PM10 concs in UK sections 2.1 to 2.4 2 Summary of PM10 projections presented in the APEG report

2.1 Receptor modelling
A receptor modelling technique has been developed which enables the measured daily mean PM10 concentration at a monitoring site to be divided into three components (APEG, 1998, Stedman, 1997):
A multiple regression analysis is carried out to determine the coefficients A and B for primary combustion and secondary particle concentrations. Either black smoke measurements or oxides of nitrogen (NOx) measurements are used as an indicator for primary combustion particles and rural sulphate measurements are used as an indicator for secondary particles:

[measured PM10] = Abs [measured black smoke] + B [measured sulphate] + C
 or
[measured PM10] = ANOx [measured NOx] + B [measured sulphate] + C
 The daily mean concentration of 'other' particles is determined by difference.

2.2 NAQS projections for 8 sites based on 1995 and 1996 data

The PM10 receptor model that has been developed enables us to make relatively sophisticated estimates of both annual mean and high percentile PM10 concentrations for future years. The key advantage of this method is that the PM10 concentrations have been separated into their component parts and appropriate reductions can be applied to these components, based on an understanding of the likely impact of current policies on future levels. The emission reduction factors that have been used to calculate the predictions based on 1996 data are summarised in Table 2.1. The factors for 1995 are slightly different.

Table 2.1. Emission reduction factors used to calculate projections for 2005 and 2010, relative to 1996 values.

Year Urban traffic exhaust Other Urban primary Urban primary combustion, GB Urban primary combustion, NI Secondary particles 'other' particles
2005 0.51 1.00 0.63 0.75 0.81 1.00
2010 0.37 1.00 0.53 0.69 0.70 1.00

Projections for urban road traffic exhaust emissions of PM10 within the UK National Atmospheric Emissions Inventory (NAEI) show that these emissions are expected to reduce to approximately half their 1996 values by 2005 (Salway et al, 1997, Goodwin et al, 1997, Murrells, 1998). We have assumed that 75% of 1996 primary emissions in UK cities were from traffic exhaust and 25% from other local sources and the latter emissions were assumed to remain at 1996 levels in 2005 and 2010. This 75% is the average split between traffic and other sources across major cities in Great Britain within the NAEI. Across all of the urban areas in Great Britain the split is 67% whereas in London 91% of primary emissions is from traffic exhausts sources. Coal use currently contributes substantially more to primary particle concentrations in urban areas of Northern Ireland (NI) than is generally the case in urban areas in GB. We have assumed that non-traffic exhaust emissions contributed 50% of the total of urban emissions in 1996 in NI and that these emissions will remain at 1996 levels.

The factors for secondary particle concentrations in future years in the UK have been based on the results of the EMEP modelling of secondary particles over Europe (Tarrason and Tsyro, 1998), which showed that levels in 2010 are likely to be about 70% of 1996 levels on the basis of current policies. It has been assumed that this reduction up to 2010 will be linear, leading to an estimate of concentration in 2005 being equal to 81% the current values.

It has been assumed that the 'other' particle concentration will remain at 1996 levels in all future years.

This analysis directly provides an estimate of the 99th percentile of daily mean PM10 concentrations, so one additional step is required to provide an estimate of the value required for comparison with the NAQS objective. The 99th percentile of the daily maximum of running 24-hour mean PM10 concentration (NAQS 99th percentile) is generally slightly higher than the 99th percentile of daily values, by a factor of about 1.16 (Figure 2.1). This is largely due to primary particle episodes, which are generally of shorter duration than secondary particle episodes causing daily maximum 24-hour running mean concentrations to be higher than fixed daily means. There is uncertainty associated with the use of this factor. It tends to overestimate the NAQS 99th percentile in years dominated by secondary particle episodes such as 1996, while underestimating values in years such as 1995, where high percentiles were less dominated by secondary particle episodes. It should also be noted that this factor of 1.16 is unlikely to remain constant in future years and may reduce as the traffic exhaust contribution to urban PM10 becomes smaller. The number of days per year with running 24-hour PM10 concentrations greater than 50 mgm-3 (NAQS days) is also greater than the number of days with fixed daily means above this threshold. The number of NAQS days is currently approximately 1.8 times the number of fixed days above 50 mgm-3. Since this is a large correction factor and it is likely to change in future years, we have not applied it to the number of days with concentrations above 50 mgm-3 presented in this report and have presented a value for fixed daily means only.

The estimated PM10 concentrations for 2005 and 2010 presented in the APEG report for the business as usual scenario are reproduced in Table 2.2. This table shows predictions for comparison with the NAQS objective.

The estimated NAQS 99th percentiles for 2005 based on both 1996 and 1995 mreasurements are higher than 50 mg m-3 (much higher than 50 mg m-3 for 1996 measurements) indicating that current policies are unlikely to deliver compliance with the objective for 2005 at city centre locations. The projected concentrations based on 1996 monitoring data are higher than those based on 1995 data because of the elevated secondary particle concentrations during the early part of 1996 (Stedman, 1997).

2.3 EU Daughter directive projections for 8 sites based in 1995 and 1996 data

The estimates of PM10 concentrations presented in Table 2.3 and Table 2.4 have been multiplied by 1.3 for comparison with the EU limit values. This is to take into account the difference between TEOM (Tapered Element Oscillating Microbalance) measurements of PM10 and the EU reference gravimetric method. While there is considerable uncertainty to the exact relationship between TEOM and gravimetric measurements, we have chosen to use a factor of 1.3 as a conservative approach.
In contrast to the projections calculated for comparison with the NAQS objective, the projections of percentiles and numbers of days above 50 mgm-3 can be calculated directly from the results of the receptor models because the limit value is for fixed daily means. The EU Stage 1 and Stage 2 limit values will apply for 1 January 2005 and 1 January 2010 respectively. Projections for comparison with these limit values should therefore ideally be calculated for 2004 and 2009, rather than 2005 and 2010. Projections for 2005 and 2010 for comparison with the EU limit values were, however, presented in the APEG report for easy of computation and consistency with the NAQS projections. We have therefore adopted this approach for all of the 1996 and 1995 base year projections presented in this report. Projections based on 1997 data have been calculated for 2005 for comparison with the NAQS and for 2004 and 2009 for comparison with the EU limit values.
Predictions of PM10 concentrations for 2005 and 2010 for comparison with the 'Daughter Directive' have been made based on 1995 and 1996 analysis. A "business as usual" primary emissions reduction scenario, blanket reduction of secondary particle concentrations using EMEP coefficients have been used in the calculations. Current national policies are likely to deliver concentrations lower than the Stage 1 90th percentile limit value except possibly in central London. For Stage 2 the 98th percentile and the annual limit value are likely to be exceeded at all sites (Table 2.4).



Back to section 1To contents pageTo chapter 1To sections 2.4 and 2.5