 |
NARSTO Workshop on Innovative Methods for Emission Inventory Development and Evaluation University of Texas, Austin October 14-17, 2003 |
 |
Stochastic Modeling Of Ambient Ozone Formation With Emission Variability
Yosuke Kimura, Junsang Nam, David Allen
University of Texas
Variability in emissions from large point sources can cause variability in the processes that control ambient ozone formation and the potential effectiveness of control strategies, if the control strategies are based on annual average data on emissions of reactive hydrocarbons and oxides of nitrogen. Ambient measurements and field studies in the Houston-Galveston area have shown that extreme variability in point source reactive hydrocarbon emissions can occur.
This presentation will describe the relationships between “upset” release events and extremely high ambient ozone concentrations. A stochastic approach to modeling emissions has been used to model emission variability from multiple sources. Variability in emissions from individual sources is first characterized as random phenomena, with average values and standard deviations of emission rates based on long-term monitoring data. The randomized emissions are then used as an input for a Monte-Carloanalysis where a photochemical simulation is repeated for thousands of emission scenarios, in order to derive the distribution of ozone concentrations expected with variability in precursor emissions. The modeling framework is used to examine the effectiveness of various control strategies.