Air pollution dispersion model

 Air quality dispersion model

Models may be described according to the chemical reactions involved. So-called nonreactive models are applied to pollutants such as CO and SO₂ because of the simple manner in which their chemical reactions can be represented. Reactive models address complex multiple-species chemical mechanism common to atmospheric photochemistry and apply to pollutants such as NO, NO2, and O₃.

Models can be described as simple or advanced based on the assumptions used and the degree of sophisticated with which the important variables are treated. Advanced models have been developed for such problems as photochemical pollution, dispersion in complex terrain, long-range transport, and point sources over flat terrain. The most widely used models for predicting the impact of relative unreactive gases, such as SO₂, released from smokestacks are based on Gaussian diffusion.

In Gaussian models, the spread of a plume in vertical horizontal directions is assumed to occur by simple diffusion along the direction of the mean wind. The maximum ground level concentration is calculated by means of the following Equation.

The parameters σ_y and σ_z describe horizontal and vertical dispersion characteristics of a plume at various distances downwind of a source as function of different atmospheric stability conditions. Values are determined from the graphs found n the figure.

The effective stack height H is equal to the physical stack height (h) plus the height of the plume (plume rises, Δh) determined from where the plume bends over. Plume rises must be calculated from model equations before the effective stack height can be calculated.

For purposes of illustration, let us determine the ground level concentration (C_x) at some downwind distance (x). For the following conditions let us calculate the ground level concentrations at 10 km directly downwind.

A power plant burning 9 tons of 2.5% sulfur coal/hr emits SO₂ at a rate of 113 g/sec. The effective stack height is 100 m, and the wind speed is 3 m/sec. It is 1 hour before sunrise, and the sky is clear. Since the off centerline distance (Y) in this case is equal to O, the following equation reduces to:

From table 1, the atmospheric stability classes for the condition described is F. It represent a nighttime condition with <37.5% cloud cover. The horizontal dispersion coefficient σ_y for a downtime distance of 5 km for atmospheric stability class F is approximately 90 m (figure 1); the vertical dispersion coefficient σ_z is approximately 20 m (figure 2)

Therefore :

The ground level concentration of SO₂ from this source would be approximately 44 μg/m³ under the conditions given.

Although the use of air quality models is the subject of considerable controversy, there's a general agreement that there a few alternatives to the use of models, particulately to make decisions on an action which is know in advance to pose  potential environmental problem. The debate arises as to which models should be used, and the interpretation of models results. The underlying question such in debates is how well, or how accurately, does the model predict concentrations under the specific circumstances, since model accuracy may vary from 30% to a factor of 2 or more? If a model is conservative , i.e., it over-predicts ground level concentrations, a source may be required to install costly control equipment unnecessarily. Less conservative models may under-predict concentrations and thus violations of air quality standards may occur. The uncertainty associated with input variables, such as wind data, and source emission data. Such data are usually estimated and not well documented.

Source: Air Quality 2nd edition, Thad Goddish