Page 14
3
The Effects of Changing Global Atmospheric Composition on Air Quality
It has long been recognized that pollution emissions can affect air quality beyond national borders. For instance, field studies have shown that the prevailing westerly winds typically carry ozone and its precursors from the eastern United States into Canada, the North Atlantic, and beyond (e.g., Prados et al., 1999); and in turn, it is known that air masses reaching the United States can carry pollution originating from many other parts of the world. The following is a discussion about some of the current and possible future effects of cross-border and intercontinental transport of air pollutants, with an emphasis on assessing the potential implications for U.S. air quality.
Observations of Current Impacts
Long-range transport of aerosols and trace gases from Asia is known to alter the composition of the remote Pacific troposphere (e.g., Uematsu et al., 1983; Xiao et al., 1997). In spring, when storm and frontal activity in Asia is most prevalent, outflows of continental pollution and dust are observed in both in situ observations and in satellite studies (e.g., Prospero and Savoie, 1989; Merrill et al., 1989; Herman et al., 1997; Wilkening et al., 2000).
There is growing observational evidence that the effects of pollutant outflow from Asia extend to North America. One study found evidence that Asian boundary layer air can be transported to the upper troposphere over California in two to four days (Kritz, 1990). Surface observations of CO coupled with air transport “back trajectories” were used to infer that anthropogenic emissions from Asia
Page 15
were transported to the surface of North America during the spring of 1997 (Jaffe et al., 1999). In the spring of 1999, CO and ozone attributed to emissions originating in Asia were again detected off the coast of Washington State (Jaffe et al., 2001). Analysis of ozone data from U.S. air quality monitoring networks reveals that the low end of the ozone probability distribution has been increasing over the past two decades, which investigators attribute to rising levels of background pollution transported from outside the United States (Lin et al., 2000).
The long-range transport of dust aerosols to the United States from Africa has been well documented. For instance, Saharan dust transported to the southeastern U.S. during summer months has been observed to impact atmospheric PM concentrations in Miami (Prospero, 1999), and has also been detected in Great Smoky Mountains National Park and even further inland (Perry et al., 1997).
The long-range transport of gases and aerosols from biomass burning has also been detected. Emissions from large forest fires in Canada increased both CO and aerosol concentrations along the eastern seaboard of the United States in 1995 (Wotawa and Trainer, 2000). In May 1998, smoke from numerous fires burning in Mexico and Central America was transported into the southern U.S., prompting Texas and other states to issue public health advisories. Smoke from these fires was documented in Oklahoma (Peppler, 2000), and satellite and ground-based observations showed this smoke plume extending northeastward into the Smoky Mountains (Kreidenweis et al., 2001). Analyses of monitoring data from Big Bend National Park in Texas suggest that smoke transport to the park was an annual spring event (Gebhart et al., 2000).
Long-range pollution transport is known to affect even very remote regions of the earth. The wintertime phenomenon known as arctic haze has been extensively studied since its first identification in the 1950s. These studies revealed evidence that the observed haze was caused by mid-latitude emissions from fossil fuel combustion, smelting, and other industry, primarily from Eurasia (Barrie, 1986). Since then, a series of field studies have greatly improved our understanding of how pollution transported to the Arctic affects atmospheric chemistry and climate within this region (for instance, see Harriss et al., 1992; MacCracken et al., 1986; Blanchet, 1989).
Modeling Studies of Current/Future Impacts
A number of three-dimensional global chemical transport models (GCTMs) have been used to study the issue of long-range pollution transport. One modeling study predicted that future developments in Asia could lead to emissions that will significantly perturb free tropospheric ozone chemistry over the Pacific (Berntsen et al., 1996). More recently, it was calculated that Asian emissions have already raised mean background springtime levels of CO and O3 by an average of 34 and 4 ppb, respectively, over the eastern North Pacific (Berntsen et
Page 16
al.,1999). 2 Another study predicted that if Asian NOx emissions were to triple between 1985 to 2010, this increase could enhance surface background ozone over the western United States by 2-6 ppb (Jacob et al., 1999).
While these simulations estimate monthly or seasonal mean changes, many observations suggest that Asian pollution outflow is highly episodic and may not be well represented as a uniform background enhancement. For instance, recent observations of CO at an observatory on the western coast of Washington State found strong synoptic-scale fluctuations with changes of 20 to 50 ppb over a few days (Jaffe et al., 2001). A model simulation by Yienger et al. (2000) predicted these types of large synoptic-scale fluctuations in CO. In this study it was also estimated that the current Asian contributions to episodic ozone events over the western United States are in the range of 3-10 ppb, and that future ozone episodes along the U.S. west coast could have Asian contribution as high as 40 ppb (based on a “worst case” emission scenario wherein Asian NOx emissions increase by a factor of four from 1990 to 2020).
Long-Term Projections
Studies carried out for the Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC) have extended estimates of future air quality changes to the year 2100. These studies are based on projections of future pollution emissions which, in turn, depend on changes in driving forces such as population, social and economic development, and technology. Attempts to predict changes in these driving forces and resulting emissions are highly uncertain when their time horizon extends as far as a hundred years into the future. Recognizing these uncertainties, however, one can examine the implications of a range of possible future emissions by developing a set of alternative scenarios.
A comprehensive set of long-term emission scenarios has recently been developed under the auspices of the IPCC Special Report on Emission Scenarios (Nakicenovic and Swart, 2000), which are based on a range of assumptions about the rate and direction of economic and technological change and the degree of globalization. These scenarios include projections for species that influence air quality, including SO2, NOx, CH4, VOCs, and CO. The IPCC selected six “illustrative” scenarios (out of the 40 that were originally developed) to represent a range of assumptions and modeling approaches. None of these scenarios is identified as more probable than any other. The results of the six illustrative scenarios for emissions of SO2 and NOx over the next 100 years are displayed in
Figure 3-1 marine boundary layer) to be 151 ppb, and the average concentrations of ozone to be 31 ppb. . Note that the total emissions of SO2 generally decrease over time in these scenarios, due to an assumed widespread implementation of emission control technologies.
2 For reference, Berntsen et al. estimated the average springtime concentrations of CO (in the marine boundary layer) to be 151 ppb, and the average concentrations of ozone to be 31 ppb.
Page 17
~ enlarge ~
FIGURE 3-1 IPCC emission scenarios for NOX and for SO2. The “A” scenarios represent materially oriented development, whereas the “B” scenarios involve a transition to information and service-oriented economies. The “1” scenarios represent globally oriented development, and the “2” scenarios represent locally and regionally oriented development. Within the A1 family, the three variants are based on different energy sector developments: fossil fuel intensive (A1F); transition to non-fossil energy sources (A1T); and balance across all energy sources (A1B). See Nakcenovic and Swart (2000) for more details.
Page 18
Box 3-1Current Policy Approaches to Addressing Transboundary Air Pollution
The importance of air pollution transport across national boundaries has been recognized in some regions for more than twenty years. The original focus of attention was acid deposition in Europe, which led to the Convention on Long-Range Transboundary Air Pollution (LRTAP) signed in 1979. Since then the Convention has been extended by eight protocols, most recently the 1999 Protocol to Abate Acidification, Eutrophication, and Ground-level Ozone. This Protocol makes regional air quality a central focus of the LRTAP Convention and includes a schedule of specific emission limits on SOX, NOX, VOCs, and ammonia for countries throughout Europe. It also establishes air quality objectives for Europe similar to U.S. National Ambient Air Quality Standards. The United States and Canada also participate in the air quality protocols of the LRTAP convention, although U.S. and Canadian obligations are simply designed to mirror pre-existing national laws. More pertinent in North America are the U.S.-Canada Air Quality Agreement, signed in 1991, and the Commission on Environmental Cooperation, established pursuant to the North America Free Trade Agreement. While the Air Quality Agreement initially addressed only acid deposition, the United States and Canada recently negotiated a new annex to the agreement in response to findings that there is significant cross-border transport of ozone in eastern North America (as discussed in the report Ground-Level Ozone: Occurrence and Transport in Eastern North America by the United States-Canada Air Quality Committee, 1999). |
These scenarios were used with a variety of GCTMs to explore how global air quality and climate could be affected over the next century by the projected increases in pollution emissions. When using a “high end” emission scenario (A2, which assumes the largest emissions of CH4 and ozone precursors of the six IPCC illustrative scenarios), all of the models projected large increases in global tropospheric ozone levels. Specifically, it was estimated that zonal mean surface ozone concentrations over northern midlatitudes for the month of July would increase from current values (estimated at about 40 ppb) to more than 70 ppb, and more than 80 ppb near the tropopause, by the year 2100 (Ehhalt and Prather, 2001). These projected changes, which result from the cumulative impact of all emissions, could have serious consequences for the air quality of most of the northern hemisphere.
It should be emphasized, however, that alternatives to this high-end emission scenario do not lead to such large-scale air quality changes. Moreover, although all models predicted significant responses of ozone to changes in precursor emissions, estimates of the overall magnitude and geographical distribution of these impacts varied widely. Also, the models used for these studies do not yet include some important processes such as the changing chemistry of the
Page 19
stratosphere and the impacts of climate and land-use change on biogenic emissions of ozone precursors. Nevertheless, these studies illustrate that changes in air pollution emissions occurring over the next several decades could potentially have impacts far beyond the regions in which they are emitted, and that in the 21st century, a global perspective may be needed to meet regional air quality objectives.
