The regression analysis was carried out for a calendar year of monitoring data for each site to determine the coefficients A and B:

[measured PM₁₀ (mg m^-3)] = A.[measured NO_x (mg m^-3)] + B.[measured sulphate (mg m^-3)] + C (mg m^-3)

These coefficients were then used to divide the measured concentration into the three components. This analysis has been completed for the years 1996 to 1999 inclusive at a range of UK national network monitoring sites.

4.2 Projecting concentrations

Each component of the daily average PM₁₀ concentration was then projected from the measurements in 1996, 1997, 1998 and 1999 to provide estimates of concentrations in 2010.

Maps of primary PM₁₀ emissions for 1997 from the NAEI (Goodwin et al, 1999) were used to determine the sectors (road traffic, domestic and services, industry, others) contributing to local primary combustion PM₁₀ at each monitoring site location. Road traffic typically contributed about 75% of emissions in large urban areas in 1997. The emissions for each sector were projected forwards by reference to emissions for non-traffic sector for different years. Published emission estimates for each sector are available from the NAEI for the years up to and including 1998 (Goodwin et al, 1999). Emissions from non-traffic sources for years between 1998 and 2010 have been derived from DTI (2000) and estimated within the NAEI. These estimates therefore incorporate an assumed growth in economic activity of about 2.5% per year and the continuation of current trends towards greater use of natural gas and cleaner technologies DTI(2000). We have estimated that non-traffic emissions in 2010 will be approximately 70% of those in 1998. Estimates of road traffic emissions were taken from Table 3.

Secondary particle concentrations in previous years were derived from the network means of rural measurements of sulphate (at 8 sites) and nitrate (total inorganic nitrate at 2 sites). Concentrations of sulphate and nitrate in future years were derived from European scale modelling work for 1997 and 2010 carried out at Imperial College. The calculated values for 2010 incorporate the emissions reductions set out within the so-called ‘Gothenburg Protocol’ to Abate Acidification, Eutrophication and Ground-level Ozone. We have assumed a linear decline in concentrations from 1997 to 2010. Table 4 shows the measured and estimated sulphate and nitrate concentrations that we have used, normalised to 1997.

The split between sulphate and nitrate at each site for each year from 1996 to 1999 was derived from the receptor model coefficient B. This coefficient relates secondary PM₁₀ concentrations to measured sulphate concentrations. If all of the measured secondary PM₁₀ were ammonium sulphate, then this coefficient would be approximately 1.3. This coefficient was generally found to be in the range from 2 to 3, the remaining secondary PM₁₀ was assumed to be nitrate.

We assumed that there will be no change in coarse particle concentrations.

Figure 4 shows illustrative site specific projections of annual mean PM₁₀ concentrations for the London Bloomsbury site. It is clear that there is good agreement between the projections for the years 1993 to 1999 and the measured values for these years. The projections track both the year to year variability in concentrations due to changes in the meteorology that influences secondary particle concentrations and the changes in concentrations due to reductions in emissions. The projections based on 1996 monitoring data are the highest; projections based on 1998 and 1999 data are lower.

The site specific analysis for PM₁₀ is of daily means, which allows us to estimate future daily concentrations directly, and thus the number of exceedences of 50 mgm^-3. The analysis is based on TEOM data, so we have applied a scaling factor of 1.3 to all data before comparing with the limit value (which is based on gravimetric measurement), as suggested by APEG (1999). Figure 5 shows illustrative site specific projections of the number of days with PM₁₀ concentrations greater than or equal to 50 mgm^-3 (gravimetric) at London Bloomsbury. The measured number of exceedences is relatively constant from 1992 to 1996, which 1996 having the largest number of exceedences. The number of days with concentrations greater than or equal to 50 mgm^-3 then declined rapidly to about 20 in 1998 and 1999. The projected concentrations also show this decline, although projections based on 1996 or 1997 are consistently higher than those based on 1998 or 1999. High concentrations during 1997 were dominated by primary particle episodes, as demonstrated by the steep decline from 1993 to 1998 for projections based on 1997.

Figure 6 shows site specific projections of the number of days with PM₁₀ concentrations greater than or equal to 50 mgm^-3 (gravimetric) at Bristol Centre. Once again it is clear that the combination of the APEG receptor model and the emissions estimates and projections can be used to explain the trend in the measured concentrations.

The projections of traffic emissions used to calculate the illustrative results shown in shown in Figures 4 to 7 are NAEI estimates (Murrells, 2000), which are based 1997 National Road Traffic Forecasts for Great Britain (DETR 1997). These projections pre-date the area type specific emissions projections for England listed in Table 3 and do not reflect any of the measures specified in the 10 Year Plan. The advantage of these NAEI projections is that they are available for all years from 1990 to 2010 and facilitate the comparison with measurement data (the 10 Year Plan emissions estimates are currently only available for 1996, 2000 and 2010). This enables the profile of projected concentrations changes to be compared with past trends an assessment of concentrations in years such as 2004 for which AQS objectives have been set The NAEI projections for 2010 are similar to those for the baseline scenario within the 10 Year Plan. However the assumptions underlying the two are not the same. For example the NAEI estimates assumed that the previous policy of the fuel duty escalator would continue until 2002. The 10 year Plan baseline only includes the impacts of the fuel duty escalator to 1999, after which the policy was changed. In addition, the NAEI estimates also do not take into account the effect on road traffic volumes of the policies in the Government’s Integrated Transport White Paper (DETR, 1998). More details on the Plan baseline can be found in the background document (DETR 2000b).

4.3 Projections for roadside monitoring sites

Site specific projections for roadside monitoring sites have also been calculated. Daily averages of measured PM₁₀ at a nearby background monitoring site have been subtracted from the concentrations measured at roadside monitoring sites in order to determine the roadside increment of daily PM₁₀. It is not possible to determine the split of the roadside increment between traffic exhaust emissions and re-suspended dusts from current network measurements. Analyses of PM₁₀ and PM_2.5 monitoring data presented in the APEG report suggested that re-suspended component could be 50% of the total. It is likely that this is an overestimate because exhaust emission may include some particles of diameter greater than 2.5 mm. We have assumed that re-suspended dust contributed 25% of the roadside increment of PM₁₀ concentrations in 1997. We have projected this roadside increment forward on the basis that this component will not change in future years, unlike the remaining 75% of the roadside increment, which will decline in line with reductions in exhaust emissions.

Figure 7 shows projections for the Bury Roadside monitoring site. The projections closely mirror the steep decline in the measured numbers of exceedences.

Chapter 3           Chapter 5