Urban scale integrated assessment for London: Which emission reduction strategies are more effective in attaining prescribed PM₁₀ air quality standards by 2005?

Abstract

Tightening of air quality standards for populated urban areas has led to increasing attention to assessment of air quality management areas (AQMAs) where exceedance occurs, and development of control strategies to eliminate such exceedance. Software tools that bring together data on pollutant sources, their respective contributions to atmospheric concentrations and human exposure, together with information on potential technological and other measures that may be used to reduce concentrations and their economic costs, can be used to identify cost-effective strategies for improving air quality, and hence aid policy development. The Urban Scale Integrated Assessment Model (USIAM) has been developed in this context, illustrated in this paper by application to traffic emissions which are a major contributor to exceedance of air quality objectives for fine particulate matter, PM₁₀, in London. In such a multidisciplinary approach the aim has been to provide a tool for rapid assessment of a wide range of scenarios to identify those that are most cost effective, maintaining a balance between the level of sophistication in the model and the uncertainties and assumptions involved.

Introduction

Development of strategies to combat urban air pollution is a complex process involving a wide range of people with different expertise and interests. The setting of air quality standards for protection of human health requires medical expertise. The extent to which such standards are exceeded requires measurements and monitoring data. Careful identification of the sources and construction of an emissions inventory is a precursor to modelling of atmospheric dispersion and validation against measurements. Information is then needed on the technological and planning options available and how much these can help to reduce emissions and at what cost. By bringing such information together integrated assessment modelling provides a quantitative tool for investigating the effectiveness of different abatement strategies, and ranking them according to their costs and effectiveness in improving air quality. Such an approach is transparent to both the responsible authorities and to the public and people affected. However, it is also important to make clear the assumptions and uncertainties and their impact on the robustness of the results.

This paper illustrates such an approach in the Urban Scale Integrated Assessment Model (USIAM) (Mediavilla-Sahagún, 2002, Mediavilla-Sahagún et al., 2002, Mediavilla-Sahagún and ApSimon, 2003) applied here to the control of the contribution from traffic to concentrations of fine particulate matter, PM₁₀,¹ in London, due to their proven adverse health effects (Pope et al., 1995, Künzli et al., 2000), by exploring potential emission reduction strategies that are successful in eliminating exceedance of Air Quality Standards for 2005 and in reducing population exposure, selecting and prioritising them in a cost-effective context. The problem of modelling concentrations across such a major city due to different types of vehicles on the extensive network of road links throughout the meteorological conditions of a single year is already heavy on computer resources. The aim in integrated assessment modelling is to provide a tool that can assess potential abatement strategies in a matter of minutes, and potentially be used interactively. In USIAM this is partly achieved by identifying different sources and locations, and a distribution of receptor locations, linked by pre-calculated dispersion matrices (using ADMS-Urban,² McHugh et al., 1997) to define source–receptor relationships for rapid calculation of the contribution of each source to the concentration at each receptor.

The current derived EU and UK air quality objectives aiming towards Air Quality Standards for PM₁₀ are defined as a limit of 35 exceedances per year of 50 μg m⁻³ over 24-h periods, and 40 μg m⁻³ for the annual mean. To avoid separate consideration of episodes and annual limits, an empirical relationship has been used between the 90th percentile of daily means (which corresponds to approximately 35 exceedances per year) and the annual average concentrations. This has been derived from an analysis of monitoring data at UK automatic sites between 1992 and 1997 (DETR, 2000a, Fig. 8.1 and Table 8.1). From this relationship it is deduced that the proposed 24-h objective is therefore unlikely to be exceeded if the annual mean concentration is below 28 μg m⁻³, gravimetric (DETR, 2000a). This is used as the derived target in USIAM for annual average concentrations of PM₁₀.

In the London metropolitan area, the main sources of primary PM₁₀ are by far the road vehicles (77%), and within these the Heavy Goods Vehicles (37.17%) and Buses (19.23%) are the largest contributors (LRC, 1997), followed by the petrol cars (12.26%). The emphasis has thus been on reducing the contribution from these sources. However, these emissions have to be superimposed on other contributions from stationary sources, and the observed coarser fraction of particles in the 2.5–10 μm size range which is not accounted for in current emission inventories. The former can be explicitly modelled, but the latter has to be added very crudely based on observations pending further research. In addition there is the secondary component, originating both from other national and trans-boundary sources that emit the chemical precursors required for its formation (e.g. nitrogen oxides, sulphur dioxide, ammonia and volatile organic compounds), which remains relatively constant throughout the south-eastern part of the UK (APEG, 1999), and cannot be controlled by action within London.

For that reason, this study focuses entirely on predicting future annual average PM₁₀ concentrations by modelling explicitly the dispersion of the primary component only. The secondary and coarse components are treated as part of the background superimposed on the modelled primary.

By ranking different options for abatement of emissions in terms of implementation costs and either the improvement in the area achieving Air Quality Objectives or the average population exposure, the USIAM model can also select and prioritise different potential strategies. It should be noted that these two indications of benefit may give a different ranking.