Benzene releases from gasoline stations - Implications for human health: Uncertainties in evaluating risk

On this page

Uncertainties in evaluating the risk to human health

Recent Canadian benzene concentrations based on monitoring data at or near gasoline station operations were not available. Estimates of benzene concentrations emitted from gasoline stations were based on modelled values of total evaporative losses, which were then used with air dispersion models to estimate the benzene exposure concentrations for residents living within the vicinity of gasoline stations. Modelling inherently includes uncertainties relating to the assumptions, input data and modelling tools used. Model validation with monitoring of benzene concentrations in air near gasoline stations under a variety of conditions would further advance the understanding of potential exposures to benzene near gasoline stations. To evaluate the impact of model choice, an alternative analysis using AERSCREEN, an alternate air-dispersion model, was conducted (refer to the following section on air dispersion model).

The assumption is made that an individual is continuously exposed and that the indoor air concentration of benzene resulting from gasoline station emissions is equal to the modelled outdoor air concentration.

This assessment only considers exposure to benzene associated with evaporative emissions from gasoline station operations and with vapour release from underground storage tanks during tanker truck unloading. Additional exposures to benzene from spills or environmental contamination associated with gasoline stations were not considered in this report. Also, other sources of benzene exposure, such as smoking or the presence of an attached garage (Health Canada 2013), were not considered. Furthermore, this assessment only considered the exposure from a single gasoline station. At busy urban intersections, it is not unusual to find multiple gas stations within close proximity of one another. In these cases, the emissions from these multiple gas stations may reach the same residential areas, which would increase the exposures of members of the population living there (Hsieh et al. 2021).

The TC05 for benzene was based on a lifetime of exposure estimated using adult breathing rates and body masses. Infants and children may be more affected by the benzene concentrations estimated in this analysis due to differences in breathing rate and body weights. In addition, Health Canada (2013) identified a range of potential cancer potencies associated with a 1 in 100,000 risk level (0.6 to 4.5 µg/m3) based on Canada (1993), OEHHA (2001) and U.S. EPA (1998). An alternative analysis was undertaken to evaluate the impact of selecting the most health-protective value from this range on the health risks from gasoline station benzene emissions (see the following section on cancer potency factor).

In June 2021, the Government of Canada announced that it is setting a mandatory target for all new light-duty cars and passenger trucks sales to be zero-emission by 2035, to help put Canada on a path to achieving its long-term goal of net zero emissions by 2050 (Transport Canada 2021). The reduction of gasoline powered vehicles in the marketplace would be expected to result in a corresponding reduction in gasoline station benzene emissions and related risks to health.

Air dispersion model: alternative analysis

SCREEN3 is a screening-level Gaussian air dispersion model that was developed based on the Industrial Source Complex (ISC) model for assessing pollutant concentrations from various sources in an industrial complex. SCREEN3 is designed to estimate maximum concentrations of chemicals at chosen receptor heights and at various distances from a release source for a given continuous emission event. AERSCREEN (US EPA 2011) was also used to conduct an alternative analysis to determine the effect of the air dispersion model on the predicted exposure concentrations. AERSCREEN is the screening model based on AERMOD (US EPA 2011). AERSCREEN produces estimates of "worst-case" 1-hr concentrations for a single source, without the need for hourly site-specific meteorological data, and also includes conversion factors to estimate "worst-case" 3-hr, 8-hr, 24-hr and annual concentrations. AERSCREEN is intended to produce concentration estimates that are equal to or greater than the estimates produced by AERMOD, without a fully developed set of meteorological and terrain data (US EPA 2011).

Calculations were conducted with AERSCREEN to determine the long-term benzene exposures from a gasoline station for those living in the vicinity of the station. The input variables for the AERSCREEN calculations are given in Table B-5 of Appendix B. For a median throughput gasoline station, at a distance of 25 m from the station fenceline, average annualized contributions to modelled benzene concentrations were estimated to be 1.1 µg/m3. When compared to the TC05 value of 14,700 μg/m3 (Canada 1993), this concentration leads to a MOE of 13,400, an incremental risk of cancer of 4 per million population, which is less than the risk of 12 per million estimated using SCREEN3. The MOEs from the SCREEN3 and AERMOD models are given in Table 6. The distance from the gasoline station fenceline at which the incremental benzene concentration from the gasoline station corresponds to the 1 in 1,000,000 risk level of 0.29 µg/m3 is estimated at 75 m for the median-throughput gasoline station scenario using AERSCREEN (Table B-6 of Appendix B), which is about half the distance estimated using SCREEN3 of 160 m.

Short-term benzene exposure concentration based on the benzene release rate and the exposure factors were determined using AERSCREEN (Table B-7 of Appendix B). In addition, maximum incremental benzene concentrations for 1-hr of exposure at different distances from the vent pipe array without vapour recovery from the unloading of a 35,000 L delivery tanker truck were also calculated using AERSCREEN and are provided in Table B-8 of Appendix B. Comparing an estimate of ~260 µg/m3 at a distance of 25 m from the fenceline with the effect level of 16 mg/m3 based on developmental hematotoxicity in mice results in a MOE of 62. This is larger than the MOE of 20 obtained with SCREEN3 (see Table 3). The AERSCREEN modelled incremental benzene releases from the vent stacks are less than the AREL of 27 μg/m3 at a distance of 150 m from the gasoline station fenceline for the 35,000 L tanker trucks. This is similar to the estimated distance of 210 m based on SCREEN3.

Table 3. Comparison of the MOEs and risks for long-term and short-term exposures calculated with SCREEN3 and AERSCREEN
MOE Cancer risk per million populationFootnote a MOE Cancer risk per million populationFootnote a
Long-term exposure from median throughput gasoline station 4,200 12 13,400Footnote b 4
Short-term exposure from unloading from 35,000 L tanker truck 20 - 62Footnote c -


Cancer risk per million is not relevant to the short-term exposure estimates

Return to footnote a referrer


20 m from the fenceline

Return to footnote b referrer


25 m from the fenceline

Return to footnote c referrer

The results of the AERSCREEN calculations are in agreement with the results obtained using SCREEN3 for the exposure scenarios of interest. For both long-term and short-term exposures, the estimates from SCREEN3 and AERSCREEN indicate that the benzene exposures attributable to gasoline station emissions may pose unacceptable risks to human health for the general population living in the vicinity of gasoline stations.

Cancer potency factor: alternative analysis

An alternative analysis was undertaken in which a 1 in 100,000 cancer risk is associated with a chronic benzene exposure level of 0.6 µg/m3, which represents the most health protective value from a range of estimates (0.6 µg/m3 to 4.5 µg/m3) identified in Health Canada (2013) and which is derived from OEHHA (2001). This corresponds to a concentration of 0.06 µg/m3 associated with a cancer risk of 1 in 1,000,000.

Based on this cancer risk and benzene concentrations generated with SCREEN3 (Appendix B, Table B-2), gasoline station benzene emissions would contribute to cancer risks above 1 in 100,000 at distances up to 30 m, 105 m and 190 m for baseline, median and high throughput stations, respectively. Corresponding distances for cancer risks above 1 in 1,000,000 are up to 180 m, 400 m and 700 m (Table 4).

Table 4. Distance to gasoline station fenceline for concentrations representing 1 in 100,000 cancer risk or 1 in 1,000,000 cancer risk based on 2 cancer potency values
Cancer potency Gasoline station throughput Distance (m) from fenceline for concentration with 1 in 100,000 cancer risk Distance (m) from fenceline for concentration with 1 in 1,000,000 cancer risk
Derived from TC05 (Canada 1993) Baseline - 70
Median 25 160
High 70 290
Low end of range identified in Health Canada (2013) Baseline 30 180
Median 105 400
High 190 700

The results indicate that using the cancer potency factor from the low end of the range of factors identified in Health Canada (2013) would increase the distance within which benzene emissions from gasoline stations increase the risk of cancer for nearby residents.

Page details

Date modified: