Ozone/Precursor Relationships in the Detroit Metropolitan Area Derived from Captive-Air Irradiations and an Empirical Photochemical Model

Abstract

The ozone generating potential of ambient Detroit air was characterized using transparent bags as outdoor smog chambers. Air samples gathered in downtown Detroit on 12 sunny days during the summer of 1981 were irradiated and ozone production was measured throughout the day. Morning additions of nonmethane organic compounds (NMOC), nitrogen oxides (NO _x ), or clean air were made to some bags to assess their effects and determine how different emission control strategies will perturb the ozone formation process. The results show that the photochemical process is hydrocarbon limited and the most effective ozone reduction strategy is one that reduces NMOC while leaving NO _x unchanged. Reductions in NO _x will likely increase maximum ozone in the Detroit area since NO _x additions were found to suppress ozone formation in the outdoor smog chambers.

The smog chamber results were compared to predictions of the EPA EKMA model that was used by the State of Michigan to forecast the effects of NMOC and NO _x controls. The model provided a reasonable approximation for the ozone maxima in the chambers and predicted the relative changes in ozone due to changes in NMOC or NO _x . The observed NO _x -inhibition and NMOC-enhancement effects on the ozone maxima in the bags were correctly described by the model. However, the model underpredicted ozone formation for the limited number of mixtures diluted initially with clean air. Finally, the effects of experimental error in a key input to EKMA (the early morning NMOC-to-NO _x ratio in the downtown area) were small enough that they had a minimal impact on the results. Thus, our analyses support the use of EKMA in Michigan's State Implementation Plan.