August 10, 2007

Future Air Pollution Levels and Climate Change: A Step Toward Realism

Filed under: Aerosols, Climate Forcings

Guest Commentary

Joel Schwartz
Visiting Fellow
American Enterprise Institute

What happens to future air pollution if the climate warms? Efthimios Tagaris and colleagues (Tagaris et al., 2007) have come closer than anyone before them in providing a realistic answer to this question. They predict that between 2001 and 2050, mean summer 8-hour ozone levels over the U.S. will decline by 11% to 28%, depending on the region, with an average decline of 20%. Fine particulate matter (PM2.5) will decline by 9% to 32%, with an average decline of 23%.

Tagaris et al.’s results stand in marked contrast to previous studies, nearly all of which predict higher air pollution levels over the U.S. in the future (e.g., Knowlton et al. (2004), Mickley et al. (2004), and Sitch et al. (2007)).

What explains the difference? Tagaris et al. is the only study to base its predictions of future air pollution levels on somewhat realistic estimates of future air pollutant emissions. Tagaris et al.’s modeling includes the effect of required declines in pollution emissions that have been occurring and will continue over the next few decades. On the other hand, previous studies have assumed constant or increasing emissions of air pollutants. In fact, as I show below, even Tagaris et al.’s predictions are too pessimistic. Air pollution will decline even more than they suggest.

Climate and regional pollution models predict that, assuming no changes in pollutant emissions, a warmer climate will tend to increase air pollution in some areas of the U.S. A major reason for this is a greater likelihood of multi-day “stagnation” events, creating more opportunities for pollution to build up over several days. Warmer temperatures also increase natural emissions of volatile organic compounds (VOC). But climate change won’t be all bad for air pollution. Higher temperatures also reduce particulate levels by preventing the condensation of “semi-volatile” compounds, like ammonium nitrate and some organics. More frequent summer rains in some areas would also tend to reduce ozone levels.

Overall, Tagaris et al. project that climate change alone will have little net effect on national-average levels of ozone or PM2.5, though there would likely be regional variation, with some areas experiencing higher and some lower pollution levels. Even this is an advance over previous studies, which have tended to focus only on regions (i.e., the Northeast) and pollutants (i.e., ozone) where climate warming is most likely to increase future levels, while ignoring potential air pollution benefits from warming.

But Tagaris et al.’s unique contribution is their attempt to base their estimate of future pollution levels on a realistic projection of future emissions. They assume, for example, that North American emissions of oxides of nitrogen (NOx, the sum of NO and NO2) and sulfur dioxide (SO2) will decline 50% between 2001 and 2050, while anthropogenic non-methane volatile organic compounds (NMVOC) will decline a little over 40%. On the other hand, natural NMVOC emissions increase more than 20% in their model, due to warmer temperatures, resulting in no net change in total NMVOC emissions over time.

Based on these assumptions, combined with the IPCC’s A1B scenario for future greenhouse gas emissions and consequent warming, Tagaris et al. conclude that between 2001 and 2050 the number of days per year exceeding the federal 85 ppb, 8-hour ozone standard will decline between 60% and 100%, depending on the region/metropolitan area. Tagaris et al.’s model projects large declines in particulate levels as well—a 34% to 80% reduction in the number of days per year exceeding EPA’s new daily PM2.5 standard of 35 micrograms per cubic meter (µg/m^3) (the entire nation already complies with the previous standard of 65 µg/m^3).

So why should we believe Tagaris et al.’s prediction of declining pollution emissions and not previous studies’ prediction of increasing emissions? First of all, declining emissions is what you would predict based on current trends. Figure 1 shows EPA’s estimate of pollution emissions from 1970-2006. All values are scaled to a common index of 1.0 in 1990, so that percentage changes over time can be compared on the same scale. The underlying emissions estimates for the chart are available from the EPA. These trends in estimated emissions are confirmed by observed ongoing declines in ambient levels of these air pollutants (see here for trends in levels of various air pollutants through 2006).

Figure 1. Trends in Estimated U.S. Air Pollutant Emissions, 1970-2006 (Source: Environmental Protection Agency, 2007a).

Second, based on existing regulatory requirements, we should expect declining pollution levels to continue for decades to come. EPA’s Clean Air Interstate Rule (CAIR) requires power plant SO2 and NOx emissions to decline more than 70% and 60%, respectively, during the next two decades, when compared with 2003 emissions. This is a cap on total emissions from power plants that remains in place independent of growth in electricity demand.

Recently implemented requirements for new automobiles and diesel trucks, and upcoming standards for new off-road diesel equipment will eliminate more than 80% of their VOC, NOx, and soot emissions during the next few decades, even after accounting for growth in total driving. Dozens of other federal and state requirements will eliminate most remaining emissions from other sources of air pollution.

Third, real-world measurements show that, if anything, Tagaris et al. understate future improvements in pollution emissions and levels. You can see this by comparing current trends to Tagaris et al.’s predictions. For example, Tagaris et al. assume total NOx emissions will decline 14% between 2001 and 2020 (Woo et al., 2006). In fact, NOx emissions already declined 18% between 2000 and 2006 (Environmental Protection Agency, 2007a).

Tagaris et al. assume VOC emissions will decline 22% between 2001 and 2020. But EPA estimates VOC emissions already declined by 12% between 2000 and 2006. Furthermore, most VOC emissions come from gasoline vehicles and on-road measurements show these are dropping at a rate of about 10% per year (that includes the offsetting effect of growth in total driving) (Kean et al., 2002; Pokharel et al., 2003; Schwartz, 2003).

Ambient pollution trends also show faster declines than Tagaris et al. assumptions suggest. According to Tagaris et al., national-average PM2.5 levels will drop about 23% between 2001 and 2050. But average PM2.5 dropped 10% from just 2001 to 2006 (Environmental Protection Agency, 2007b). Tagaris et al. predict a 20% decline in summer-average ozone during the same period. But summer-average ozone levels for 2004-06 were already 6% lower than for 2000-2002 (Tagaris et al.’s baseline period). In addition, ozone-precursor pollutants—NOx, VOC, and CO—are dropping faster than Tagaris et al. assume. Thus, even though Tagaris et al. are the first researchers to even attempt a realistic appraisal of future air pollution levels, their estimates should be considered a worst-case upper limit on future emissions.

In contrast to Tagaris et al., previous studies of climate change and air pollution are patently unrealistic. I showed in my last WCR commentary that Sitch et al. (2007)’s assumption of increasing air pollution emissions is the polar opposite of observed trends and regulatory requirements.

Mickley et al. (2004) hold CO and black carbon (BC; i.e., “soot”) emissions constant and then use the GISS general circulation model to predict that peak CO and BC levels in the Northeast would rise 5% to 10% in 2050 under the IPCC’s A1B scenario. Although the introduction to their paper makes a passing mention of the importance of future pollutant emissions, the rest of the article leads readers to believe that future air pollution levels will be higher than current levels. Mickley also gave reporters this impression in a press release on her research sub-titled “Summertime pollution may intensify in the northeastern and midwestern U.S. due to global warming” (Mickley, 2005). Mickley was more forthcoming in a guest column on Real Climate, but still omitted the fact that future pollution emissions will be much lower than today and that this is far more important than climate change for future air pollution levels.

Knowlton et al. (2004) used EPA’s NOx and VOC emissions estimate for 1996 to predict ozone levels after 2050. Pat Michaels, Robert Davis and I sent a letter to Environmental Health Perspectives, which published Knowlton et al., pointing out this lack of realism in Knowlton et al.’s emission “projection” (Schwartz et al., 2005). In response, Knowlton et al. claimed that their study wasn’t actually intended to predict likely future ozone levels, but was merely a feasibility study (though they also expressed the mistaken view that NOx and VOC emissions would increase in the future). This was a direct contradiction of their own paper, which portrayed their results as a prediction that ozone would increase in the future.

Knowlton et al. performed their study on behalf of the Natural Resources Defense Council, an environmental activist group, and several members of the Knowlton et al. team also co-authored the NRDC report Heat Advisory: How Global Warming Causes More Bad Air Days (Patz et al., 2004), leaving little doubt about the true purpose of the research. Although Knowlton et al. appeared in a prestigious peer-reviewed scientific journal, the whole exercise was pre-determined to produce the ideologically desired answer—air pollution will increase in the future due to climate change—rather than the scientifically justifiable answer—air pollution will decline in the future, regardless of climate change.

In some cases, researchers studying the effect of climate change on air pollution justify holding pollutant emissions constant in order to isolate the effect of climate change on air pollution. Fair enough. The problem is that they hold emissions constant at current levels and then apply a warmer future climate to current emissions. Thus, what they are really assessing is how current ambient pollution levels would change given current emissions, if the current climate were a few degrees warmer—a counterfactual scenario that is of little policy or even scientific relevance.

Instead, to provide policy-relevant information, they should begin with a realistic estimate (or reasonable range of estimates) for future pollutant emissions and then apply these emissions to two climate scenarios: a baseline with no future warming, and then a scenario (or scenarios) with warming. This would isolate the effect of future warming on future air pollution.

Even Tagaris et al. doesn’t do this, because it doesn’t include a “no warming” baseline. However, Tagaris et al.’s results do suggest that there would be some places where climate warming would cause air pollution to decline somewhat less than it otherwise would, while in other areas air pollution would decline somewhat more than it otherwise would. That’s not nearly as scary as saying climate change will cause air pollution to increase above current levels, and that probably explains why there has been so little effort at realistic modeling on this question.

There’s an interesting irony here. Climate alarmists act as if their model projections represent relatively certain predictions of future climate parameters, and they vigorously defend the purported realism of their results. Yet when it comes to future air pollutant emissions, this ostensible quest for realism suddenly disappears, in favor of scenarios that are patently at odds with reality. If anything, the proposition that air pollutant emissions will sharply decline in the future is far more certain than any predictions of how and why the Earth’s climate will change due to greenhouse gas emissions. Fortunately, Tagaris et al. have begun nudging climate scientists toward a more realistic assessment of future air pollution levels under a changing climate.


Environmental Protection Agency, 2007a. Air Quality and Emissions – Progress Continues in 2006,

Environmental Protection Agency, 2007b. Air Trends,

Kean, A.J., Sawyer, R.F., Harley, R.A. and Kendall, G.R., 2002. Trends in Exhaust Emissions from In-Use California Light-Duty Vehicles, 1994-2001. Society of Automotive Engineers, Warrendale, Penn.

Knowlton, K., Rosenthal, J.E., Hogrefe, C., Lynn, B., Gaffin, S., Goldberg, R., Rosenzweig, C., Civerolo, K., Ku, J.Y. and Kinney, P.L., 2004. Assessing ozone-related health impacts under a changing climate. Environmental Health Perspectives, 112(15), 1557-63.

Mickley, L.J., 2005. “Effects of future climate change on regional air pollution episodes in the United States.” Press release for “The Pollution-Climate Connection,” a presentation in the AAAS session “Climate Change is in the Air.” Harvard University, February 19,

Mickley, L.J., Jacob, D.J., Field, B.D. and Rind, D., 2004. Effects of future climate change on regional air pollution episodes in the United States. Geophysical Research Letters, 31, L24103.

Patz, J.A., Kinney, P.L., Bell, M.L., Ellis, H., Goldberg, R., Hogrefe, C., Khoury, S., Knowlton, K., Rosenthal, J., Rosenzweig, C. and Ziska, L., 2004. Heat Advisory: How Global Warming Causes More Bad Air Days. Natural Resources Defense Council,

Pokharel, S.S., Bishop, G.A., Stedman, D.H. and Slott, R., 2003. Emissions Reductions as a Result of Automobile Improvement. Environmental Science and Technology, 37, 5097-5101.

Schwartz, J., 2003. No Way Back: Why Air Pollution Will Continue to Decline. American Enterprise Institute,

Schwartz, J., Michaels, P. and Davis, R.E., 2005. Ozone: unrealistic scenarios. Environmental Health Perspectives, 113(2), A86-7; author reply A87.

Sitch, S., Cox, P.M., Collins, W.J. and Huntingford, C., 2007. Indirect radiative forcing of climate change through ozone effects on the land-carbon sink. Nature, 448, doi:10.1038/nature06059.

Tagaris, E., Manomaiphiboon, K., Liao, K.-J., Leung, L.R., Woo, J.-H., He, S., Amar, P. and Russell, A.G., 2007. Impacts of global climate change and emissions on regional ozone and fine particulate matter concentrations over the United States. Journal of Geophysical Research, 112, doi:10.1029/2006JD008262.

Woo, J.-H., He, S. and Amar, P., 2006. Development of Mid-Century Anthropogenic Emissions Inventory in Support of Regional Air Quality Modeling under Influence of Climate Change, EPA 15th International Emission Inventory Conference, May 15-18, New Orleans.

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress