1. Topic

2. Introduction

Dispersion models require data on what and how much pollutant is being released into the atmosphere and from what location. More advanced models require information on how the pollution is being released and any variations in time. For time series modelling meteorological data over time will be required.

3. Discussion

The data required for dispersion models can be broadly summarised as:

· Map: For the more advanced models which can deal with several sources a digital map of the area is required. This enables the point and line sources to be accurately located. It may also have layers of additional information such as the land and building height and the surface type.

· Meteorology: wind speed, wind direction, temperature and stability determine how emissions disperse. Some models will require the stability category whereas more modern models calculate vertical turbulence and mixing heights based on other parameters such as cloud cover or vertical wind speed. The frequency of data required will depend on the air quality standard, i.e. hourly limits will require hourly average met data.

· Local environment: complex topography and surface roughness is used to calculate the ground turbulence. The built environment can also effect dispersion so building heights may also be required. It is also necessary for a model to predict concentrations at defined locations. This may involve defining a grid over the area where the pollution concentration is calculated in each grid square. This enables the model to display iso-lines of pollution concentrations.

· Emission from point sources: These are usually chimney stacks. The substance, rate of release, temperature of gas, velocity of rise, stack dimensions, and time of release may all be required plus the dimensions of the building if the model can calculate the effect of down wash from the stack.

· Area sources: These have similar data requirements to point sources but without the dimensions of a chimney stack.

· Emission data from line sources: These will usually be road transport although in some areas rail may be significant. The data required includes traffic flow, vehicle type, vehicle speed, daily/monthly/annual variation, location and length of roads, and emission factors. Some models may also be able to deal with street canyons, elevation or cuttings so the dimensions of the streets may also be required.

· Background concentrations: Dispersion models can only predict ground level concentrations from sources which have been included in the model. There may be situations where pollution is coming in from outside the area. There may also be background sources such are housing areas or minor roads which have a small but significant effect. These may be important when air quality objectives are exceeded. See Further Reading.

4. Recommendation / Conclusion

· Depending on the model selected contact the department responsible for industrial processes and / or transport data. If necessary, find the nearest available meteorological monitoring station. If there are no local weather stations weather data may need to be monitored by the municipality itself. Monitored data will be needed to correlate with the model especially if used to calculate the background concentration. Information from outside the area may be useful for calculating background concentrations.

· There will inevitably be large assumptions made with data types which will lead to uncertainties. All models will only be as accurate as the data input.

5. Examples / Further Reading

• Meteorological data required for Airviro and ADMS models

• The use of meteorological data in dispersion models in Birmingham

6. Additional Documents / Web Links

· LAQM.TG2(00) - Review and assessment: Estimating emissions http://www.defra.gov.uk/.

· LAQM.TG3(00) - Review and assessment: Selection and use of dispersion models http://www.defra.gov.uk/.

Last Updated

25th January 2005