Box Modeling

Modeling the Mexico City Outflow : Simulations with the NCAR Master Mechanism


The chemical evolution of Mexico City outflow air was modeled with the NCAR Master Mechanism, a box (0-dimensional) model with detailed gas phase chemistry. The model was run for the environment described in Table 1, with initial (midnight of first day) chemical concentrations as given in Table 2. Photolysis coefficients (Table 3) were calculated using the Tropospheric Ultraviolet-Visible (TUV) model, version 4.2. The model simulates the time dependence of the chemical composition of the initial air parcel, assuming no additional emissions, no dilution, and no heterogeneous processes.


Figures 1 shows the computed concentrations of various species. Figure 2 contains non-radical species. Figure 3 shows the pseudo-first order rate coefficient for the removal of OH by various species X, i.e. kOH/X[X]. Figure 4 shows the rates of formation of radicals by photolytic processes.


The complex behavior on day one (Fig 1) is due to both the high initial hydrocarbon and NOx concentration, and to "spinning-up" of the concentrations of oxidation intermediates. Days two through five show much more regular diurnal cycle patterns. Daytime values of the radicals OH, HO2, RO2, and RCOO2 are essentially unchanged over days two thorugh five, with maximum local noon values OH ~3e6, HO2 ~1.5e9, RO2 ~2.5e9, and RCOO2 ~2.5e8 molec cm-3 (Fig 1). Nighttime concentrations of these radicals are smaller but not negligible (esp. for the RO2 and RCOO2), and show a decreasing trend over days two through five. Daytime NO and nighttime NO3 radical concentrations are similar (~1e9 molec cm-3 on days one through two) but decrease gradually decrease gradually over the five days, tracking the decrease in the concentration of peroxyacyl nitrates (PANs, Fig 2).

O3 reaches a maximum of ~400 ppb on the first day (Fig 2), decreasing to ~300 ppb by day five. The other major inorganic products are HNO3 and H2O2 (50 and 30 ppb respectively, at end of day five), while organic products include ketones (120 ppbv), organic hydroperoxides (40 ppbv), aldehydes (15 ppbv including six ppb of CH2O), organic acids and peracids (15 and 10 ppbv respectively), organic nitrates (14 ppbv), and peroxyacyl nitrates (PANs, 3.3 ppbv). NO2 is seen to be in steady-state with PANs, decreasing to about 0.5 ppb by the end of the simulation. The decreases in PANs and NO2 are very sensitive to temperature (as confirmed in separate simulations) due to PANs decomposition. Table 4 summarizes the major product species.

Figure 3 shows the pseudo-first order rate coefficient for the removal of OH by various species X, i.e. kOH/X[X]. After several days, this reactivity is dominated by non-methane organic compounds (nmVOCs), which include the non-methane hydrocarbons (nmHCs) and the intermediates of their oxidation. Although the reactivity of the nmHCs decreases rapidly (especially in the first day), the total organic reactivity is seen to persist for the entire simulation time, and exceeds that of CO. Methane reactions are relatively unimportant for these conditions, and NO2+OH is significant only during the first day.

The photolysis of organic compounds dominates on the first day, while the photolysis of O3 (to O1D, followed by its reaction with H2O) dominates on days two through five (Fig 4). The photolysis of H2O2 increases in importance as the concentration of H2O2 grows over the five days. Dilution with cleaner drier air masses, though not considered in this simulation, is expected to increase the importance of organic and H2O2 photolysis relative to O3 photolysis, because the latter depends on both O3 and H2O concentrations.

Implications for MIRAGE-Mex

  1. Large concentrations of intermediates from the gas phase oxidation of hydrocarbon are expected in the outflow. These include ketones, aldehydes, PANs, alkohols, as well as organic peroxides, nitrates, and acids. The predicted concentrations range from a few ppbv to several tens of ppbv, and are well above the detection limit of current instruments.
  2. The reactivity (with OH) of these intermediates persists for at least 5 days, and sustains the production of peroxy radicals (HO2, RO2, and RCOO2). These peroxy radicals contribute to the production of O3 and H2O2.
  3. H2O2 is expected to be a major gas phase product in the outflow, with concentrations reaching several tens of ppb even in the presence of ppb-level NOx concentrations.
  4. Photolysis of the organic intermediates is a major source of new radicals.
  5. Gas-aerosol interactions could be important to both phases: Removal of low-vapor-pressure intermediates could reduce their importance to the gas phase radical budgets, while at the same time contributing to the organic content of the aerosols.
  6. Substantial nighttime concentrations of NO3, RO2, and RCOO2 are expected. In the case of RO2, the large nighttime values result partly from slower gas-phase removal (compared to daytime removal by NO and HO2). Other less well-known losses of RO2, e.g. reactions with aerosols, could become relatively more important at night.