- Discussion
(continued)
6.5 Influence of Stage I and II Controls
The results indicate that Stage I controls are effective in reducing benzene concentrations by approximately 65% within the forecourt area, although at some stations (E, J and K) concentrations still remained above the AQS. Stage I controls in combination with Stage II resulted in a much greater (approximately 93%) reduction in forecourt measurements, which were significantly below the AQS.
Stage I controls were fitted at Station H where continuous benzene measurements were made in close proximity to the delivery tanks. The lack of a causal relationship between short term peaks in benzene concentrations recorded at this site and delivery dates may be attributable to the operation of Stage I controls during delivery which would have caused recycling of displaced benzene vapour and prevented significant emissions to the atmosphere. However, in the absence of information concerning wind speed and direction it is not possible to examine in detail the relationship between peak concentrations and the petrol delivery schedule. For example, there may have been occasions when petrol was delivered and the prevailing wind was in the opposite direction to the inlet for the BTX analyser, thereby significantly reducing the likelihood of the analyser detecting any change in benzene levels associated with delivery.
6.6 Comparison with Other Studies
Comparison of the data measured in this study with those reported for other petrol station sites is not straightforward as concentrations are influenced by the size of the station, fuel sales, traffic flow along adjacent roads and type of emission controls fitted. Furthermore, differences in methodology, in particular exposure periods and types of adsorbent, means that only broad comparisons can be made.
The range of values measured at sites located at the petrol pumps in the present study are consistent with previously reported data; Little and Cram and CONCAWE reported mean values of 12.8 ppb and 15.1 ppb respectively; these compare with levels of 1.9 to 19.1 measured in this study.
In the study currently being conducted by Bristol City Council, benzene levels are being monitored at two stations using identical methodology to that employed in the present study. The characteristics of the two Bristol stations, in terms of numbers of service bays, are similar to five stations (D, I, J, K and L) in this study. Comparison of the data measured in the forecourt areas of these sites indicate that values for these stations were lower than those reported for the Bristol sites, e.g. a maximum forecourt mean of 8 ppb recorded in the present study compares with means of 10.4 and 28.4 ppb recorded at the two Bristol sites.
Measured forecourt means in the present study were consistent with means of 4.5 to 6.3 ppb reported for boundary sites in the preliminary CONCAWE study although these data were collected over a period of 24 hours and thus may not be representative of more typical conditions. When a more detailed one year study was undertaken using methodology identical to that employed in the present study benzene levels recorded at three sites at the sides and rear of a semi-rural petrol station were lower than in the preliminary study, ranging from 1.7 to 3.2 ppb. Such a range of values is consistent with data reported for the majority of the stations in this study with the exception of Station A (4.8 to 8 ppb), D (3.8 to 5.6 ppb), J (4.5 to 7 ppb) and K (4.8 to 8 ppb). Data in the present study are also consistent with levels reported by Stanger Science and Environment for site perimeters of petrol stations in the London Area (annual means ranging from 2 to 5 ppb).
There are no reports in the literature of continuous real time measurements at petrol forecourts, although CONCAWE have previously used active sampling techniques to provide 12 hour and 24 hour averages. In these studies 12 hour means ranged from 1 to 99.12 ppb with 24 hour benzene concentrations ranging from 5.2 ppb to 387 ppb. Although the methodology was different from that used in the present study, the peak values reported by CONCAWE were well in excess of both the 30 minute and daily averages reported here. The range of daily means recorded by the BTX analyser in the present study (1.7 to 4.5 ppb) were also lower than the range of 8 to 13 ppb suggested by Little and Cram as being typical of ambient mean daily concentrations at petrol stations.
Measured benzene levels at the roadside sites in the present study are generally consistent with previous data reported for similar sites in the vicinity of petrol stations. For example average mean values recorded at four off-site roadside locations in the Bristol study ranged from 1.9 to 4.8 ppb, compared with the range of 1.2 to 4.7 ppb at the 24 roadside sites in the present study. Similarly the mean value of 4.4 ppb recorded in the 12 service station CONCAWE study is within the range of values reported in the present study. However, in the second CONCAWE study, which took place at a semi-rural station during 1993, mean roadside values were much lower than in the present study, ranging from 0.3 to 0.5 ppb. These lower values are likely to be a reflection of the lower traffic volumes and hence lower benzene emissions characteristic of semi-rural locations. Benzene concentrations recorded in the present study are also typical of values normally measured at roadside locations in the Greater London Area. During 1995 measured benzene concentrations at 44 roadside sites varied from 2 to 6 ppb, consistent with the range reported in the present study.
As would be expected mean benzene concentrations recorded within the petrol station forecourts and at the roadside sites are higher than measurements previously reported for urban background locations. The overall mean for all roadside and forecourt sites of 2.8 ppb and 3.8 ppb compare with a mean of 2 ppb recorded using diffusion tubes at 32 background sites across London in 1995. Similarly the mean benzene concentration measured by the BTX analyser of 2.8 ppb is higher than the range of 0.7 ppb to 1.8 ppb recorded as 1994 annual averages at the National Hydrocarbon Network sites which are situated at urban background locations, however this is only an approximate comparison as a strict comparison between a 3 month average and annual average is not possible. The maximum hourly mean of 9.4 ppb recorded by the BTX analyser is comparable with maximum hourly means recorded by real time analysers as part of the National Hydrocarbon Network during 1994 which ranged from 10.7 ppb to 44.6 ppb at the AUN stations
6.7 Assessment of Seasonal Variation
Meteorological conditions which determine the dilution and dispersion of atmospheric pollutants generally result in higher ground level concentrations in winter months. Winter to summer ratios for benzene data collected in 1994 and 1995 at the sites in the UK Hydrocarbon Network are provided in Table 12 and clearly show winter months to be characterised by higher concentrations. Such a seasonal difference is likely to be attributable to more stable atmospheric conditions in the winter months resulting in higher ground level concentrations.
Table 12 Winter/Summer Ratios at the National Hydrocarbon Network Sites
| Site Name |
1994 |
1995* |
| Birmingham East |
1.8 |
1.4 |
| Middlesborough |
1.1 |
1.3 |
| London UCL |
1.3 |
1.3 |
| London Eltham |
1.3 |
1.5 |
| Edinburgh |
1.5 |
1.3 |
| Belfast South |
1.5 |
1.6 |
| Cardiff East |
1.6 |
1.3 |
Note - Summer represents the period from 1 April to 30 September and Winter represents the period from 1 October to 31 March
* Data derived from the National Archive
The influence of meteorological conditions on benzene levels at petrol stations is likely to be more complex than at the UK Hydrocarbon Network sites. At petrol stations benzene levels are influenced by both vehicle exhaust emissions and evaporative emissions, whereas at the network sites levels will be predominantly influenced by exhaust emissions. Consequently peaks in benzene levels may occur at petrol stations during both winter and summer months, the former being driven by stable atmospheric conditions and the latter by high temperatures causing greater evaporative emissions.
6.8 The Auto Oil Programme
On the basis of only three months data, as obtained in the present study, it is not possible to assess any influence of prevailing meteorological conditions on temporal variation in the data set. Previous studies6 in which benzene has been monitored at petrol station sites have shown ambient benzene levels recorded around boundary fences in winter to be twice as high as the corresponding concentrations in summer.
In response to European Directive 94/12/EC covering emissions from cars the European Commission, vehicle manufacturers and the oil industry embarked upon a collaborative programme known as Auto Oil.
The aim of this programme was to work out air quality targets for the next century, and the most cost effective way of achieving them.
The three year programme included:
- a study of the nature and extent of likely future air quality problems in relation to WHO thinking on air pollution standards. This included modelling conditions in Athens, Cologne, the Hague, London, Lyon, Madrid and Milan;
- the European Programme on emissions, fuels and engine technology, a research project to clarify the relationship between fuel and engine technology in determining emission levels, and the potential for improving one or both;
- the contribution of better vehicle maintenance and more stringent annual inspections;
- local initiatives to reduce the number of vehicles on the road and journey distances.
The strategy that emerged from the programme in June 1996 included new air quality objectives as well as proposals for two technical directives on car emissions and fuel composition respectively.
In terms of benzene the Commission report that modelling results obtained as part of the Auto Oil programme have shown that by the year 2010 atmospheric benzene levels will be sufficiently low even without the application of additional measures. In light of this the Commission does not view it necessary to impose a limit of benzene in petrol below 2%.
Report prepared by Stanger Science and Environment
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
