Constructed Climates

Figure 4.2: At left, annual pollution emissions from human sources show gradual decreases over the last several decades (data from the U.S. EPA). Pollutants include volatile organic compounds (VOCs), carbon monoxide (CO), ammonia (NH₃), reactive nitrogen (NO_x= NO + NO₂), particulate matter (with condensables) below 10 and 2.5 microns in size (PM₁₀ and PM_2.5), and sulfur dioxide (SO₂). The right plot compares pollutant levels before, during, and after the 1996 Atlanta Summer Olympics, during which authorities reduced traffic by 22.5% (after Friedman et al. 2001). Ozone, carbon monoxide, and PM₁₀ (particulate matter below 10 microns diameter) levels were reduced significantly during the Olympics. NAAQS stands for the National Ambient Air Quality Standards set by the U.S. Environmental Protection Agency.

The left graph in Figure 4.2 shows, on a national scale, that Americans put quite a bit of pollutants into the atmosphere. The EPA data separate these numbers into many source categories, sometimes humorously so, for example, with around 85% of NH₃ and PM₁₀ emissions labeled “miscellaneous.”[9] Notable successes lead to the decreasing trend, giving hope for further air quality improvements. Implementation of the 1970 Clean Air Act[10] reduced the various pollutants over this period, showing the power of effective legislative regulation. Most impressively, perhaps, legislation reduced VOC emissions from highway vehicles from 17 million tons in 1970 to 3.9 million tons in 2006.

Note also the change in the U.S. economy, seen in Figure 3.4, from 40% goods-producing industries in 1950 to 20% in 2000. Service industries took up the slack. Perhaps a service economy has fewer emissions than one based on producing goods: Could the emissions shifts seen here be, in part, a reflection of this economic change? One problem with this hypothesis is that transportation produces much of our NO_x and VOC emissions, and people’s travels don’t decrease with a service economy.[11]

One example highlights traffic’s role. Atlanta imposed tight traffic restrictions during the 1996 Atlanta Summer Olympics, and the experiments, summarized in the right plot, took advantage of the situation.[12] Gasoline sales decreased by 3.9% for the month, including purchases by 1 million visitors. During the games, weekday peak morning traffic counts were 22.5% lower than usual, though total weekday traffic decreased only 2.8%, but public transportation more than doubled with 17 million additional trips!

These data represent the findings as a percentage of the National Ambient Air Quality Standards (NAAQS)[13] and clearly demonstrate that emissions affect air quality. During the games carbon monoxide and PM₁₀[14] drops were significant, and ozone levels were 28% lower. In addition, there were significant correlations between peak morning traffic counts and peak ozone concentration on each day. However, after correcting for four weather variables, temporal trends, and day of week, ozone levels were just 13% lower within Atlanta compared to 2-7% lower in neighboring areas.

What do these numbers mean on a personal level? Consider, for example, human emissions of VOCs, amounting to roughly 400 billion pounds per year. These emissions run the gamut of solvent use, highway vehicles, other fuel combustion, and so on. Dividing this huge number by the U.S. population gives a bit more than 1,000 pounds per person per year, or VOC emissions of about 3 pounds per American per day.

——————————-

[9]Emissions data from the U.S. EPA are available at epa.gov/ttn/chief/trends. The U.S. EPA compiled these emissions estimates from many sources; the sharp changes arise from new methodologies, measurements, and regulations, not from actual increases or reductions. Likens et al. (2005) discuss some of the mysterious, sharp changes. The numbers I plot for particulate matter include “condensables,” which means vapor emissions, not really solid or liquid particles at the moment of emission. This vapor may later condense into very small particles. EPA data also include numbers for particulate matter excluding condensables, showing that about 40% of particulate matter constitutes condensable emissions. In my plot, values for several pollutants must be multiplied by a factor of 10 or 100 (for example, “4” would read “40” or “400”).

[10]The EPA website, epa.gov/air/caa, provides an overview of the 1990 Clean Air Act.

[11]I thank an anonymous reviewer for the counterargument that travel doesn’t decrease with a service economy, which is certainly true. Yet, I wonder: There must be some emissions signature due to this economic shift!

[12]Air pollution measurements in downtown Atlanta during the 1996 Atlanta Summer Olympics are discussed in Friedman et al. (2001). Georgia’s Environmental Protection Division monitored air quality before, during, and after the Olympics. The 2008 Beijing Olympics “repeats” these “experiments,” but I have not seen any results.

[13]The Clean Air Act of 1990 requires the U.S. Environmental Protection Agency to prepare the National Ambient Air Quality Standard (NAAQS) for many different pollutants.

[14]PM₁₀ stands for “particulate matter” smaller than 10 microns — a micron is 10⁻⁶ meters — really just small particles of junk. Air quality experts now measure even smaller particles, those less than 2.5 microns, which can get trapped deep in the lungs. Their density is denoted by PM_2.5.