br probably due do to the slight
6 probably due do to the slight variations of the traﬃc flow in their vicinity among cold and warm periods.
3.2. Allocation of ILCR in atmospheric circulation regimes
ILCR values exceeded the 10−6 threshold at all stations indicating a potential health risk due to inhalation of benzene (Table 5). Moreover, the ILCR values were considered as not recommendable (> 10−5) at 9 out of the 11 sampling sites with the exception of two background areas: PRA and MAD. The third background station (ROM), is located inside a large green area within the city of Rome (Villa Ada Park), about 200 m from the nearest street, and is considered representative of the background air quality within the Rome conurbation (Silibello et al., 2015). Benzene concentrations measured inside the Villa Ada Park are considered as an indicator of the vehicular traﬃc emissions (Avino and Brocco, 2004); hence a stronger eﬀect of traﬃc was revealed at ROM, in comparison with the other two background stations (PRA and MAD).
The allocation of these ILCR values in distinct atmospheric circu-lation clusters is also presented in Table 5. Cluster 1 incorporated 11.8%–15% of the estimated ILCR at the studied areas while cluster 2 gathered higher fractions of ILCR at MAD, ATH, ROM and PAN sam-pling sites where its influence was also connected to increased benzene concentrations (Table 3). The percentage of ILCR corresponding to cluster 3 varies from 8.7% to 14.1% at PAN and HAM respectively. Cluster 4 which had a significant impact on benzene levels at PRA, HAM, CLF, STR, WAR and GOT was associated with low/moderate contributions to the total ILCR at these areas due to its rarity of oc-currence (5.9% of total days). The impact of cluster 4 was more intense at PRA and HAM whilst the prevalence of cluster 5, which was also characterized by the influence of the anticyclonic activity over NWR, accounted for the highest fractions of ILCR among all clusters at these two areas. Despite the low benzene concentrations observed in cluster 6 (Table 3), high proportions of ILCR were associated with this Dorsomorphin synoptic pattern due to its increased likelihood of appearance (17.9% of total days). Cluster 7 participated significantly to the ILCR at ATH, ROM, WAR, PAN, PRA and STR, despite its rareness of occurrence (7.4% of total days), whilst cluster 8 and cluster 9 included important fractions
Total ILCR (highlighted red) and its allocation in distinct synoptic patterns. The percentage (%) in parentheses is the proportion of days classified in each atmospheric circulation regime.
of ILCR at all stations due to their enhanced frequency of appearance (13.6% and 13.7% of total days respectively).
At this point Incompatibility is necessary to describe some uncertainties which exist in our estimation of ILCR due to the inclusion of certain para-meters. Initially, the used BW and BR values correspond to an average adult and thus the calculated ILCR at each area may vary significantly for each person depending on his/her way of life and body functions. In addition, benzene levels in each city were obtained by only one station and thus do not reflect the actual exposure concentration of each person which strongly depends on his/her habits and programming. For ex-ample tobacco smoke has been identified as one of the main potential sources of benzene exposure (Fiebelkorn and Meredith, 2018) which was not considered in the present risk assessment. Moreover, due to data availability, most of the sites used in this research are traﬃc sta-tions, and usually urban background stations are the ones representing the general breathing exposure. Finally the selection of daily benzene concentrations for this work produced a set of days which belonged mostly to the cold periods (61.7%) and less to the warm periods (38.3%) of the studied years and therefore the calculated ILCR may be slightly overestimated due to higher benzene emissions within cold seasons. For all these reasons the calculated ILCR values are indicative but further analysis with more accurately estimated exposure con-centrations is needed.
This paper studied atmospheric levels of benzene in 11 cities across Europe in terms of synoptic circulation. Nine atmospheric circulation patterns were created from the applied cluster analysis procedure and the adverse eﬀect of anticyclonic activity on benzene related air quality was underlined. Increased benzene concentrations in central con-tinental Europe and Northern Europe were related to the influence of extended high pressure systems located over NWR, whereas weak pressure gradient fields, which are markers of light winds, were
connected with the degradation of air quality across Southern Europe due to air stagnation. Low benzene concentrations at 9 out of 11 sam-pling sites were associated with a specific circulation regime, which usually appeared within warm periods. Hence, the impact of intensified anthropogenic combustion on benzene levels during cold periods was identified.