Published in: 2nd conf. Environment & Transport, incl. 15th conf. Transport and Air Pollution. Reims, France, 12-14 June 2006. Proceedings n○107, Vol. 2, Inrets ed., Arceuil, France, 2006, p. 264-269
Unexpectedly low decrease of NO2 air pollution – Correlation with ozone concentration and altered exhaust emissions Werner SCHOLZ* & Peter RABL** * LUBW • State Institute for Environment, Measurements and Nature Conservation BadenWürttemberg, D-76185 Karlsruhe, Germany – email:
[email protected] ** Bavarian Environmental Protection Agency, D-86177 Augsburg, Germany Abstract An analysis of air pollution data of nitrogen oxides and ozone over the last 16 years was performed for the German federal states of Baden-Württemberg and Bavaria, covering more than 60 air quality monitoring stations. These stations were clustered into four groups according to their distance to road traffic. At highly traffic-influenced sites the total nitrogen oxides (NOx) and also nitrogen monoxide (NO) show a distinct decrease, in accordance with the reduction of NOx emissions of vehicles, whereas nitrogen dioxide (NO2) concentrations reveal only a modest decline or even a slight increase at some sites. This is an obvious discrepancy to the calculated development of NOx emissions. The ratio NO2/NOx exhibits a pronounced increase at urban monitoring sites. Ozone concentrations in cities show a comparable increase in the mean, probably because of reduced degradation reactions. One explanation for these divergent trends is based on the photochemical balance between NO, NO2 and ozone. Another, and probably important, reason seems to be a large increase of direct NO2 exhaust emissions as a result of the growing number of modern private diesel cars equipped with oxidation catalysts, producing high NO2/NOx ratios by the oxidation of NO. The database concerning reliable NO2 (not NOx) emission data of modern cars and exhaust treatment equipment is yet insufficient. From the NO2 and ozone concentration trends over 16 years it is concluded that the EU air quality limits of NO2, valid from 2010, may then be exceeded at locations with a heavy traffic burden. Keys-words: Air pollution trend, traffic-related pollution, nitrogen oxides, nitrogen dioxide, diesel passenger cars, oxidation catalyst, exhaust after treatment. Résumé Une analyse des moyennes annuelles en oxydes d'azote et en ozone fut effectuée pour la période de 1990 à 2005 pour les stations de mesure de la qualité de l'air de Bade-Wurtemberg et de la Bavière. Tandis que le niveau de pollution atmosphérique de NOx et de NO a diminué distinctement, en conformité avec le développement des émissions de NOx, seulement une petite diminution, partiellement aussi une augmentation, s'est manifestée aux concentrations de NO2 près de rues très fréquentées. Une explication pour cet effet est vue dans l'équilibre photochimique entre NO, NO2 et ozone. Une autre raison est la composition changée du gaz résiduel des voitures diesel. Les voitures particuliers modernes diesel sont équipées avec un pot catalytique d'oxydation qui transforme NO en NO2, de sorte qu'une partie importante de l'oxyde d'azote est directement émis comme NO2. Par conséquent, un dépassement des valeurs limites pour NO2 en 2010 sera vraisemblable à proximité de rues très fréquentées. Mots-clefs : qualité de l'air, oxyde d'azote, dioxyde d'azote, équilibre photochimique.
2
Introduction The development of NO2 air concentrations at highly traffic-influenced sites is contrary to the decreasing trend of nitrogen oxide (NOx) emissions. A trend analysis of air concentration data from monitoring sites in Baden-Württemberg and Bavaria was performed to identify the reasons. 1 - Development of nitrogen oxide emissions Road traffic is the main source of nitrogen oxide (NOx) emissions. In 2000, road transport accounted for 48% of the total NOx emissions in Germany (Umweltbundesamt, 2004) and for 51% in Baden-Württemberg (UMEG, 2004). Between 1990 and 2002, total NOx as well as traffic NOx emissions showed a continuous decrease of about 47-48% (figure 1). Development of NOx emissions in Germany (kt NO2/a) 3000
Agriculture Military
2500
Industrial Processes 2000 Small scale combustion units
1500
Manufacturing and Construction Energy Industries
1000
Offroad transportation
500
Road transportation
Figure 1: Figure 1:
20 02
20 01
20 00
19 99
19 98
19 97
19 96
19 95
19 94
19 93
19 92
19 91
19 90
0
Development of NOx emissions in Germany, in kt NO2/a (Source: Umweltbundesamt, 2004) Evolution des émissions d’oxyde d’azote en Allemagne, en kt NO2/a
2 - Development of nitrogen oxide and ozone air concentrations Based on the calculated decrease of NOx emissions, a corresponding decrease in NOx air concentrations should obviously be expected. For the trend analysis, the monitoring sites were classified into four groups according to their distance from road traffic. The results for BadenWürttemberg are presented below; the analysis of data for monitoring sites in Bavaria yielded similar results (shown in Rabl & Scholz (2005)). NOx air concentrations show a distinct decrease which is most pronounced in main roads and at urban sites with traffic influence (figure 2). The observed decrease in these groups is by a factor of 53% and 51% respectively between 1990 and 2005, in good correspondence with the calculated emission trend. The graph of the development over time is reasonably fitted by a linear approximation. NO air concentrations exhibit an even steeper downward trend at main roads and urban sites than the NOx concentrations. Only in the rural background NO remains essentially unaltered at a very low level (figure 3a). In contrast, NO2 air concentrations do not exhibit a corresponding decrease (figure 3b): In urban areas the downward trend is more moderate than that one of NOx and NO (-33% to -37%), whereas in direct vicinity of highly frequented streets almost stagnation is observed, in some cases even an increase of NO2 concentrations.
3 NOx air concentrations in Baden-Württemberg 250
NOx as NO2 (µg/m³)
200 main roads (6) y = -7,499x + 15149
150
urban traffic influence (5) urban background (18)
100
y = -3,449x + 6971
rural background (4)
50
0 1990 1992 1994 1996 1998 2000 2002 2004
Figure 2: Figure 2:
Development of NOx air concentrations in Baden-Württemberg 1990-2005 (annual means, calculated as NO2; number of monitoring sites in parentheses) Evolution des niveaux d’oxyde d’azote (NOx) à Baden-Württemberg de 1990 à 2005 (moyennes annuelles, calculé comme NO2; nombre de stations en parenthèses)
This trend of NO2 concentrations will turn into a problem for the air quality situation in many cities during the next years. As can be seen from figure 3b, the NO2 annual mean concentrations at main roads are considerably above the EU air quality limit of 40 µg/m³ valid from 2010. Even a further considerable reduction of motor vehicle NOx emissions would only have a perceptible effect several years later. a)
NO air concentrations
b)
NO2 air concentrations in Baden-Württemberg 80
140
70
120
60
y = -6,9979x + 14093
80 60
y = -0,5012x + 1055,6
main roads (6)
y = -2,4882x + 5012,7
NO2 (µg/m³)
NO as NO2 (µg/m³)
100
50 y = -0,9608x + 1957,9
40 30
40 20 0 1990 1992 1994 1996 1998 2000 2002 2004
urban traffic influence (5) urban background (18) rural background (4)
20 10 0 1990 1992 1994 1996 1998 2000 2002 2004
Figure 3a, b): Development of a) NO and b) NO2 air concentrations in Baden-Württemberg 19902005 (annual means, number of monitoring sites in parentheses) Figure 3: Evolution des niveaux a) de NO et b) de NO2 à Baden-Württemberg de 1990 à 2005 (moyennes annuelles, nombre de stations en parenthèses) Nitrogen oxides are emitted from combustion processes mainly as NO and subsequently oxidized to NO2, usually in the atmosphere. Therefore the ratio NO2/NOx increases with increasing distance from the emission source. The analysis of the measured NO2/NOx ratios shows pronounced increases at the urban and traffic influenced monitoring sites (figure 4). At main roads, in 1990 about 25% of the NOx consisted of NO2 (Rabl & Scholz, 2005), whereas in 2005 the proportion of NO2 was increased to more than 40%.
4 NO2/NOx ratio in Baden-Württemberg
1.0
0.8 rural background (4)
0.6
urban background (18)
urban traffic influence (5)
0.4
main roads (6)
0.2
0.0 1990 1992 1994 1996 1998 2000 2002 2004
Figure 4: Figure 4:
Development of the NO2/NOx ratio in Baden-Württemberg 1990-2005 (annual means, number of monitoring sites in parentheses) Evolution de la relation NO2/NOx à Baden-Württemberg de 1990 à 2005 (moyennes annuelles, nombre de stations en parenthèses)
A similar picture is obtained from the analysis of the ozone trend (figure 5). For the site type "rural background", a slow decrease with annual variations is observed, whereas in the cities - at a lower level - a distinct increase in the annual mean is recorded, suggesting a gradual adjustment of the urban and the rural situation.
Ozone concentrations in Baden-Württemberg 90 80
µg/m³
70 60
rural background (4)
50
urban background (18)
40
urban traffic influence (5)
30
main roads (2)
20 10 0 1990
1992 1994
Figure 5: Figure 5:
1996
1998
2000 2002
2004
Development of ozone concentrations in Baden-Württemberg 1990-2005 (annual means, number of monitoring sites in parentheses) Evolution des concentrations d'ozone à Baden-Württemberg de 1990 à 2005 (moyennes annuelles, nombre de stations en parenthèses)
The increase of the ozone annual mean values in cities is not interpreted as an enhanced formation, but as a reduced degradation of ozone due to reduced emissions of primary pollutants, mainly NO.
5
3 - First explanation: Photochemical balance The parallelism in the development over time of the NO2/NOx-ratio and ozone points to a photochemical correlation between nitrogen oxides and ozone. A simple approach is the wellknown photo-stationary equilibrium: NO2 + O2
NO + O3
According to the law of mass action the following relation results: [NO] · [O3]
with k = constant of equilibrium.
k= [NO2]
Figure 6 shows that this explanation applies quite well. The product [NO] · [O3] is plotted against the corresponding NO2-concentration for the air quality monitoring sites of Baden-Württemberg between 1990 and 2003. Each point in the graph represents the mean values of one site and one year. A satisfying linear correlation is obtained for urban and main road sites (with k ≈ 35). The deviation in rural background is probably caused by the measuring inaccuracy of low NO-/NO2concentrations. This result underlines the photochemical balance between NO, NO2 and ozone, which is observed with annual mean values, but also with daytime values (shown in Rabl & Scholz (2005)). 3000
Law of mass action at measuring sites in Baden-Württemberg 1990-2003 main roads (2)
2500
NO*O3 (µg/m³)²
urban traffic influence (5)
2000
urban background (18) rural background (4)
1500
y = 34,997x - 318,1
Trend all sites except rural BG
R2 = 0,74
1000
500
0 0
10
20
30
40
50
60
70
NO2 (µg/m³)
Figure 6: Figure 6:
Relationship between NO, NO2 and ozone according to the law of mass action at air quality monitoring sites in Baden-Württemberg 1990-2003 (annual means). Equilibre photochimique entre NO, NO2 et ozone aux stations de qualité de l'air à Baden-Württemberg de 1990 à 2003 (moyennes annuelles).
The importance of this relationship for the situation in the cities is that the three components are tightly coupled and changes in one component affect the others. The increased ozone concentrations due to reduced NO emissions represent an elevated oxidative potential in the atmosphere, resulting in a constant or even enhanced formation of NO2 according to the product [NO] · [O3]. 4 - Second explanation: Changes in diesel vehicle exhaust emissions The NO2/NOx-ratio in the exhaust emission of diesel cars has also changed over the last years. Modern diesel cars, which have been significantly increasing in number for several years, are equipped with an oxidation catalytic converter in order to keep within the emission limits for HC
6
and CO. As a side reaction, these catalytic converters oxidize NO to NO2, with the result that a considerable part of the vehicle's NOx emission is emitted directly as NO2. This oxidation of NO to NO2 will probably be performed by any noble metal catalyzed exhaust treatment device, as well as by catalytically active particle filters. Unfortunately, up to now the database concerning reliable NO2 (not NOx) emission data of modern cars and exhaust treatment equipment is insufficient. An experimental proof for the changed NO2/NOx-ratio in the exhaust emission can be deduced from the results of a measuring campaign with two measuring sites positioned at the same location on opposite sides of the highly frequented national road "B 10" in Karlsruhe. The winter situation presented in figure 7 is characterized by: a) wind in constant windward-leeward-direction, perpendicular to the road, with an average wind speed of 1-3 m/s, b) a very low ozone concentration of 1 µg/m³ on average at both sides. Under these conditions, the concentration differences between the two measuring sites represent the contribution directly produced by the road traffic. From the NOand NO2-differences a NO2/NOx-ratio of about 0.2 – 0.3 is obtained, which means that about 20% of the NOx is directly emitted as NO2 during the daytime (with a high proportion of trucks) and about 30% of the NOx is emitted as NO2 at night (mainly cars). This is far more than the proportion of 5% which has been assumed for a long time as the direct NO2 emission of vehicles. NO2/NOx B10 3./4.12.2003 0,5
NO2/NOx
0,4 0,3 0,2 0,1
Figure 7: Figure 7:
04.12.2003 24:00
04.12.2003 22:00
04.12.2003 20:00
04.12.2003 18:00
04.12.2003 16:00
04.12.2003 14:00
04.12.2003 12:00
04.12.2003 10:00
04.12.2003 08:00
04.12.2003 06:00
04.12.2003 04:00
04.12.2003 02:00
03.12.2003 24:00
03.12.2003 22:00
03.12.2003 20:00
03.12.2003 18:00
03.12.2003 16:00
03.12.2003 14:00
03.12.2003 12:00
03.12.2003 10:00
03.12.2003 08:00
03.12.2003 06:00
0,0
NO2/NOx-ratio for the concentration difference between opposing measuring sites on the national road "B 10" in Karlsruhe (details see text) Relation NO2/NOx pour la différence de concentrations de deux stations de mesure opposé à la route nationale "B 10" à Karlsruhe (détails en texte)
Conclusion The changes in diesel vehicle exhaust emissions seem to be the main reason for the slow decrease or even increase of NO2 concentrations at highly traffic-influenced sites. Since diesel powered cars are still increasing in number, and NOx, but not NO2-emissions are regulated, it is very probable that in 2010 the EU limit values for NO2 will be exceeded at locations with heavy traffic burden. Only a further reduction of motor vehicle NOx emissions or a separate limitation of NO2 emissions will be effective in reducing the NO2 air concentrations. References
Rabl P. & W. Scholz (2005): Wechselbeziehungen zwischen Stickstoffoxid- und Ozon-Immissionen, Datenanalysen aus Baden-Württemberg und Bayern 1990-2003. Immissionsschutz, Heft 1, p 21-25. UMEG (2004): Emissionskataster Baden-Württemberg, www.emissionserfassung.de/emi/ee_index.html Umweltbundesamt, Germany (2004): Berichterstattung Emissionen
