4.1 Introduction    4.2 Source apportionment of PM10    4.3 Map of secondary PM10 for 1996    4.4 A map of total PM10 for 2004
4.5 An alternative map of total PM10 for 2004

4 Mapping studies

4.1 INTRODUCTION
The mapping of background PM₁₀concentrations across the UK is a complex task because of the variation in composition and source apportionment of particles between different locations. The PM₁₀mapping methods used within the DETR research programme have developed considerably over recent years as the understanding of the source apportionment of PM₁₀has improved (Stedman et al, 1997, Stedman, 1998). A map of estimated annual mean PM₁₀for 1996 has been published by Stedman (1998) along with maps of the other National Air Quality Strategy (NAQS) pollutants. These mapping techniques were later extended by incorporating the findings of the recently published report of the Airborne Particles Expert Group (APEG, 1999) to calculate daily maps of PM₁₀within the review of the NAQS (Stedman et al, 1998, DETR et al, 1999).

While a number of uncertainties remain within the mapping methods that have been developed, it is clear that the most important requirement for the calculation of maps of estimated pollutant concentrations is that they should be fit for purpose. The maps of estimated PM₁₀ concentration that have been calculated here for use within the review and assessment process have been developed within the framework of the drafting of the technical information. A priority in the choice of mapping methods is that the maps are fully integrated into the review and assessment process and have been calculated in a manner consistent with the recommended procedures for review and assessment. This enables calculations to be carried out consistently throughout the procedure, so that, for example, the treatment of coarse particle concentrations will be identical whether background concentrations are derived from the maps or from ambient monitoring data.

Two maps are required for the review and assessment procedure for the proposed PM₁₀objectives and these maps will be made available within the National Air Quality Information Archive on the World Wide Web:

www.aeat.co.uk/netcen/airqual/

These maps are

• background annual mean secondary PM₁₀in 1996
  
• background annual mean total PM₁₀in 2004.
  

Both maps show background PM₁₀concentrations as measured by a gravimetric or equivalent method.


4.2 SOURCE APPORTIONMENT OF PM₁₀

The source apportionment of PM₁₀ has recently been reviewed by APEG (1999) and the sources that are thought to make significant contribution to ambient background concentrations in the UK are listed in Table 5.

Table 5: Source apportionment of background PM₁₀

Size	Main Categories	Main Source Types
Fine (< 2.5 mm)	Primary	local vehicle exhaust
		local combustion sources
		regional vehicle exhaust
		regional combustion sources
	Secondary	regional sulphates and nitrates
Coarse (2.5 - 10 mm)	Natural	regional resuspended soil and dust, sea salt
	Man made	resuspended dusts, mechanically derived particles

A map of estimated PM₁₀ concentration therefore need to take each of these sources into account.

4.3 A MAP OF SECONDARY PM₁₀ FOR 1996
Figure 4 is a map of estimated secondary PM₁₀concentrations for 1996. This map has been calculated from the map presented as Figure 4.8 in the APEG report by multiplying by 1.3. The APEG map was derived from rural sulphate measurements using the methods recommended by QUARG (1996). The factor of 1.3 has been applied because the receptor modelling relationships used by APEG were based on TEOM monitoring data:

secondary PM₁₀(gravimetric units) = 1.3 x secondary PM₁₀(TEOM units)

1996 was chosen as the base year for the review and assessment procedure due to the unusual 'worst case' secondary particle episodes experienced in the early part of the year (APEG 1999).


4.4 A MAP OF TOTAL PM₁₀ FOR 2004

4.4.1 Introduction
The 2004 map has been calculated in two stages:

• a map of total PM₁₀for 1996 has been estimated
  
• appropriate emission reduction factors were applied to the individual components of the map
  

Click here to see Figures 5a to 5d

4.4.2 1996
This map is shown in Figure 5a and has been derived using the following equation:

estimated PM10(gravimetric units)	= 1.3 x secondary PM10(TEOM units)
	+ 1.3 x k1 x local primary emissions (kTonnes per 25 km2 per year)
	+ 1.3 x coarse(TEOM units)

The treatment of secondary PM₁₀ is the same as used to derive Figure 4. The primary component has been calculated from local primary emissions from area sources (but excluding all industrial sources) estimated within the National Atmospheric Emissions Inventory. The coefficient, k₁, has been derived by regression analysis of TEOM monitoring data from the DETR national air quality monitoring networks. The value of this coefficient has been allowed to vary as follows, in order to achieve best fit to the monitoring data and take into account any systematic errors in the emissions estimates:

Northern Ireland:	0.024
London:	0.032
Elsewhere:	0.038

These values can be compared with a similar coefficient for the local contribution to NO_x concentrations (for which the source apportionment and emissions are well understood) of 0.033 in consistent units (Stedman, 1998).

The value of the coarse particle concentration has been set to 8 mgm^-3 (TEOM units) or 10.5 mgm^-3 (gravimetric units). This is the typical value found for the coarse particle concentration derived as the residual in the receptor modelling of TEOM data undertaken by APEG (1999) and Stedman et al (1998).


4.4.3 2004              Click here for Figures 5a to 5d
The following emissions reduction factors (derived from APEG, 1999) were applied in order to project to 1996 map forward to the 2004 map shown in Figure 5b:

primary 2004	= primary 1996 x 0.651 (GB)
primary 2004	= primary 1996 x 0.767 (Northern Ireland)
secondary 2004	= secondary 1996 x 0.829
coarse 2004	= coarse 1996

The primary emission reduction factors were derived from forecasts of road traffic exhaust emissions; which are assumed to contribute 75% of total primary combustion emissions in GB and 50% in Northern Ireland; non road traffic exhaust primary emissions are assumed to remain at 1996 levels. The secondary emission reduction factors were derived from secondary particle modelling carried out by EMEP.

This map (Figure 5b) is the one recommended for use within the review and assessment procedure.


4.5 AN ALTERNATIVE MAP OF TOTAL PM₁₀ FOR 2004           Click here for Figures 5a to 5d
The map presented in Figure 5b was chosen for inclusion in the review and assessment procedure because the simplifying assumptions incorporated in its calculation enable the review and assessment calculations to be reasonably straight forward.

The treatment of both primary and coarse particle concentrations have, however, been considerably simplified in comparison to the source apportionment findings of APEG. Figures 5c and 5d shown alternative maps of total PM₁₀concentrations for both 1996 and 2004 based on the following more complete breakdown, in which these components have been split into local and regional contributions :

estimated PM10(gravimetric units)	= 1.3 x secondary PM10(TEOM units)
	+ 1.3 x k2 x local primary emissions (kTonnes per 25 km2 per year)
	+ 1.3 x regional primary (TEOM units)
	+ 1.3 x regional coarse(TEOM units)
	+ 1.3 x local coarse (TEOM units)

where

regional primary has been estimated from rural NO₂ measurements by multiplying by a factor of 0.15. This gives a value that varies from about 2 mgm-3 (TEOM units) in the south east to virtually zero in north west Scotland. This is reasonably consistent with both the modelled estimates of long range transported PM₁₀concentrations of 1.9 mgm-3 in London and 1.0 mgm^-3 and long range transported black smoke concentration of about 1.5 mgm^-3 in East Anglia cited by APEG (1999).

regional coarse has been assumed to be 6 mgm^-3 across the whole of the country.

local coarse has been estimated by multiplying estimates of construction derived PM₁₀emission from the NAEI across an area of 25 km² by 0.20. This factor has been chosen to give an additional urban contribution to coarse particle concentrations of about 2 mgm^-3 (TEOM units) in central London and zero in rural areas. It is recognised that the local contribution to coarse particle concentrations is due to a range of urban activities including resuspended dusts due to traffic and construction activity, and construction emissions have been used as an indicator of these activities. This indicator was chosen to be consistent with the limited information from measurements of the concentration of coarse particles at sites with co-located PM₁₀and PM_2.5 monitors (APEG, 1999), which indicated urban levels of about 8 mgm^-3 in London and Birmingham but only about 5 mgm^-3 at the rural Harwell site.

The values of the coefficient k2 are slightly lower than the k1 values used in Figures 5a and 5b because the regional primary component is treated explicitly, rather than rolled into the local primary:

Northern Ireland:	0.020
London:	0.028
Elsewhere:	0.032

The regional primary concentration has been treated the same as the local primary concentration and the local coarse concentration has been treated the same as the regional coarse concentration in projecting the 1996 map (Figure 5c) forward to 2004 (Figure 5d).

This alternative map has slightly higher estimated total PM₁₀ concentrations in 2004 in some urban areas (up to about 2 mgm^-3 in central London) and lower concentrations in rural areas (by up to about 3 mgm^-3). While this map provides a more realistic picture of the likely variation in background annual mean PM₁₀in 2004, it is not compatible with the rest of the review and assessment procedure unless the method is made considerably more complicated. The procedure in the technical information report (Moorcroft et al, 1999) has the advantage that the local primary contribution to current PM₁₀can be calculated by subtracting the mapped secondary contribution and a constant concentration of coarse particles from a measured or mapped value. This simplicity would be lost with the more complex map. This underlines that it is possible to calculate several different maps of the same statistic and the importance of ensuring that each map is 'fit for purpose'.


Section 3 (Industrial Stacks)          References

---

Report and site prepared by the National Environmental Technology Centre, part of AEA Technology, on behalf of the UK Department of the Environment, Transport and the Regions