources. That assumption relies on new technologies such as superconductors and improved electrical storage devices and either the building of new infras- tructures or the elimination of existing infrastructures or both. Tier 3 is clearly beyond current capability. CUSTOMER VALUE AND FUEL STRATEGY If a conversion to alternative fuels is to be successful, the vehicles that use these fuels must represent value to the customer or they will not sell.

THE AUTOMOB LE AND THE ATMOSPHERE 137 Customer surveys, such as the 1987 Automotive Consumer Profile survey of 5,000 members of the driving-age public, show that reliability and quality lead the list of consumer '~wants" (see Figure 11), with safety and service (availability of parts and labor) also high on the list (Power Report, 1988~. About 70 percent or more of the respondents said that these features were "very or somewhat important" in the purchase of their next vehicles. Purchase price, operating cost, and resale value (a factor in operating cost) are also perceived as important. Based on new car buyers' perceptions of how these features would be affected by a switch to methanol fuel (see Figure 12), the perceived benefits of methanol are in areas of relatively low importance to consumers. Vehicles that use alternative fuels must be fully competitive in value with current transportation alternatives. In fact, it could be argued that they must offer an advantage to overcome any concerns a customer might have. This advantage may have to take the form of buyer incentives (at least in the beginning) to induce consumer risk invesunent by compensating for the perceived potential disadvantages associated with the convenience, reputation, and resale of alternative-fuel vehicles in relation to gasoline- fueled vehicles. It is critical that consumer concerns be addressed at the outset. 1 400 1 200 >` a, 1 000 Q CO Hi; In z o In CO 400 Level for Ozone Attainment - 200 HO _ Nonhighway Automobiles Vehicles Trucks ~ Petroleum Operations NOxi O Residential ~ Electric Power Manufacturing O Service- Commerce FIGURE 10 South Coast Air Basin emissions lay source for 1985. In 1985 the ozone produced by precursor sources unrelated to transportation was sufficient to exceed the standard. Level for NO Attainment

138 Reliability Quality Safety Service Comfort Reputation Low Price Fuel Economy Appearance Power Technology Small Size JOHN ~ SHILLER 1 '1 ~:q .:.:.:.:.1 1987-1988 Buyers n Potential New-Car I I Consumers 0 50 1 00 PERCENT SAYING VERY/SOMEWHAT IMPORTANT FIGURE 11 Important factors affecting new-car purchase decisions. From the customer's point of view, the availability of vehicles capable of using alternative energy sources will alone not produce a change in the present transportation system. There must also be a substantial increase in the production capacity for the alternative fuel in question and a distribution network to make such fuel readily available. Therefore, the development of vehicles having "fuel flexibility" seems imperative in any program to introduce a new fuel. CONCLUSIONS Significant advances have been made in reducing emissions from current-technology vehicles. Further reductions are expected as newer vehicles replace old. Even with the progress to date, certain areas of the

THE AUTOMOBILE AND THE ATMOSPHERE 139 country still did not attain all air quality standards by the August 1988 congressional deadline. The Los Angeles area faces the most difficult task to achieve the national air quality standards. Because of the special air quality situation in Los Angeles, the potential environmental benefits of alternative fuels are being studied to determine what can be accomplished. 1b date, the relatively low cost of gasoline and diesel fuel and their other inherent advantages have restricted the use of other fuel sources. The renewed interest in alcohol as a fuel in the United States results mainly from a hope that it may reduce environmentally undesirable emissions. Studies suggest that conversion of light~uty vehicles to methanol fuel has some potential to reduce ozone levels if emission assumptions are correct, but they also suggest that in certain areas it is not possible to achieve the ozone standard with any strategy that targets only the emissions from the motor vehicle transportation system. Much remains to be done to ascertain the validity of the assumptions that underlie the conclusions reached thus far and to gain a better understanding of currently Reliability Quality Safety Service Comfort Reputation Low Price Fuel Economy Appearance Power Technology Small Size ~//.///////~/////////~////////////~ ///// Perception of Methanol Impact Positive Neutral 1 Unknown Negative 1 0 20 30 40 50 60 70 80 90 PERCENT SAYING VERY/SOMEWHAT IMPORTANT FIGURE 12 Consumer perceptions of the effect of methanol-fuel capability on factors important to a new-car purchase.

140 JOHN ~ SHILLER unregulated emissions. The improvements made in conventional technology also need careful evaluation for comparison. Surveys of motor vehicle consumer preferences indicate that the per- ceived benefits of methanol are relatively less important than other vehicle attnbutes. This situation poses an even greater challenge for vehicle man- ufacturers and governmental regulators to deliver these new technology vehicles in a way that provides equal, if not greater, value to the customer. REFERENCES Acurex. 1986. California's Methanol Program, Evaluation Report, Vol. I, Executive Summary. November. Mountain View, Calif.: Acurex Corp. American Petroleum Institute (API). 1987. Ozone Concentration Data Analysis. Washington, D.C.: API. Bolt, J. A. 1980. A Survey of Alcohol as a Motor Fuel. SAE Technical Paper Series, SAE/pt-80/19. Warrendale, Pa.: Society of Automotive Engineers. Brunelle, M. F., J. E. Dickinson, and W. J. Hamming. 1966. Effectiveness of Organic Solvents in Photochemical Smog Formation: Solvent Project, Final Report. Los Angeles County Air Pollution Control District. California Air Resources Board (CARB). 1986. Methodology to Calculate Emission Factors for On-Road Motor Vehicles. CARB Technical Support Division, Emission Inventory Branch, Motor Vehicle Emissions and Projections Section. Sacramento, Calif.: CARB. California Air Resources Board. 1988. Selected draft of Carnegie Mellon University study results released in response to a request for information in preparation for the June 1, 1988, CARB workshop on formaldehyde emission standards. Carter, W. P. lo, R. Atkinson, W. D. Long, L. Lo N. Parker, and M. ~ Dodd. 1986. Effects of Methanol Fuel Substitution on Multi-Day Air Pollution Episodes. University of California for California Air Resources Board, Contract No. A3-125-32. Sacramento, Calif.: CARB. Chang, R. Y., S. J. Rudy, G. Kuntasal, and R. A. Gorse, Jr. 1989. Impact of methanol vehicles on ozone air quality. Atmospheric Environment (In Press). Chemical Rubber Company. 1983. Handbook of Chemistry and Physics, 64th ed. Boca Raton, Fla.: CRC Press. Computers in Science. 1988. Carnegie Mellon University. January-February. Dodge, M. C 1984. Combined effects of organic reactivity and NMHC/NO ~ ratio on photochemical oxidant formation—A modeling study. Atmosphenc Environment 18~8~:1657-1665. Faith, W. L., and ~ A. Atkisson, Jr. 1972. Air Pollution, 2d ed. New York: John Wiley & Sons. Gold, M. D., and C. E. Moulis. 1988. Effects of emission standards on methanol vehicle- related ozone, formaldehyde, and methanol exposure. APCA Paper No. 88-41.4, SDSB/EPA. Pittsburgh, Pa.: Air Pollution Control Association. Haagen-Smit, A. J. 1952. Chemistry and physiology of Los Angeles smog. Industrial and Engineenng Chemistry 44(6):1342. Hagen, D. Lo 1977. Methanol as a fuel: A review with bibliography. SAE Technical Paper Senes, SAE-770792. Warrendale, Pa.: Society of Automotive Engineers. Harris, J. N., A. G. Russell, and J. B. Milford. 1988. Air Quality Implications for Methanol Fuel Utilization. SAE Technical Paper Series, SAE-881198. Wa.~ndale, Pa.: Society of Automotive Engineers. Health Effects Institute (HEI). 1985. Letter to Dr. Bernard Goldstein, Assistant Admin- istrator for Research and Development, EPA, from Thomas P. G~umbley. August 12. Ingersoll, E. P. 1895. The Horseless Age (November).

THE AUTOMOB LE AND THE ATMOSPHERE 141 Ingersoll, E. P. 1897a. The oil famine bugaboo, a gunpowder motor. The Horseless Age II(3~. Ingersoll, E. P. 1897b. Acetylene as a motive agent, motor cabs (electric) in New York. The Horseless Age (February). Ingersoll, E. P. 1897c. Acetylene motors. The Horseless Age (March). Ingersoll, E. P., ed. 1897d. Riker electric Victoria carbonic acid carriage motor, the Worthley steam carriage, new motor (electric) fire (fighting) apparatus, alcohol as a fuel for motors. The Horseless Age (September). Ingersol, E. P. 1897e. Compressed air vehicles of the Pneumatic Carriage Company. The Horseless Age (October). Lonneman, W. A., S. A. Meeks, and R. L. Sella. 1986. Non-methane organic composition in the Lincoln Tunnel. Environmental Science and Technology 20:79~796. Miron, W. L., R. A. Ragazzi, T. W. Hollman, and G. L. Gallagher. 1986. Ethanol- blended fuel as a CO reduction strategy at high altitude. SAE Technical Paper Series, SAE-860530. Warrendale, Pa.: Society of Automotive Engineers. Motor Vehicle Manufacturers Association (MVMA). 1987. MVMA Motor Vehicle Facts and Figures '87. Washington, D.C.: MVMA. Murrell, J. D., and G. K Piotrowski. 1987. Fuel economy and emissions of Toyota T-LCS-M methanol prototype vehicle. SAE Technical Paper Series, SAE-871090. Warrendale, Pa.: Society of Automotive Engineem. Nichols, R. J. 1986. The flexible fuel vehicle: The bridge to methanol. Paper FL-86-103. National Petroleum Refiners Association, Fuels and Lubricants Meeting, Houston, Texas. November. Nichols, R. J., and J. M. Norbeck. 1985. Assessment of emissions from methanol-fueled vehicles: Implications for ozone air quality. APCA Paper No. 85-38.3. Pittsburgh, Pa.: Air Pollution Control Association. Nichols, R. J., E. Clinton, E. T. King, C. S. Smith, and R. J. Moreland. 1988. A view of FFV aldehyde emissions. SAE Future Transportation Technology Conference and Exposition, August 8-11. O'Toole, R., E. Dutzi, R. Gershman, W. Heft, W. Kalema, and D. Maynard. 1983a. California Methanol Assessment, Vol. 1: Summary Report: Jet Propulsion Laboratory and California Institute of Technology, N83-33340. Springfield, Va.: National Technical Information Service. O'Toole, R., E. Dutzi, R. Gershman, W. Heft, W. Kalema, and D. Maynard. 1983b. California Methanol Assessment, Vol. 2: Technical Report: Jet Propulsion Laboratory and California Institute of Technology, N8~33341. Springfield, Va.: National Technical Information Service. Pefley, R. K, B. Pullman, and G. Whitten. 1984. The impact of alcohol fuels on urban air pollution: Methanol photochemistry study. University of Santa Clara and Systems Applications, Inc., for Department of Energy, DOE/CE/50036-1. November. Springfield, Va.: National Technical Information Service. Power Report. 1988. liouble-free, quality, safe cam important to consumers. The Power Report Newsletter (March). Regulation No. 13. 1987. The reduction of carbon monoxide emissions from gasoline powered motor vehicles through the use of oxygenated fuels. Adopted by the Colorado Air Quality Control Commission on July 29, 1987, and submitted to the Colorado Secretary of State for publication in the Colorado Register on July 10, 1987. South Coast Air Quality Management District (SCAQMD) and Southern California Associ- ation of Governments (SCAG). 1987. The Path to Clean Air Attainment Strategies. December. Los Angeles, Calif.: SCAG. South Coast Air Quality Management District (SCAQMD) and Southern California Asso- ciation of Governments (SCAG). 1988. The Path to Clean Air Policy Proposals for the 1988 Air Quality Management Plan Revision. June. Los Angeles, Calif.: SCAG. Shiller, J. W. 1987. The role of alternative fuels in air quality planning: Methanol fuels in light duty vehicles, does it help or hurt? Proceedings, Air Pollution Control Association Specialty Workshop on Post-1987 Ozone Issues, Golden West Chapter, San Francisco, November.

142 JOHN ~ SlIlLLER Staner, H. W., ed. 1905. Alcohol as a fuel for motor cars. The Autocar XIV(482~(January 14~. Szwarc, A., and G. M. Branco. 1987. Automotive emissions The Brazilian control program. SAE Technical Paper Series, SAE-871073. Warrendale, Pa.: Society of Automotive Engineers. U.S. Department of Energy. 1988. Assessment of Costs and Benefits of Flexible and Alternative Fuel Use in the U.S. Transportation Sector, Progress Report One: Context and Analytical Framework, DOE/PE-0080. Washington, D.C.: U.S. Department of Energy. U.S. Energy Information Administration. 1988. Monthly Energy Review (July). Washington, D.C.: U.S. Department of Energy. U.S. Environmental Protection Agency (EPA). 1977. Uses, Limitations and Technical Basis of Procedures for Quantifying Relationships Between Photochemical Oxidants and Precursors. EPA-450/2-M-021a. EPA Research Triangle Park, N.C. U.S. Environmental Protection Agency. 1985. Outdoor Smog Chamber Experiments: Reactivity of Methanol Exhaust. EPA Offlce of Mobile Sources. EPA 460/3-85-009a & b. September. U.S. Environmental Protection Agency. 1986a. Regulatory Support Document, Proposed Organic Emission Standards and Test Procedures for 1988 and Later Methanol Vehicles and Engines. July. U.S. Environmental Protection Agency. 1986b. Emission Standards for Methanol-Fueled Motor Vehicles and Motor Vehicle Engines, FR 51 No. 166. August. U.S. Environmental Protection Agency. 1987a. Air Quality Benefits of Alternative Fuels. Report prepared for the Vice President's disk Force on Alternative Fuels, EPA Offlce of Mobile Sources. July. U.S. Environmental Protection Agency. 1987b. Guidance on Estimating Motor Vehicle Emission Reductions from the Use of Alternate Fuels and Fuel Blends. Draft Technical Report EPA-AA-TSS-PA-87~. EPA Emission Control Technology Division, Ann Arbor, Mich. July. U.S. Environmental Protection Agency. 1987c. Note to correspondents, EPA release of 1986 air quality data. EPA Offlce of Air Quality Planning and Standards. August 27. U.S. Environmental Protection Agency. 1988a. National Air Quality and Emission Trends Report, 1988. EPA450/4-88-001. U.S. Environmental Protection Agency. 1988b. EPA lists areas failing to meet ozone or carbon monoxide standards. EPA press release, Washington, D.C., May 3. U.S. Environmental Protection Agency. 1988c. Guidance on Estimating Motor Vehicle Emission Reductions from the Use of Alternative Fuels and Fuel Blends. EPA-AA- TSS-PA-87~. Ann Arbor, Mich.: EPA Emission Control Technology Division. Whitten, G. Z., and H. Hogo. 1983. Impact of methanol on smog: A preliminary estimate. Systems Applications, Inc., San Rafael, Calif., for ARCO Petroleum Products Co., SAI Publication No. 83044. February. Whitten, G. Z., T. C. Myers, and N. Yonkow. 1986. Photochemical modeling of methanol- use scenarios in Philadelphia. Systems Applications, Inc., San Rafael, Cali£, for EPA Emission Control Technology Division, Ann Arbor, Mich. EPA 460/3-86-001. March.

Next: 3. Evolving Vulnerbilities and Opportunities »
Energy: Production, Consumption, and Consequences Get This Book
×
Buy Paperback | $85.00
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

Energy provides a fresh, multidisciplinary approach to energy analysis. Leading experts from diverse fields examine the evolving structure of our energy system from several perspectives. They explore the changing patterns of supply and demand, offer insights into the forces that are driving the changes, and discuss energy planning strategies that take advantage of such insights.

The book addresses several major issues, including the growing vulnerabilities in the U.S. energy system, the influence of technological change, and the role of electricity in meeting social objectives. The strongest of the book's themes is the growing influence of environmental concerns on the global energy system.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!