Air quality study: assessing variations in roadside air quality with sampling height

5 Discussion and Conclusions

The findings of the Glasgow study provided numerous insights into not only spatial variations of key air pollutants in Glasgow City centre and the influence of exposure height on concentrations observed. In addition, the study also helped to elucidate correlations between observed concentrations of different pollutants at different exposure heights and also provide useful information regarding the application of mobile monitoring and the application of personal sampler/ sensor technology for assessing human exposure to air pollutants in an urban environment. From the literature review it was evident that the study was rather unique in nature, with no other directly comparable published study identified.

A primary driver of the study was to assess the concentrations of pollutants, primarily particulate matter at child (0.80 m) and adult breathing heights (1.68 m) at a number of locations within an urban environment. However, due to its nature it was identified at the outset that it would be possible to also include other important pollutants in the study, notably NO ₂, black carbon and ultrafine particulates and enhance research into the variation of these pollutants within Glasgow City Centre. The mobile and multiple parameter nature of the study resulted in the generation of large volumes of data which were used to investigate 4 principal areas of research, these being:

• Mobile vs Fixed Monitoring
• Pollutant Concentrations vs Height
• Cross-Pollutant Correlations
• Spatial Distribution of Pollutants

5.1 Mobile vs Fixed Monitoring

Scottish, UK and European policy on assessing exposure to air pollution is currently reliant upon fixed location monitoring "in the breathing zone", which is typically accepted to be up to about 4 m above the ground. In practical terms, it is difficult to routinely measure at heights much below 1.5 m, due partly to analyser infrastructure requirements and also the potential risk of vandalism or interference with the equipment. Furthermore, the scale and technical requirements of air quality monitors that are used for LAQM and EC Directive compliance means that in general this equipment must typically operate in a fixed location, or on a large motor vehicle. Whilst this requirement provides continuity in terms of data reporting and the data quality it is recognised that such fixed monitoring may not be provide a representative measure of personal exposure to air pollution. Consequently, numerous Governments and international expert groups are currently exploring the future opportunity to utilise innovative technologies for ambient air quality monitoring purposes (Defra, 2015).

In the current study, the need to undertake mobile monitoring at multiple heights and for multiple pollutants necessitated the use of a selection of portable analysers and pervasive samplers. Whilst these samplers are not accredited for compliance monitoring purposes, the study provided the opportunity to test the technology available against data from fixed AURN monitoring sites following co-location.

From the study it was identified that concentrations of pollutants recorded by the mobile trolley along the prescribed route reported different concentrations than those reported at Glasgow Kerbside over the comparable period. In addition, the findings indicated that the observed differences were pollutant specific and in some cases influenced by the monitoring height. Notably it was identified that consistently higher concentrations of PM _2.5 (42% higher) and PM ₁₀ (47-63% higher) were recorded by the mobile monitoring trolley than at Glasgow Kerbside AURN.

In contrast, comparison of average concentrations of NO also identified that lower concentrations were observed by the mobile trolley (89%) over the prescribed study than by those recorded by the Glasgow Kerbside monitoring station. When NO ₂ was considered, the findings of the study were shown to be influenced by the inlet height on the monitoring trolley, with slightly higher concentrations (8%) being recorded at 1.68 m, and slightly lower concentrations (11%) recorded at 0.80 m when compared to concentrations recorded at Glasgow Kerbside over the same study period.

Overall, the findings appear to indicate that in certain circumstances, fixed monitoring may either under- or over-estimate the concentrations of pollutants to which members of the public are routinely exposed to in urban environments. However, this is perhaps not overly surprising, given that by being mobile, the trolley experiences greater variation in its proximity to pollutant sources (e.g. road traffic, commercial sources and fugitive emissions), but also exposure to localised meteorology. This finding may also be influenced by the height of the inlet at the fixed monitoring site which is significantly higher than either of the inlet heights on the mobile monitoring trolley. There may thus be some benefit in evaluating the influence of inlet height on concentrations measured at a fixed monitoring site to ascertain the influence on concentrations reported.

The finding that average concentrations of PM _2.5 and PM ₁₀ were significantly higher along the mobile monitoring route is particularly interesting. In many circumstances, due to their proximity to road traffic emissions, kerbside and roadside monitoring stations are often considered to represent 'worst case' urban pollution. However, this finding questions the validity of this general belief and merits further study to help confirm these findings.

5.2 Pollutant Concentrations vs Height

One of the principal objectives of the study was to determine the nature of the relationship between pollutant concentrations and height; in this case, 2 exposure heights, average child and adult breathing height.

The study investigated the relationship between exposure height and pollutant concentration across the entire mobile study route, but also sought to undertake a basic evaluation of the potential influence of different environment types along the route on any relationships identified. These environment types were:

• A busy street canyon where traffic is dominated by high numbers of buses and taxis (Hope Street)
• A busy A-road with high volumes of traffic (High Street)
• A predominantly pedestrianised section of the route (Sauchiehall Street/ Buchanan Street)

The data from the study demonstrated the complexity of pollutant concentrations in the ambient environment, but several findings were consistently observed. Data obtained for the entire study route and each of the individual sections outlined above indicated that concentrations of CO and PM _2.5 observed were not significantly influenced by the inlet height (height of exposure), and the observed influence of inlet height on concentrations was too variable to draw a clear conclusion. In contrast, average concentrations of PM ₁₀ were shown to be consistently higher (up to 12.6% higher) at 0.80 m than at 1.68 m throughout the study route. This finding is significant, as it appears to indicate that in the urban environment children may be exposed to higher concentrations of PM ₁₀ on average than adults. This may reflect the view that the coarse PM fraction is deposited faster and thus may exert a more significant local impact than the finer fraction which may be transported over longer distances.

However, as no significant influence of height was observed for concentrations of PM _2.5, the respirable fraction of PM, it is recommended that further research is undertaken to confirm the findings and ascertain if this observed effect is restricted to PM with an aerodynamic diameter of more than 2.5 m, or whether the low concentrations of PM _2.5 simply make identifying the effect more challenging.

For NO ₂, at lower ambient concentrations, no consistent influence of exposure height was identified on observed concentrations. However, at higher ambient concentrations, such as that may be observed in close proximity to busy road junctions, significantly higher concentrations were observed at adult breathing height (1.68 m) than at child breathing height (0.80 m). This finding is significant, as it appears to indicate that in the urban environment adults may be exposed to higher concentrations of NO ₂ on average than children. Given that this influence was only noted at relatively high ambient concentrations, it is possible that the finding is influenced by proximity to local sources, notably elevated road traffic, as emissions of NO ₂ will be associated with exhaust gas emissions, which being hot may rise upwards immediately following release from the exhaust before dispersing more widely in the environment. However, given that this finding was also observed on Sauchiehall Street and Buchanan Street sections of the study route which are largely pedestrianised this possibility must be questioned. Before any definitive conclusions are drawn from these results, it is recommended that further research is undertaken to confirm and help elucidate the conclusions.

5.3 Cross-Pollutant Correlations

The ambient urban environment commonly includes a complex mixture of pollutants from a range of sources. As such, humans are commonly exposed to a cocktail of pollutants, and in some situations, the presence of certain pollutants is often used as a marker to indicate the presence of other 'associated' pollutants. For example, whilst NO ₂ has been associated with adverse effects on hospital admissions, detrimental effects on lung function and prevalence of asthma, in the past it has often been questioned whether these associations are caused by NO ₂ itself, or due to some other pollutant with which it is correlated with in ambient air ( COMEAP, 2015)

Data from the study demonstrated the extreme complexity of intra-pollutant relationships. Inconsistent or weaker correlations were observed between most of the pollutants studied, whilst consistently moderately-strong to strong correlations were identified with the PM fractions PM _0.5 with PM _1.0 and PM _2.5, and separately PM _2.5 with PM _5.0 and PM ₁₀. These findings are perhaps not surprising but suggest that PM behaviour is related particle size, with the course fraction being deposited faster than the fraction <PM _2.5.

5.4 Spatial Distribution of Pollutants

Perhaps the most interesting findings from the study were generated by the spatial visualisation of the concentrations of pollutants measured by the mobile monitoring trolley along the prescribed monitoring route. In general, higher concentrations of pollutants were readily observed in areas where high road traffic volumes and traffic congestion were observed. Some of the most notably areas where elevated concentrations of pollutants were observed included Hope Street, High Street and around George Square, all 3 of which are areas of the study route where high traffic volumes and congestion are commonly observed.

Elevated concentrations of BC and UFP were more pronounced and visualised more clearly than other pollutants. This may be related to the way that these pollutants are reported or alternatively may indicate that road traffic is the main local source for these pollutants. Concentrations of BC and UFP were found to decrease along Sauchiehall Street and Buchanan Street which are largely pedestrianised, however, this finding was not mirrored in PM _2.5 concentrations, where increased concentrations were observed. These findings perhaps indicated different sources for each pollutant, or more likely reflect the different ways that the pollutants disperse. Fugitive sources such as re-suspension of particulates also may have a greater influence on the spatial variability of PM _2.5 concentrations.

The spatial visualisation of pollutants through the use of the mobile monitoring trolley shows significant potential for the identification or confirmation of pollutant hotspots within urban environments which are most commonly predicted through the use of dispersion modelling. Through the application of such a mobile monitoring 'screening' approach, authorities may be able to confirm the presence and extent of pollution hotspots, and thus investigate, design and implement appropriate mitigation measures. Furthermore, the capacity of the mobile monitoring approach to identify not only increased concentrations of pollutants associated with road traffic, but also elevated concentrations of PM _2.5 and BC in close proximity to the construction site at Strathclyde University indicates the potential application of the technology for assessing the potential air quality impact of proposed developments. This could relate not only to the construction and demolition processes, but also pre- and post-construction impacts which are currently challenging to monitor particularly when PM is a key consideration.

5.5 Possible Implications for Policy

A wide range of air quality monitoring is undertaken by the Scottish Government to fulfil the requirements of EU Directive 2008/50/ EC on ambient air quality, and by local authorities under the Local Air Quality Management regime ( LAQM) as set out in the Environment Act 1995 and associated regulations.

Although both the Directives and the LAQM regime are primarily focused on protecting human health, current monitoring approaches do not consider potential variations in air quality with height above the ground. The heights of monitoring station sampling points vary depending on local conditions and with type of equipment in use. The findings of the study indicate that in Glasgow City Centre, concentrations of key AQS pollutants, notably PM ₁₀ and NO ₂ reported by fixed monitoring sites may not provide a good estimate of an individual person's exposure to these pollutants in such an urban environment. However, in reality, compliance monitoring sites are not designed or expected to represent individual personal exposure, but provide reliable and consistent data of concentrations of air quality pollutants representative of different environment types (e.g. kerbside and urban background) to inform compliance reporting and support the Scottish Government, UK Government and European Commission in targeting overall improvements in ambient air quality. The results from the Glasgow study are very informative and provide valuable insight into the potential variations of air pollutants in the urban environment. However, the study itself does not aim to, and does not provide a true reflection of personal exposure, as individuals don't tend to walk around urban environments constantly, but tend to spend time travelling and indoors at home or at work where they will be exposed to other sources of pollution.

A potentially important finding of the study was the suggestion that concentrations of some pollutants are influenced by the height of exposure/ monitoring inlet height. As the Air Quality Directives and the UK Air Quality Strategy are primarily concerned with protecting human health, the finding that in some urban environments that children may be exposed to higher concentrations of PM ₁₀ than adults is something that merits further investigation. In terms of monitoring policy, sample inlets used at fixed site are in many cases are set at a height much greater than 1.68 m; typically between 2.5 m and 3.5 m. The results from this study indicate that pollutant gradients exist for some pollutants between 0.80 m and 1.68 m and as a result, this may need to be taken into account in the future policy. For example, it is not known if a concentration gradient exists between 0.8 and 3.5 m: could a relationship be derived between pollutant concentrations and the height at fixed monitoring sites up to 3.5 m, and would this be representative of specific site types (e.g. Kerbside or Urban Background)? This could potentially provide a method to improve estimates of pollutant concentrations below the sample inlet height of a particular station.

Another important finding of the Glasgow study was that higher concentrations of PM _2.5 were observed in pedestrian sections of the monitoring route than in close proximity to many of the roads. As the Scottish Government is currently considering the addition of PM _2.5 to the LAQM regime in Scotland, and the associated design of a PM _2.5 monitoring network, the findings from this and related studies may help inform this process, particularly the geographical spread of monitoring sites. However, it is recommended that this finding is evaluated in other pedestrianised/ urban background locations including in other towns and cities to determine whether this finding is more widely observed.

In passing, the study has also undertaken an evaluation of some personal air quality samplers and pervasive sensors which don't currently meet the requirements for air quality compliance monitoring. This applicability of such analysers for ambient air quality monitoring is currently the focus of considerable research across the UK and Europe and consequently not definitive conclusions should be drawn from this study. However, whilst the sensors are not currently suitable for compliance monitoring purposes, or for the declaration of an Air Quality Management Area, for example, in certain circumstances such technology could provide valuable insight into the spatial variation of pollutant concentrations, which with careful processing could potentially become valuable screening techniques (similar to NO ₂ diffusion tubes). With appropriate research and the development of guidance, some of these technologies have the future potential to support the refinement of local dispersion models and the national compliance model and thus enhance the assessment of population exposure.

5.6 Recommendations for Further Research

The findings of this study are limited by the number of monitoring exercises carried out, with a total 8 mobile monitoring exercises and 11 co-location exercises during a six month period. As a result, it was not possible to capture all conditions (e.g. traffic, weather and pollution episodes), although the potential for doing so was maximised through monitoring on different days of the week; in different months and seasons of the year; and in a variety of urban microenvironments. Therefore, the results can only be considered to represent a snapshot of conditions within Glasgow City Centre.

There is obviously opportunity to enhance and refined the findings of the study by repeating the study and undertaking a more extensive mobile monitoring exercise using a number of towns and cities around Scotland throughout a calendar year. There are also a large number of potential studies that could be undertaken and bring value to air quality research in Scotland. However, for the purposes of this report there is potentially more benefit from focussing on confirming some of the key findings of the Glasgow study and where these may support the current and future air quality policy of the Scottish Government. Therefore, it is recommended that further research could focus on:

• Investigating the variation in concentrations of pollutants (most notably NO ₂, PM ₁₀ and PM _2.5) with height at a number of fixed monitoring sites with different site classifications (e.g. kerbside, roadside and urban background) to determine whether the variations in pollutant concentrations observed by the mobile monitoring trolley are also observed at compliance monitoring sites.
• Investigating the spatial differences in ambient concentrations of PM _2.5 and PM ₁₀ observed in the current study. Such a study could help to inform Scottish Government policy with respect to PM, notably the potential inclusion of PM _2.5 within the LAQM regime and the associated development of a PM _2.5 monitoring network.