|
| Block 8: Environmental Issues and Public Health - Air Pollution Chapter 6: Methods Of Assessing Air Quality - Measuring, Monitoring And Modelling Ambient Pollutant Concentrations. |
 |
 |
 |
 |
 |
Air quality modelling, particularly dispersion modelling, is used to predict air pollution concentrations in the modelled region using data on pollutant emission sources and meteorology as inputs. There are several different types of atmospheric models, ranging from a simple input-output ‘box’ model to very complex models (and difficult to develop and use) that attempt to account for dispersion, atmospheric chemistry and local terrain variations. Statistical dispersion models are intermediate in complexity and ease of use, and the most widely used. Air quality models are mathematical procedures that combine emission data from one or several sources (some models are capable of handling several thousand sources, including stationary and mobile sources) and meteorological data to predict ambient concentrations of pollutants as a function of time and location. The output of all models should be validated, and calibrated if necessary, against data from appropriately located monitoring stations. Dispersion models are able to interpolate values between monitoring stations, and may be used to predict (forecast) near-term concentrations, providing a coherent integrated picture of the link between the sources of pollution and the ambient air quality. Atmospheric models are therefore as essential aspect of air quality assessment, but their limitations must understood and accounted for. The calibration and validation of models is essential.
A great value of air pollution modelling is that it can be used to evaluate the effectiveness of interventions such as the use of cleaner fuels, the us gas cleaning equipment on factory stacks or the emission reduction benefits of improving public transport and simultaneously reducing private vehicle use. Modelling may also be used to assess the likely impact of new industrial developments. Model results may thus be used to make air quality management decisions.
The validity of a dispersion model is dependent on the quality of pollutant source data. As in the case of an Emission Inventory, the necessary regulatory and administrative infrastructure for the periodic collection of source data is essential.
Air pollution modelling may also be used to study the potential impact of a single emission source, illustrated in the following example.
Consider the impact of a large industrial boiler using coal as a fuel source. For modelling purposes, assume a stack height of 40m, a stack exit velocity of 6m/s and an exit gas temperature of 300 ºC. A dispersion model (in this case a simplified statistical (Gaussian) model was used) is able to predict concentrations in a specified area, for given meteorological conditions.
Figures 6.2 and 6.3 illustrate the substantial differences in ambient concentrations as a function of downwind location, and the location of the point of highest concentration, that may occur due to differences in wind speed and atmospheric stability. (Stable or very stable conditions may occur at night, particular during winter, under conditions of low wind speed. Unstable conditions may occur during the day with strong insolation (heating by the sun) and moderate wind speed.) Under stable or poor dispersion conditions (Figure 6.2), maximum concentrations are high (about 28 ug/m3) and occur over a comparatively large area; under unstable conditions (good dispersion, Figure 6.3), maximum concentrations are lower (about 20 µg/m3) and occur over a smaller area. Note also that the point of maximum concentration is not at the source but is some distance downwind because the pollutants are discharged at a height of 40m. Note also that the location of the point of maximum concentration is substantially different for the two cases.
 |
| Figure 2: Case 1: Emission rate 2 g/s, 40 m stack, 10 km/h wind, atmospheric stability class f |
 |
| Figure 3: Case 2: Emission rate 2 g/s, 40 m stack, 20 km/h wind, atmospheric stability class a |
Basic dispersion models account for the influence of meteorology – wind strength and direction, cloud cover - and source strength and other characteristics such as release height, temperature and exit velocity. More complex models are needed to account for atmospheric chemistry and the influence of terrain on wind direction.
For modelling purposes, assume a stack height of 40m, a stack exit velocity of 5m/s and an exit gas temperature of 300 ºC. A dispersion model (in this case a simple Gaussian plume model was used) is able to predict concentrations in a specified area, for stable (Stability Class F, 8km/h wind speed), Figure 6.4a) and moderately unstable conditions (Stability Class B, 16km/h wind speed), Figure 6.4b.
 |
Figure 6.4a: Pollutant concentrations for Stability Class F, 8km/h wind speed |
The change in ambient concentrations due to a change in meteorological conditions may be illustrated by modelling the same emission source under neutral or unstable conditions:
 |
Figure 6.4b: Pollutant concentrations for Stability Class B, 16km/h wind speed) |
The maximum ambient concentrations are about 35% lower in the second case; the region subjected to the highest concentration (about 36 µg/m3) is much smaller compared to the very stable case.
Wind direction plays an even greater role - in both case the concentrations upwind are essentially zero compared to the significant concentraiuons downwind of the source.