~nnwphcric Eiwirmm~~nt zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Vol. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 20, No.8. pp. 1589-15%. 19% wo6-6%l~t3.00+0.00 Printed in Orcat Britain. PapmoaJourrulr Lti. SIMULATIONS OF THE OZONE FORMATION CAUSED BY TRAFFIC IN URBAN AREAS K. ZELLNER and N. MOUSSIOP~ULOS* Institut fiir Tcchnischc Thcrmodynamik, Univcrsitit Karlsruhe, D-7500 Karlsruhe, W. Germany (First received 26 July 1985 und receivcdfir publication 21 zyxwvutsrqponmlkjihgfedcbaZYXW February 1986) Abstract-A box model and a vertical all model for simulations of the ozone formation causal by trafEcin urban areas arc presented.Diurnal variations of emission rates arc calculated on the basis of assumed diurnal trafliccycles. In both models the critical case of vanishing horizontal adva%ivc transport is considered. In the vertical all model turbulent diffusion is described using suitable paramctcrizations for the aidy diffusivity and the mixing layer depth. Photochemical smog formation is modeled using alternatively a compact and a detailed reaction mechanism. The compact mechanism of Eschcnrocda and Martinez (EM mechanism, Ado. Chum. 113, 101-167, 1972) can be regarded as an adequate basis for a qualitative description of photochemical transformations in the case that no detailed data on hydrocarbon emissions are available. Predictions of the 0, formation using this mechanism are fairly similar to corresponding predictions achieved with the detailed reaction mechanism of Atkinson, Lloyd and Winges (ALW mechanism, Atmospheric Environment 16,1341-1355,1982). Hence, ifthc ALW mozhanism is valid, the EM mazhanism may be used to obtain a reasonable impression of the diurnal variation of the OS formation in an urban airshed.The simulation results illustratethe effectsof traffic density and verticalturbulent diffusion on the 0, formation in such an airshcd. Key word index: Photosmog modeling, ozone formation, box model, all model, motorcar emissions. zyxwvutsrqponm 1. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA INTRODUCHON of a compact mechanism as a basis for a qualitative Increased hydrocarbon (HC) and nitrogen oxide description of photochemical transformations might be of interest in the case when no detailed information (NO,) emissions lead to a considerable contribution of on the subdivision of HC emissions is available. The photochemical smog to environmental pollution. Examples for high ozone concentrations observed in comparison of the chosen compact reaction mechan- ism with the more detailed reaction mechanism shows urban areas of the Federal Republic of Germany during fair weather periods in summer are shown in that the prediction of ozone formation is fairly similar for the two me,-.hnisms. Figs 1 and 2 (Rudolf, 1979; Umweltbundesamt Berichte, 1983). High 0, peak concentrations in urban areas are accompanied by increased levels of the ozone concen- tration in the low mountain range (e.g. Bavarian Forest, Black Forest). Elevated permanent concen- trations and occasional OJ peaks in wooded areas are regarded as stress factors for plants, especially for trees (Ashmore et al., 1985; McLaughlin, 1985). Often the maximum O3 concentrations observed in forests can be related to windward urban areas. Due to the preference of individual traffic to public transport, private cars are the main contributors to HC and NO, emissions in urban areas. For photochemical smog formation it is of special importance that motor vehicles emit both reactive HCs and NO,. In this paper two models for the simulation of photochemical smog formation caused by automotive exhaust in urban areas are presented. Photochemical smog formation is modeled using alternatively two reaction mechanisms of different complexity. The use *To whom correspondence should be addressed. ,o 0 zyxwvutsrqponmlkjihgfedcbaZYXWV 6 12 16 Time (h) -_ Fig. 1. NO-. NO?- and O,-coneentrations observed in the city of Frankfurt on 4 July 1973. 1589