FIGURE 16.39 Relationship between observed concentrations of O 3 and reacted oxides of nitrogen expressed as (NOv — NOt) at a site downwind of Toronto, Canada (adapted from Roussel et al, 1996).

intercept in Fig. 16.39 is 26 ppb, which should represent the "background" 03. (Note that as discussed in Chapter 14, this is not the level of 03 in an atmosphere unperturbed by anthropogenic emissions but rather the current global level, which is a factor of 2-3 times higher than prior to the industrial revolution due to anthropogenic influences.) Both the slope and intercept are in good agreement with values of 6-12 reported for the slope and 24-42 reported for the intercept in other studies carried out in the eastern and southeastern United States (e.g., see Trainer et al., 1993; Kleinman et al., 1994; Olszyna et al., 1994; Hastie et al., 1996; and Ridley et al., 1998). These intercept values are also consistent with measurements of current boundary layer levels of 03 measured independently in a number of studies (e.g., Altshuller and Lefohn, 1996). Similar values for the ozone production efficiency have been reported in other areas around the world, including Athens (e.g., Peleg et al., f997) and LI. Valby, Denmark (Skov et al., 1997).

A slightly modified approach has also been used in a number of studies in which the sum of (03 + N02) is plotted against NOz (e.g., St. John et al., 1998). This minimizes the effects of short-term variations in 03 caused by its rapid reaction with NO. Thus, when 03 is titrated by the NO reaction, the measured 03 concentrations will be small; however, the N02 generated is a source of 03 through its subsequent photolysis. Hence the sum of (03 + N02) is sometimes chosen as a measure of the ultimate formation of ozone. The ozone production efficiency determined from slopes of plots of (03 + N02) against NOz in the Nashville, Tennessee, area was measured to be typically 5-6 if it was assumed that NO is not removed by other processes. The production efficiency appeared to be about the same for the general urban plume and for an air mass in which a plume from a power plant was also embedded. Including other losses for NOy such as deposition lowers the estimated production efficiency by about a factor of two (St. John et al., 1998; Nunnermacker et al., 1998).

This relationship between 03 and NOz = (NOy -NOx) in areas downwind of urban centers can be anticipated, based on the oxidation of NO to N02 by H02 and R02 radicals and the subsequent photolysis of N02 to form 03. As discussed in Chapter 6.J, at lower NOx concentrations, reactions of HOz and R02 with themselves and each other compete with their reactions with NO. However, the oxidation of NO to N02 leads to 03 formation since photolysis of N02, generating 03, is a major fate for N02. This then gives rise to the observed relationship between NOz and 03. Under these low-NOx conditions, the formation of H202 and other peroxides is important and deter

FIGURE 16.39 Relationship between observed concentrations of O 3 and reacted oxides of nitrogen expressed as (NOv — NOt) at a site downwind of Toronto, Canada (adapted from Roussel et al, 1996).

mined largely by the rate of formation of the precursor free radicals (e.g., Kleinman, 1991, 1994).

However, when the rate of NOx emissions is larger than the rate of radical production, this relationship would not be expected to be as clear. In this region, reaction of N02 with OH to form HN03 becomes important, removing both the free radical OH and NOz without forming 03, weakening the 03-N0z relationship. Under these high-NOx conditions, the HOz + H02 or R02 reactions are also less important, leading to a decreased formation of peroxides.

Kleinman and co-workers (Kleinman, 1991, 1994; Kleinman et al., 1997) examined the utility of treating VOC-NOx chemistry in terms of these two regimes defined in terms of the relative rates of free radical production compared to emissions of NOx. Given that the rates of free radical formation vary rapidly and from location to location, transitions from one regime to another can occur diurnally and seasonally, as well as geographically. For example, such a transition can occur during the fall when photolysis of 03 to form Of'D) and subsequently OH by its reaction with water decreases due to seasonal decreases in UV and relative humidity. Depending on the relative strength of the NOx emissions, this can result in a transition from the low-NOx to the high-NOx regime, accompanied by decreases in H202 production and a weakening of the correlation between 03 and NOz.

This behavior is consistent with the observations of a number of field studies. For example, Jacob et al. (1995) report that in Shenandoah National Park in Virginia (U.S.) in early September, there was a good correlation between 03 and NOz, with a slope of f8 compared to the range of 8.5-14 observed in other studies. In the latter part of September, the correlation was weaker (r2 = 0.23 vs 0.49 earlier) and the slope was only 7. This weakening of the relationship between 03 and NOz was accompanied by a decrease in concentrations of H202 from an average of 0.86 ppb to 0.13 ppb, as expected for a transition from the NOx-limited to the VOC-limited regime.

A number of modeling studies, combined with field measurements, suggest that VOC control may be more effective than NOx in controlling 03 at some locations, primarily urban. This is consistent with both box and airshed model predictions in that if one is in effect in the low VOC/NOx regime on the top of the ridge line in Fig. 16.14a, a decrease in NOx could actually lead to an initial increase in 03 before it decreases, ft should be noted, however, that these highly polluted locations are generally not those at which the ozone peaks occur. As illustrated in Fig. 16.14, as the air parcel moves downwind from these low VOC/high NOx regions, it generally moves into the NOx-limited regime and it is in these downwind areas that 03 generally peaks. This type of behavior, i.e., transition from the VOC- to the NOx-limited regime, has been observed in many regions, including the Los Angeles area, Toronto (Fuentes and Dann, 1994), Munich (Fabian et al., 1994), and the eastern U.S. (McKeen et al., 1991b; Mathur et al., 1994). Modeling studies on the generation and transport of 03 in urban areas are also consistent with this increased sensitivity to NOx as the air mass moves downwind (e.g., Duncan and Chameides, 1998).

Based on the chemistry discussed above, the use of indicator species has been proposed to differentiate air masses in which ozone formation is more sensitive to NOx than to VOC and vice versa (e.g., Milford et al., 1994; Sillman, 1995; Sillman et al., 1997; Lu and Chang, 1998; Sillman, 1999). For example, model calculations suggest that high values of the ratios Oj/fNO^ - NOJ, HCHO/NOr and H202/HN03 reflect air masses in the NOx-sensitive regime whereas low values reflect the VOC-sensitive regime (e.g., Sillman et al., 1997; Lu and Chang, 1998). The reasons for this are found in the complex chemistry discussed above, but some generalizations can be made. For example, high 03 and low NOv suggest relatively large free radical sources and smaller radical sinks such as the N02 + OH reaction. HCHO is a measure of the oxidation products of both anthropogenic and biogenic VOC; thus, the higher HCHO relative to NO , the further to the right of the isopleths (Fig. 16.14a) is the air mass, i.e., toward the NOx-sensitive regime. Similarly, high concentrations of H202 relative to HN03 indicate that the air mass is in the high-VOC/NOx regime where ozone is most sensitive to NOx control. Some of the uncertainties in the application of such indicators in field studies are discussed by Sillman (1995) and Lu and Chang (1998).

Related to the use of indicator species is the use of "species age" in photochemical modeling studies (Venkatram et al., 1998). In this approach, the VOC and NOx are calculated at a particular point of interest in an air basin assuming no chemical reactions, i.e., only transport occurs. The age of the VOC and NOx since the time of emission is also calculated. The amount of 03 formed is then estimated using the VOC-NOx chemistry for that time period. This approach separates transport and chemistry in an explicit manner and allows the calculation of the effectiveness of various VOC and NOx reductions at a particular location.

In short, a combined VOC-NOx reduction strategy seems to be the optimum approach to controlling ozone and other secondary air pollutants, and there is evidence from the experience in southern California that this approach is effective. For differing viewpoints, however, see articles by scientists from General Motors

Research Laboratories (Chock et al., 1981, 1983; Klim-isch and Heuss, 1983; Kumar and Chock, 1984; Wolff, 1993).

Was this article helpful?

0 0

Post a comment