3.1. Case Study
reports the mass (µg) of size-fractionated PM deposited in different sections of the respiratory tract, as a function of the season.
shows the PM mass (µg) deposited in the respiratory tract, estimated for the summer and winter periods. As reported in the figure, the PM deposited mass was higher during the winter period for all PM fractions, even if the differences between the estimates for summer and winter were minimal (<1 µg for PM1
, and PM4
; >2 µg for PM10
). Moreover, the mass deposited in the upper airways (H) contributed significantly to the mass deposited in the whole airways (total) for both summer and winter (47% for PM1
, 62% for PM2.5
, 74% for PM4
, and 96% for PM10
, on average).
Regarding the estimation of the PM deposited mass as a function of the ME visited by the commuter, as reported in Table 3
, and considering the total mass deposited in the entire respiratory tract, for the winter period the maximum values were estimated in the “Other” environments and in “Underground”, for all the PM fractions considered, followed by the “Indoor” and the “Walking (LT)” environments. The lowest values were estimated in the “Car” and “Walking (LT)” environments. For the summer period, the maximum values were estimated in the “Other” and “Walking (HT)” environments. As during the winter, the lowest values were found in the “Car” and “Walking (LT) environments.
A problem stated by the scientific literature regards the lack of data to provide a systematic basis for comparing the exposure concentrations in different transportation modes, due to different sources of variability (i.e., period of the day, season, and location) [22
]. As stated by the authors, indeed, transportation mode exposure concentrations can vary in accordance with these environmental factors (i.e., season and time of day), which are related to atmospheric stability and pollutant dispersion. Moreover, exposure concentration levels in different transportation modes may be affected by the traffic flow, by proximity to emissions hotspots, and by emissions from other vehicles [11
]. For example, Frey and collaborators, in their recent paper, reported how PM2.5
exposure concentration levels are sensitive firstly to the mode of transport, followed by the time of the day and by the monitoring season [22
Not considering the “Other” environment (as it is difficult to characterize, since it includes all the periods of transition while moving from one ME to another), for the winter period the highest values of PM deposited mass were estimated in the “Underground”, “Indoor”, and “Walking (HT)” environments. Although the time spent in the “Underground” environment was small (0.4 h) and the estimated subject ventilation rate was moderate (0.66 m3
/h), this environment was characterized by the highest PM exposure concentrations [10
], probably due to the presence of indoor PM sources (e.g., abrasion of rails, wheels, and brakes and resuspension of particles) [4
]. Conversely, the time spent in the “Indoor” environment, due to the study design, was the highest among the investigated environments (>1.5 h). Finally, in the “Walking (HT)” environment, we measured the highest pulmonary ventilation rate values (1.30 m3
/h); this could justify the high inhaled dose of pollutants in this environment. During the summer, the “Walking (HT)” environment was the environment characterized by the highest PM deposited mass values, due to the combination of a high subject pulmonary ventilation rate (1.30 m3
/h) and high exposure concentration levels. During both winter and summer, the mass deposited values were lower in the “Car” and “Walking (LT)” environments; this can be justified by the reduced permanence time in these environments (<20 min for the “Walking (LT)” environment and <1 h for the “Car” environment).
These results show how the different factors taken into account for the calculation of the inhaled dose (i.e., exposure concentration, time spent in a particular environment, and lung ventilation rate) can contribute significantly to the PM deposited mass. Even if not specifically performed in this study, a sensitivity analysis was carried out by the authors in a similar study conducted in the city of Milan; the principal results show how the parameters having a major impact on the inhaled dose are the time spent in a ME and personal exposure levels. In this case, VE seems to have a low impact on the inhaled dose, both for MEs and kinds of pollutants [24
In general, a previous study [7
] suggests how the inhaled dose of pollutants is higher during active commuting compared to motorized trips: this can be explained by the subjects’ increased minute ventilation. Another study [25
] indicates that, although exposure levels are low during walking trips, pulmonary ventilation rates are generally higher if compared to other MEs; for this reason, it is particularly important to consider both variables for the estimation of the inhaled dose (e.g., exposure concentrations and ventilation rate). It should be noted that the scientific literature also reports that the residence time is an important factor to consider in the inhaled dose estimation, as well as the pulmonary ventilation rate. In fact, active transport (walking and cycling) is characterized by higher exposure levels and inhaled doses of PM2.5
than other transport modes on a comparable trip [3
3.2. General Population
Estimation of the pollutant inhaled dose was carried out on a commuter population that usually travels in the city of Milan using the methodology described in paragraph 2.2. The estimated values of the inhaled dose of size-fractionated PM segregated by ME, time spent commuting, and gender are reported in Table 4
. These data were further subdivided according to the season (summer/winter) and the commuting period of the day (morning/afternoon). As expected, Table 4
shows how the period of permanence in each ME impacts on the inhaled dose. Furthermore, as previously discussed, higher values of inhaled doses of PM were estimated during active commuting (“Cycling” and “Walking”), due to the increased pulmonary ventilation rate. In addition, due to the lower pulmonary ventilation rate in women, it seems that women inhale a lower dose of pollutants, although there is no statistically significant difference between the inhaled doses of pollutants between women and men (p
> 0.05 for all PM fractions; Mann–Whitney U
test, performed after checking the normality—resulting neither normally not log-normally distributed—of the data distribution via Kolmogorov–Smirnov test).
Statistically significant differences (p < 0.05) were not found by comparing the two monitoring periods (morning/afternoon) but as expected, occurred as a function of the considered ME.
Following the literature [11
], the non-parametric Kruskal–Wallis test was used to assess the differences (in terms of inhaled dose) among the MEs groups. Furthermore, pairwise post hoc Mann–Whitney tests were used to further investigate the data when the Kruskal–Wallis test results were found to be significant [26
]. This test allowed the statistically significant differences to be identified within the data. However, in order to limit the Type I error rate, a Bonferroni correction was applied for each post hoc Mann–Whitney test. As such, the statistically significant value of 0.05 was divided by the number of the possible comparisons among the groups (N
= 10). The resulting value was the critical value (p) considered in the post hoc Mann–Whitney test [26
In detail, as reported in Table 5
, statistically significant differences were found between the “Walking” environment and the other MEs. Moreover, there were no statistically significant differences between the two active transport methods (“Cycling” and “Walking”).
Further differences in the inhaled doses estimated across different MEs can also occur according to the season. In fact, during winter the differences between MEs corresponded with those of the entire study period (i.e., statistically significant differences were found between active and passive commuting); in summer, however, the only statistically significant differences were found for the ME “Walking” versus the MEs “Train” and “Car” (Table S1
To provide a broader perspective to the study, the information obtained from the case study and from the general population analysis was associated with the average commuting periods of the general population commuting in the city of Milan. A summary of these data (ISTAT 2011) was shown in Figure 3
. Although the permanence time (reported by the Italian census (ISTAT 2011 and used in this part of the study)) in a particular ME (15, 30, 60, 90 min) is different according to the gender, it is possible to notice how the preferred type of commuting is walking (52% and 48%, respectively, for women and men) for short trips (15 min—Figure 3
a), followed by commuting by car (24% and 25%, respectively, for women and men) and cycling (8% for both genders). Public transport is not generally chosen for short trips (<15 min). Compared to the 15 min periods, the number of subjects who choose to travel by bike for 30 min (Figure 3
b) is reduced to 6% for both women and men. On the contrary, the number of commuters walking for a period of >15 min decreases (9% and 8% for periods of 30 min (Figure 3
b) for women and men, respectively, 2% for periods of 60 min (Figure 3
c), and 5% for periods of 90 min (Figure 3
d), for both genders) while, as expected, the use of public transport (metro and buses) increases with increasing commuting times (Figure 3
The analysis of this kind of information is important to consider, especially regarding the estimation of the inhaled dose in active commuting patterns (walking and cycling), as these are preferred to passive commuting for short trips. As reported before, the inhaled dose can be strongly influenced by the time spent in a particular ME and by the subject’s pulmonary ventilation rate. In fact, active transport is thus characterized by a higher inhaled dose of pollutant, if compared with the typical passive means of transport, due to (i) the higher pulmonary ventilation rate of the subjects and to (ii) the longer period of time spent in these kinds of environments. As said, although these aspects have now been consolidated, it is still difficult to define a trend in the study of the commuters’ inhaled dose of pollutants applicable to different urban contexts, since, in addition to environmental (i.e., concentrations of pollutants), micro-environmental, and personal (i.e., physiological parameters) variability, it is necessary to consider population mobility patterns (in turn influenced by different aspects, such as the urban layout). All these aspects can therefore contribute in defining the inhaled dose of airborne pollutants and should be considered for the personal and community choice of the best solution for urban commuting, in terms of the potential impact on health. For example, in the specific case of the city of Milan (information about mobility in the city of Milan is available in a recent study [27
]), it is possible to note that active commuting is typically chosen for the quickest routes (15 min of travel). Therefore, direct comparisons with other studies are not possible; furthermore, this suggests that each specific case should be assessed.
3.3. Limits of the Study and Future Developments
This study has several limitations: (i) the inhaled doses of pollutants were estimated along a route established a priori, which although was intended to best simulate the path of an average commuter, might not be fully representative of the entire population. Moreover, these results cannot be extended to other urban areas: in fact, the concentrations of pollutants measured in different MEs and the estimation of the inhaled dose are intrinsically characterized by a high variability, especially in urban areas. Geostatistical analyses for the description of the selected route (i.e., the analysis of the population density, land use, etc.) were not conducted. In addition, (ii) the study was carried out considering a single subject, estimating the personal pulmonary ventilation rate, certainly not representative of the entire population. Moreover, (iii) due to the study design, the evening trip (return to home) did not coincide with the evening rush times, as was done for morning commuting. Finally, it is necessary to recognize that different assumptions were used to obtain data regarding the ventilation rate and the estimated inhaled dose via the MPPD model: in this way, considering the use of different levels of approximation, it is necessary to consider the presence of an intrinsic error associated with these estimates. Moreover, the worst case (in terms of deposited mass) was considered in this study, as the clearance was not evaluated or taken into account.
For these reasons, future developments could include measures also during the evening rush hours and conducted along other routes, with the aim of improving the representativeness of this study. In addition, it would be useful to evaluate the influence of micro-environmental conditions (e.g., congested conditions) on the measurement of pollutant exposure concentrations at first and, therefore, on the estimate of the pollutant inhaled dose. Finally, the commuters’ daily exposure assessments and the contextual use of biological measurements should be considered in future studies.