1.1 Polycyclic Aromatic Hydrocarbons

Polycyclic aromatic hydrocarbons (PAHs) are a group of persistent organic pollutant compounds they contain two or more benzene rings, they are generally produced through incomplete combustion or pyrolysis and sources include industrial, commercial, vehicular and residential sources. A more detailed discussion of the sources of PAH in the UK is presented in Section 1.3 where trends in the emissions of B[a]P presented.

The International Agency for Research on Cancer (IARC) has determined that B[a]P is carcinogenic to humans (group 1) (WHO, 2010) and is currently considered by IARC as the most carcinogenic PAH. The carcinogenic classification of PAH and other agents can be found on the International Agency for research on Cancer (IARC) website.

Table 1 below shows the details of PAH that are required to be measured Fourth Daughter Directive (2004/107/EC) (EC, 2005) in the UK plus benzo[ghi]perylene which was more recently included in the 2014 technical standard CEN/TS 16645:2014 (BSI, 2014). The table also included the IARC classifications and structures of the PAH.

Table 1: PAH structures and IARC Carcinogenic Classification of PAH that should be monitored according to the 4th Daughter Directive (2004/107/EC). IARC Group 1: Carcinogenic to humans, Group 2A: Probably carcinogenic to humans, Group 2B: Possibly carcinogenic to humans, Group 3: Not classifiable as to its carcinogenicity to humans

Compound	Structure	IARC
Benz[a]anthracene		2B
Benzo[b]fluoranthene		2B
Benzo[j]fluoranthene		2B
Benzo[k]fluoranthene		2B
Benzo[a]pyrene		1
Dibenz[a,h]anthracene		2A
Indeno[1,2,3-cd]pyrene		2B
Benzo[g,h,i]perylene		3

1.2 Air Quality Policy

In the UK there is a national air quality objective for B[a]P in ambient air. This is based on an annual mean concentration of 0.25 ng/m3. Details can be found in the UK in the Air Quality Strategy in 2007.

The EC Air Quality Framework Directive (Directive 96/62/EC) (EC, 1996) set a strategic framework to tackle air quality in a consistent way across Europe by setting limit and target values for air pollutants via a series of Daughter Directives. The Fourth Daughter Directive sets a target value for B[a]P of 1 ng/m³ (total content in the PM₁₀ fraction averaged over a calendar year). Mandatory measurement requirements relating to the measurement of B[a]P can be found in the Fourth Daughter Directive (Directive 2004/107/EC).

B[a]P’s suitability as a marker for the PAH mixture in ambient air by the EC Position Paper (EC, 2001) on PAH led to it being selected as the measure for monitoring in the Fourth Daughter Directive (Directive 2004/107/EC) and the more stringent UK National Air Quality Objective for PAH. Measurements of B[a]P in ambient air are covered by the European standard EN 15549 (BSI, 2008), which has been adopted as the European reference method.

Measurements of PAH in Deposition are covered by European standard EN 15980 (BSI, 2011) which details the measurement method sampling, sample preparation and analysis for benz[a]anthracene, benzo[b]fluoranthene, benzo[j]fluoranthene, benzo[k]fluoranthene, benzo[a]pyrene, dibenz[a,h]anthracene and indeno[1,2,3-cd]pyrene. There is no limit or target value related deposition of PAH in the UK or Europe.

The two sites of located at Chilbolton Observatory, Hampshire and Auchencorth Moss, Midlothian are used to support the European Monitoring and Evaluation Programme (EMEP) to Level 2. EMEP is a scientifically based and policy driven programme under the Convention on Long-range Transboundary Air Pollution (CLRTAP) which was signed by the UK in 1979. The convention aids international co-operation to solve transboundary air pollution problems, provides access to emission, measurement and modelling data and provides information on the effects of air pollution on ecosystems, health, crops and materials.

1.3 Sources of PAHs in the UK

The National Atmospheric Emissions Inventory (NAEI) has estimated the emission of PAH for the UK for many years. The inventory now estimates the emissions of 16 PAH (the US-EPA 16) which includes benzo[a]pyrene. As with all emissions inventories there is some uncertainty in the estimates as the emissions are not based solely on measurements and require some estimation of emission factors and activities being required.

There has been a significant decrease in the overall emissions of benzo[a]pyrene in the UK since the 1990’s. Figure 1 below displays UK estimated emissions of benzo[a]pyrene by source sectors with data selected from 2000 to 2016. From a review of this it is apparent that the emission of benzo[a]pyrene are dominated by the residential and commercial combustion. More historic emissions estimates are also available by double clicking on the plot.

In recent years residential and commercial combustion dominate emissions with emissions of B[a]P making up 89% of the estimated emissions of B[a]P in the UK in 2016. The residential and commercial combustion sector shows an increase in estimated emissions of B[a]p from an estimated 4.3 tonnes in 2000 to 6.7 tonnes in 2016 which represents a 54% increase.

Improvements in the calculation of emissions from the domestic wood combustion sector to take into account the age of appliances has been integrated into the NAEI emission estimates shown in this report. Therefore, whilst it is possible to compare emissions between years in the data set they should not be compared with data previously reported.

More information relating to UK Emissions of PAH and other pollutant can be accessed via the NAEI website.

2.1 Network Objectives

The objective of the PAH Network is to determine the ambient concentrations of PAHs in ambient air in the UK through monitoring and chemical analysis, and deliver or aid the delivery the following:

• A UK assessment of current concentrations of PAHs for assessment against the Fourth Daughter Directive and the UK Air Quality Strategy objectives, and provide measurement input for any future reviews.
• A Review of the measurements and trends of airborne concentrations of PAHs in representative UK industrial, urban and other areas enabling comparison the rural locations where measurements take place.
• Provide data and metadata to UK-AIR to enable demonstration of the UK’s compliance with the Fourth Daughter Directive, the OSPAR convention (OSPAR, 2017) and the UNECE Convention on Long Range Transboundary Air Pollutants (UNECE, 1979).
  
• To enable a better understanding of sources or potential sources of PAH to potentially enable improvements to the NAEI or identification areas of uncertainty.

2.2 Network Overview

The 32 monitoring stations operating in the UK PAH Network during 2017 are shown in Figure 2. Two monitoring sites where both Digitel particulate samplers and deposition samples are taken are marked with drop marker (Auchencorth Moss and Chilbolton Observatory). Other sites where only Digitel particulate samples are taken are marked with red circles.

In 2017 there were 32 ambient air sampling sites operational in the network in the UK in addition to two deposition samplers. The majority of the sampling locations are urban background however the network also includes urban industrial, rural background and a single site that is urban traffic at London Marylebone Road. Of the rural located samplers there are two sites that are located at Chilbolton Observatory, Hampshire and Auchencorth Moss, Midlothian which are used to support the European Monitoring and Evaluation Programme (EMEP).

2.3 Samplers in the PAH Network

The Network requires the sampling and analysis of two types of samples these are particulate and deposition samples.

‘PAH Digitel (solid phase)’ particulate samplers. These samples are in the PM₁₀ fraction of ambient air on a filter and are taken daily at all 30 Network stations using Digitel DHA-80 samplers with automatic filter changers. Each sample is taken for 24 h, and a sample is taken every 24 h, with the sample changeover generally occurring at midnight. The samples are bulked into groups representing calendar months for analysis. The Digitel DHA-80 samplers (see Figure 3) used throughout the Network are considered to be equivalent to the requirements of the European Standard for sampling PM₁₀ matter (EN 12341) (BSI, 2014). The samplers are therefore valid for use with the European Standard method for the measurement of B[a]P in ambient air (EN 15549). The solid phase filter samples have a measurement period of 24 hours at a flowrate of approximately 30 m³/h.

Figure 3: Digitel DHA-80 sampler deployed to measure solid phase PAH in the UK Network.

‘PAH deposition’ samplers. These Deposition samples are taken fortnightly at two rural Network stations of Auchencorth Moss and Chilbolton Observatory (previously at Harwell). Each sample is taken for 14 days using a deposition sampler (Figure 4) that meets the requirement of the European Standard for the measurement of the deposition of PAHs (EN 15980). The deposition samplers itself consist of a glass funnel and a four litre brown glass collection bottle, which are located inside a protective tube in order to minimise photochemical reactions and the degradation of PAHs. The spikes seen on the image have been fitted to the top of the protective tubes to prevent damage and contamination by bird strikes.

Figure 4: Samplers to measure deposition of PAH in the UK Network.

2.4 Network Activities During 2017

4.1 Comparison of B[a]P annual concentrations against EC target values and UK Air Quality Objective

The annual mean B[a]P concentration measured at all the PAH Digitel (solid phase) particulate samplers are shown in Figure 5.

The data capture results calculated in this analysis is based on the percentage of valid data covering the entire calendar year. All other sites running throughout 2017 had a data capture on average of 99%.

None of the UK PAH Network measurement sites exceeded the EC target value of 1 ng/m³. However 5 sites exceeded the lower and upper assessment threshold (UAT). These were Derry Brandywell, Scunthorpe Low Santon, Scunthorpe Town, Port Talbot Margam and Ballymena Ballykeel.

The more stringent UK Air Quality Objective for PAH (0.25ng/m³ B[a]P) was exceeded at eight sites:

• Derry/Londonderry Brandywell
• Scunthorpe Low Santon
• Scunthorpe Town
• Port Talbot Margam
• Ballymena Ballykeel
• Royston
• Kilmakee Leisure Centre
• Swansea Cwm Level Park

Whilst some of the above sites have specific emission sources such as steel works (Scunthorpe sites and Port Talbot Margam), other urban may have significant solid fuel/wood use contributing to the air concentrations of PAH such as Derry/Londonderry Bandywell, Ballymena Ballykeel and Kilmakee Leisure Centre. The Royston and South Hiendley sites were positioned to measure concentrations around the Coke works in Royston, this closed in December 2014 however levels at Royston continue to be above the UK Air Quality Objective this may be due to solid fuel use in the area or possible re-suspension of previous emissions.

The sites located in Northern Ireland and those around the steel works in Scunthorpe will be discussed in a later section where polar plots are used to present the data.

4.4 Review of correlation of B[a]P and other PAHs

Whilst this report concentrates on the measurements of B[a]P which is only one PAH measured in the PAH network it is important to demonstrate that this and the approach of the UK and EU to set objectives and target values for B[a]P is sensible. To investigate if B[a]P can be considered as a ‘marker’ for other PAHs, correlations between B[a]P and other PAHs at each site based on monthly average concentration from all Digitel sampler data available from 2007 to 2017 at each of the sites. This assessment is shown visually in Figure 18. The larger darker red circles indicate high correlation whereas small blue circles indicate a negative correlation of the PAH with B[a]P.

Figure 18: Correlation between monthly average B[a]P and other PAHs concentrations at each monitoring site since the network was established (see Appendix 1 for key to abbreviations).

Correlations between B[a]P and some other PAHs (from benzo[j]flouranthene to benzo[k]fluoranthene) were consistently high across all monitoring sites. This group of PAH includes the PAH that must be analysed as a minimum within the PAH network according to the Fourth Daughter Directive (benz[a]anthracene, benzo[b]fluoranthene, benzo[j]fluoranthene, benzo[k]fluoranthene, indeno[1,2,3-cd]pyrene, and dibenz[a,h]anthracene). This confirms that over the 10+ years of PAH measurement using the Digitel samplers Benzo[a]pyrene is considered representative of the mixture of the PAH that the Fourth Daughter Directive requires.

At some locations, such as Kilmakee Leisure Centre and Lynemouth 2, B[a]P correlated well with more other PAHs than at others sites. Some PAHs, such as 5-Methyl Chrysene, Dibenz[al]pyrene and dibenzo[ah]pyrene typically have no correlation with B[a]P at most monitoring sites.

4.5 Depositon (‘C’) samples

The measured fortnightly or monthly Benzo[a]pyrene concentrations measured in deposition at the Auchencorth Moss and Chilbolton sites are displayed in Appendix 2. The levels of PAH at these rural sites in the UK are very low and as reported in the previous annual report (NPL, 2016). The levels found are often below or very close to detection limits, which makes any interpretation of the data difficult. Other PAH analysed are also regularly below detection limits. The monitoring conducted at the sites does however enable the UK to meet its obligations under the Forth Daughter Directive. All deposition data is available on the UK-AIR website.

4.7 Trend Assessment at the UK PAH Network sites

To summarise the long-term each at each monitoring site, the slope of the trend for each site was calculated using TheilSen function in the openair R package (Figure 24). A positive slope means that there is an increasing trend of B[a]P annual mean and a negative slope means a decreasing trend. The slope of each site’s B[a]P concentration over time was also calculated with its 95% confidence interval, which indicates the uncertainty of the slope coefficient. If the uncertainty of the slope covers zero, it suggests that no significant trend can be concluded. Figure 24 shows separately the sites that had a significant trend and the sites that did not.

Figure 24: B[a]P concentration trend (ng m^-3 yr^-1) from 2008 to 2017.

The trend calculated from many of the monitoring sites did not show a significant decreasing or increasing trend. It was either because there was very little change in the annual concentrations (i.e. a slope was not significantly different from zero) or there were large variations between years but not in a consistent direction (as indicated by the large confidence intervals (red bars) at the two Scunthorpe sites). For all the sites where a significant trend was observed, they generally showed a very small decreasing trend and the sites showing large decreasing trend (South Hiendley, Royston and Ballymena Ballykeel) were associated with larger uncertainties but all appear to have a significant decreasing trend in B[a]P concentrations.