A new chemical kinetics mechanism for simulating urban and regional photochemistry has been developed and evaluated. The mechanism, called the Carbon Bond Mechanism IV (CBM‐IV), was derived by condensing a detailed mechanism that included the most recent kinetic, mechanistic, and photolytic information. The CBM‐IV contains extensive improvements to earlier carbon bond mechanisms in the chemical representations of aromatics, biogenic hydrocarbons, peroxyacetyl nitrates, and formaldehyde. The performance of the CBM‐IV was evaluated against data from 170 experiments conducted in three different smog chambers. These experiments included NOx‐air irradiations of individual organic compounds as well as a number of simple and complex organic mixtures. The results of the evaluation indicate substantial improvement in the ability of the CBM‐IV to simulate aromatic and isoprene systems with average overcalculation of ozone concentrations of 1% for the aromatic simulations and 6% for the isoprene simulations. The mechanism also performed well in simulating organic mixture experiments. Maximum ozone concentrations calculated for 68 of these experiments were approximately 2% greater than the observed values while formaldehyde values were low by 9%.

No abstract

Three chemical mechanisms recently developed for use in urban‐ and regional‐scale oxidant models have been compared. Approximately 400 simulations were conducted using an atmospheric model to determine the degree to which the mechanisms yield comparable predictions of ozone and other oxidants. The simulations were carried out over a range of initial conditions for a number of scenarios representative of urban and rural environments. Sensitivity studies were conducted to assess the effect of varying parameters, such as the initial ratio of volatile organic compounds (VOC) to oxides of nitrogen (NOx), the composition of the VOC mixture, and the temperature. The mechanisms were found to yield nearly identical predictions for ozone, peroxyacetylnitrate, and nitric acid under most conditions. Differences, however, were noted in simulations with (1) high NOx levels, (2) high levels of aromatic hydrocarbons, and (3) low afternoon temperatures. Predictions of hydrogen peroxide were in good agreement except at high VOC/NOx ratios where differences of as much as a factor of 3 were found.

No abstract