Title of Example

Example

Introduction

Birmingham is a large metropolitan conurbation with many sources of pollutants to atmosphere distributed throughout the city including many industrial processes and a large road network carrying traffic.

Birmingham is fortunate in benefiting from an extensive monitoring network measuring a variety of pollutants and providing a wealth of up to date and historical data. High quality monitoring equipment is expensive to acquire and to run; therefore the number of places which data can be collected is inevitably restricted. Also even the most sophisticated pollution monitoring can only give a snap shot of the situation over a particular period of time at one specific location. Dispersion models offer an alternative way of describing air quality over a large geographical area, allowing the prediction of future conditions and source apportionment.

The computer based atmospheric dispersion model supplied by INDIC Airviro has been utilised by Birmingham City Council (BCC) throughout the Review and Assessment process of air quality. The output of the modelling work along with monitoring data has enabled Birmingham City Council to make more-informed decisions in relation to air quality. This example considers how Birmingham City Council has modelled nitrogen dioxide using a dispersion model.

Modelling of Nitrogen Dioxide in the City of Birmingham.

Nitrogen dioxide (NO₂) has been highlighted as the main pollutant of concern throughout Birmingham and a lot of modelling has been done for this pollutant. The specific sources of NO₂ in Birmingham are:

· Emissions from motor vehicles (line sources)

· Emissions from industrial combustion plant (point/area sources)

· Emissions from commercial combustion plant (point sources)

· Emissions from space heating equipment in homes, hospitals and schools (point sources)

Initially two emission databases incorporating all the different sources were established, one for the base year and a second for the predicted emissions information for 2005 (the compliance year for the objective).

The levels of NO₂ have been calculated and predicted using the Airviro model, for both present time and for 2005. The model has been used to calculate two sets of predictions. Firstly, a ‘time series’ of hourly values over a full calendar year, from which average and high percentile values could be derived for specific points, and secondly, annual ‘scenario’ calculations, which were used to illustrate the distribution of NO₂ across the city.

Initial predictions suggested that there was no requirement to declare any Air Quality Management Areas, with respect to the annual average levels of NO₂. However, given that the predicted values alongside some roads including the M6 motorway and either side of the A38 (M) were very close to the objective, it was recommended that further investigations should be carried out.

By combining local knowledge and experience with the results of air pollution measurements and computer predictions, a number of ‘near worst case’ conditions were identified, for further investigation.

Modelling of NO₂ in Intensive Study Areas

Initial computer modelling was based on the full emissions information, but calculated at a medium spatial resolution of 250 x 250m. This allowed potential problem areas to be easily highlighted for further detailed assessments, or intensive study areas (ISAs).

Six areas were selected from the ‘near worst case’ situations identified. Each area comprised a half to one kilometre road links in which four or five nitrogen dioxide diffusion tubes are exposed.

All six areas were chosen to represent where people are likely to be exposed for the relevant period. The monitoring locations have been selected as being representative of near-worst case conditions, where the public are likely to be exposed.

In preparation for more detailed computer modelling of the predicted levels of nitrogen dioxide, for 2005, the information used to model air pollution was updated, wherever possible.

The road traffic data, for the six areas of interest, were cross-checked against Birmingham City Council’s Transportation Department’s and the Highways Agency’s current road traffic data.

The revised emission factors for road traffic published by the the UK National Government ( The Department of the Environment, Food and Rural Affairs (Defra) in 2002) were used in the computer modelling. The revised emission factors reflect that vehicles travelling slowly tend to emit more pollution than vehicles travelling at moderate speeds. Hence congested streets with slow moving traffic will be a cause for concern.

The road links were re- aligned to centrelines of the carriageways and the most important roads were digitised to follow their curves. This ensures that the model reflects the actual situation, and increases the emissions on some curved road links, as following the curve increases the total distance travelled.

25 x 25m grid maps were created in all of the ISAs and all other areas of interest where possible. The maps are then used to model these areas at a finer resolution of 25m. For NO₂ the modelling of the area is carried out using only road source and for the year 2005 (target year for the NO₂ objective). After completion of the model the multiplication factor derived from the verification process (of the model) is applied to the output and an appropriate background is added.

The background is obtained from a model run of the entire West Midlands at a 25m resolution with all sources included. Ten to fifteen grid squares which are approximately 0.5km (or at a distance at which the impact of the road is thought to be significant) from the road of interest are averaged. The model was also verified at background sites and the multiplication factor derived from the process is applied to the model output. Hence the model has been adjusted by comparison to background monitoring sites.

The model outputs obtained are exported as shape files from Airviro. These shape files can then be imported into any Geographical Information System (GIS) package for improved presentation. In the case of Birmingham the maps were imported into a package called Arc GIS.

Conclusions of the further modelling of NO₂ in Intensive Study Areas

The computer modelling indicates a wider spread of pollution than the measurements indicate. This seems to be a general feature of Gaussian dispersion models, and further investigations are in progress to measure the rate of reduction in pollution, with distance from the motorway, in particular.

The main conclusion of the further investigation is that air quality in Birmingham, in 2005, appears likely to exceed the air quality objective for the annual average concentrations of NO₂ at certain locations throughout the city. As a result Birmingham City Council, were required by Section 83 of the Environment Act 1995, to declare one or more Air Quality Management Areas, for those parts of the city, in which the air quality objective is likely to be exceeded.

Using the model to determine the NOx Reduction Required

Birmingham City Council declared the whole city an Air Quality Management Area as it was predicted that the 2005 objective for NO₂ would not be achieved. The next stage is to establish how NO₂ can be reduced. To do this we need to know what sources are causing the problem. Using the Airviro model the percentage contribution of the NOx from the different sources can be determined.

For example, modelling has shown that on the Stratford Road (one of the intensive study areas) the approximate contribution of various sources to NO_x concentration are as follows:

Traffic 65%

Industry 5%

Background 30%

However, across the city the scenario becomes;

Traffic 50%

Industry 20%

Background 30%

Within the Airviro model it is possible to refine the source apportionment exercise further.

Using the Airviro Model to conduct Source Apportionment

Source apportionment is the process whereby the contributions from individual sources of pollution are determined.

Road traffic has been identified as the most important source of NO₂ within Birmingham. The Airviro model has been used to determine the contribution of a number of sources within the traffic fleet to the total amounts of nitrogen oxides across the borough. Figure 1 and 2 for 2001 and 2005, respectively, illustrate the contribution of a number of sources to the total amounts of oxides across the borough.

\s Figure 1: Source Apportionment for oxides of Nitrogen across Birmingham in 2001

\s

Figure 2: Source Apportionment for oxides of Nitrogen across Birmingham in 2005

Although, the transportation sector is clearly the major contributor to the emissions of nitrogen oxides across the city (refer to Figures 1 and 2), it is noteworthy that there is a slight decrease from a traffic contribution of 79% in 2001 to 74% in 2005. It is anticipated that with the improved technology in vehicles been manufactured, this downward trend will continue to 2010.

Although Figures 1 and 2 describe the relative emissions for Birmingham as a whole, the relative amounts of pollution vary across the city. Thus, in an industrial area, the percentage of pollution from industry may be higher than the city average, whilst that from road traffic may be less significant. Conversely, in areas close to major roads, pollution from road traffic may dominate the emissions. For example, concentrations of NO₂ measured at the Selly Oak (Bristol Road) area of the city exceeded the objective; the primary source in this area is emissions from traffic. Figure 3 illustrates that petrol cars, HGV and buses made the greatest percentage contribution to the total nitrogen oxides emissions in 2001 and the same pattern is expected in 2005. Although HGVs and buses are a small part of the fleet composition, they give rise to greater amounts of pollution than cars (petrol and diesel) and therefore it might be pertinent to carefully consider them during action planning.

\s Figure 3: Source Apportionment for oxides of Nitrogen in Selly Oak 2001.

Source apportionment has indicated that central Birmingham is slightly different from other areas of the city, in that the contribution from commercial and domestic sources appears to be significant at an excess of 30% for both 2001 and 2005 (refer to Figure 4)

\s Figure 4: Source apportionment for oxides of nitrogen in Central Birmingham.

All the pie charts illustrate that petrol cars were a significant contributor to the nitrogen oxide emissions and it is anticipated that this trend will continue into 2005. Hence a more detailed analysis of this category of vehicles is warranted.

The emissions inventory is compiled such that investigations using the Airviro Model, of three types of petrol cars, namely Pre-Euro, Euro 1 and a composite of Euro 2, 3 and 4 can be carried out. For 2001, the Pre-Euro cars made up 23% of the total vehicle fleet composition (31% of the petrol car fleet) but contributed approximately 80% of the total emissions from petrol cars (refer to Table 1). Then, Table 2 illustrate that in 2005, the Pre-Euro cars are expected to make up 11% of the total vehicle fleet composition (15% of the petrol car fleet) but contributed approximately 67% of the total emissions from petrol cars. This suggests that it may be necessary to target this particular type of petrol car in a bid to reduce the emissions of nitrogen oxide.

Table 2: Contributions from the three classes of euro cars to the total emissions from petrol cars in 2005.

The above provides an overview how NO₂ can be modelled with a dispersion model and how the information produced from the model can be interpreted and used when making decisions in relation to air pollution.

For more detailed information on how Birmingham City Council continue to use modelling in the Review and Assessment of air quality throughout Birmingham you should visit the Birmingham City Council web site:

www.birmingham.gov.uk

Last Updated

13th January 2005