Info

Distance along the cross-section (km) San Bernardino Ocean Mountains

FIGURE 16.25 Model-predicted O (in pphm, where 1 pphm — 10 ppb) west to east across the Los Angeles area South Coast air basin and as a function of altitude and time for August 27, 1987 (adapted from Lu and Turco, 1996).

mation. There are a number of reasons for this, but increased thermal decomposition of PAN and H02N02 and increased biogenic emissions are believed to be significant contributors.

Clearly, the meteorological inputs have a critical impact on model predictions and effects in uncertain ties in these inputs need to be considered in assessing model predictions.

(2) Initial and boundary conditions As the name implies, initial conditions are the initial concentrations of all of the species considered in the model. Since transport from one cell to the next is also an important determinant of subsequent concentrations, boundary conditions, that is, the concentrations of species entering the region being modeled, also need to be specified.

In practice, models are usually run to simulate a multiday episode. In this case, the effect of the initial concentrations is reduced in the later days of the simulation due to the chemistry and transport that takes place, but their effect can be detected on day 2 and sometimes on day 3 of the simulation. The assumed boundary conditions, however, can be more important since they represent direct transport into the region. When one models the effects of future control strategies, there are a number of possible assumptions one can use for boundary conditions. For example, one could decrease the emissions in the modeling region but assume that the boundary conditions remain the same, or alternatively, the boundary conditions could be proportionately decreased if it is assumed that the emissions reductions in the region modeled also apply to sources outside the region that are transported in.

Figure 16.26, for example, shows ozone isopleths predicted for the Los Angeles area using the CfT Eulerian model under two different assumptions for the boundary and initial conditions (Winner et al., 1995). In Fig. 16.26a, the pollutant concentrations measured in the past were used to set the boundary and initial conditions. In Fig. 16.26b, concentrations measured in clean air over the Pacific Ocean upwind of the modeled area were used instead. The axes are given in terms of the percentage of VOC or NOx emissions compared to the baseline level on August 28, 1987, rather than in terms of absolute concentrations as for some of the isopleths seen earlier. Not only are the predicted ozone concentrations for a given percentage reduction in the precursors different, but the shapes of the isopleths also change. Under the assumption of clean boundary and initial conditions, the isopleth shape is similar to the traditional isopleths derived from box models and environmental chamber studies (e.g., see Fig. 16.14). However, they become more L-shaped when the historical boundary and initial conditions are used. This is because the highest 03 concentrations occur downwind due to the delayed production of 03 at low VOC/NOx ratios, so that the isopleths are similar to those shown in Fig. 16.16 for the eastern end of the air basin.

In a similar vein, Kuklin and Seinfeld (1995) have shown that the predicted reductions in VOC needed to attain the federal air quality standard of 0.f2 ppm for 03 in the Los Angeles area for a given level of NOx emissions can differ by almost a factor of two, depending on the initial boundary conditions assumed.

The critical point is that the optimal control strategy derived from such models will be clearly impacted considerably by which set of isopleths, i.e., which set of boundary and initial conditions, is chosen.

(3) Emissions If the meteorology, chemistry, and emissions submodels are all reasonably correct, one should be able to predict the concentrations of various species, particularly the organics that are directly emitted. An alternative that removes some of the uncertainties associated with the meteorology is to compare measured with predicted ratios of compounds.

0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

% of baseline VOC emissions % of baseline VOC emissions

FIGURE 16.26 Effect of assumed boundary and initial conditions on ozone isopleths in the Los Angeles area for reductions in VOC and NOA. from August 28, 1987 levels: (a) using historical values of measured concentrations; (b) clean air values for boundary and initial conditions (adapted from Winner et al., 1995).

0 10 20 30 40 50 60 70 80 90 100 0 10 20 30 40 50 60 70 80 90 100

% of baseline VOC emissions % of baseline VOC emissions

FIGURE 16.26 Effect of assumed boundary and initial conditions on ozone isopleths in the Los Angeles area for reductions in VOC and NOA. from August 28, 1987 levels: (a) using historical values of measured concentrations; (b) clean air values for boundary and initial conditions (adapted from Winner et al., 1995).

Was this article helpful?

0 0

Post a comment