Various epidemiological studies have shown the significant impacts of urban air pollution on human health. More concretely, several epidemiological studies have shown that the adverse health effects are related to both short- and long-term exposure to inhalable particulate matter [1
During the last decade, many researchers have used the AirQ2.2.3 model, which has been developed by the World Health Organization (WHO) in order to investigate the influence of air pollution on public health and to estimate the adverse effects of air pollution on human health, especially in an urban populated environment [6
]. Shakour et al. [6
] investigated the levels of particulate matter in selected sites in Cairo city, Egypt. The selected sites were chosen to present different activities at north and south areas of Cairo city. They applied the AirQ2.2.3 model in order to calculate the risk on human communities as a result of particulate matter exposure. The output results showed that quantifying the impact of air pollution on the public’s health has become an increasingly critical component in policy discussion.
Fattore et al. [7
] applied the AirQ2.2.3 model in two different small municipalities in a highly industrialized area of Northern Italy in order to investigate the influence of particulate matter, ozone and nitrogen dioxide exposure on human health. According to their results the AirQ2.2.3 model seems to be an effective tool for such researches, helpful in decision-making.
Goudarzi et al. [8
] used the AirQ2.2.3 model to evaluate adverse health effects caused by nitrogen dioxide exposure in Ahvaz city, Iran during 2009. They found that approximately 3% of total cardiovascular mortality, acute myocardial infarction, and hospital admission for chronic obstructive pulmonary disease happened when the nitrogen dioxide concentrations were more than 20 µg/m3
. Low percentage of the observed health endpoints was associated with low concentration of measured nitrogen dioxide.
Naddafi et al. [9
] applied the AirQ2.2.3 model in Tehran city, Iran in order to assess adverse health effects due to the exposure to particulate matter, ozone, nitrogen dioxide and sulfur dioxide. Results indicated that the estimated magnitude of the health impact underscores the need for urgent actions to reduce the health burden of air pollution.
], in order to estimate the short-term mortality impact of air pollution, applied the approach suggested by the World Health Organization (WHO), using the AirQ2.2.3 model. Daily concentrations of particulate matter, ozone, nitrogen dioxide and sulfur dioxide were used to assess human exposure and health effects, in terms of attributable proportion of the health outcome, annual number of excess cases of mortality for all causes, and cardiovascular and respiratory diseases. The results are also in line with those of other international studies that apply the AirQ model.
] investigated the health impacts of atmospheric particles with an aerodynamic diameter of 10 microns or less (PM10
) in Makkah city, Kingdom of Saudi Arabia, applying the AirQ2.2.3 model. The results of the model were discussed and compared with several studies conducted in other countries around the world.
Moustris et al. [12
] applied the AirQ2.2.3 model in order to evaluate adverse health effects by PM10
pollution in the coastal city of Volos, Greece, during a 5-year period (2007–2011). The results of the study indicated that when the mean annual PM10
concentration exceeds the corresponding European Union (EU) threshold value [13
], the number of hospital admissions for respiratory diseases (HARD) associated with PM10
increases by 25% on average. Also, they estimated an increase of about 2.5% in HARD compared to the expected annual HARD cases for Volos. Furthermore, a strong correlation was found between the number of days exceeding the EU daily threshold concentration ([PM10
] ≥ 50 μg/m3
] and the annual HARD cases.
Finally, Omidi et al. [14
] assessed the health risks associated with nitrogen dioxide in the city of Kermanshah, the capital of Kermanshah province, Iran. According to their findings, several immediate measures should be taken by the government to control the levels of nitrogen dioxide in the atmosphere to confine the related cases of mortality and morbidity.
In this study, an effort was made to assess the annual number of hospital admissions for respiratory diseases (HARD) due to the exposure of inhalable particulate matter (PM10), within the greater Athens area (GAA), Greece, for a 13-year period 2001–2013.
3. Results and Discussion
Initially, a statistical processing of the mean daily concentrations of PM10
during the examined period 2001–2013 was carried out. Figure 1
depicts the variation of the mean annual PM10
concentrations for the six examined locations within the GAA.
According to Figure 1
, the time series of the mean annual PM10
concentrations depict significant decreasing trends in all six examined sites (see Table 1
). It was found that the annual decreasing trend ranges between −0.3 μg/m3
per year (THR) and −2.4 μg/m3
per year (LYK). Furthermore, MAR, LYK ARI and PIR seem to present mean annual concentrations of PM10
above the EU annual threshold (40 μg/m3
) until 2010, however PM10
concentrations for all the examined stations remain below the EU annual threshold during the period 2011–2013.
shows the mean monthly PM10
concentrations within the GAA during the period 2001–2013. It is obvious that in all cases, there is a significant seasonality. More specifically, during the cold period of the year (October–April) in four of the examined sites (ARI, LYK, MAR and PIR) the mean monthly PM10
concentrations are higher than in the warm period of the year (May–September). This is mainly due to the traffic and secondarily due to the use of oil and wood heating systems as well. Concerning the two remaining sites (AGP and THR) things look totally different. During the cold period of the year, the mean monthly PM10
concentrations are less than the corresponding concentrations during the warm period of the year. In our opinion, this is due to different sources of PM10
that appeared in these areas. In four of them (ARI, LYK, MAR and PIR), which are characterized as traffic (T) or city center (CC) areas (see Table 2
), it seems that the PM10
sources concern mainly vehicles’ traffic air pollution in contrast to the two other suburban-background (S-BG) areas (AGP and THR). Higher concentrations may also be due to different meteorological conditions in warm and cold periods of the year, such as different mixing heights, different wind speeds, etc. The suburban-background areas are not influenced by a specific type of source, but by a synergy of sources (e.g., wind erosion, industry, traffic, heating, etc.).
In Figure 3
, the mean intra weekly variation of PM10
concentrations within the GAA during the warm (a) and the cold (b) period of the year is depicted. Higher PM10
concentrations appear during the cold period of the year against the warm period of the year, except AGP and THR. However, lower PM10
concentrations dominate during the weekend against working days of the week, in all examined areas. This leads to the conclusion that the vehicular traffic is the main PM10
source; during weekends on one hand, many inhabitants are likely to leave the city center for short excursions in the countryside and on the other hand, most of the others who remain in the city center do not use their vehicles.
The EU Directive 2008/50/EC [13
] on ambient air quality and cleaner air for Europe determines 50 μg/m3
as a daily threshold concentration of PM10
. Thus, every day when the mean daily PM10
concentration is above 50 μg/m3
is considered as an exceedance day. According to the EU Directive 2008/50/EC [13
], the number of exceedance days must be less than 35 days per year (in other words less than 9.6% of the days of a year). The annual number of the exceedances was calculated for each site during the period 2001–2013 (Figure 4
According to Figure 4
, it is clear that the annual number of exceedances appears to be higher than the EU threshold of about 35 days per year (dot line in Figure 4
) in four of the six examined sites within the GAA (ARI, LYK, MAR and PIR). However, through the years, a significant decreasing trend is presented. Regarding the AGP site, one can see that the exceedances appearing during the period 2001–2005 are eliminated thereafter. There is not any major problem concerning the annual number of exceedance days that appeared in THR site; for most of the examined years, it is lower than the EU corresponding threshold, except the period 2001–2004 where the exceedances stay around the EU annual threshold and 2010 where they are slightly higher than the EU annual threshold.
Subsequently, the AirQ2.2.3 model was applied for each one of the six examined sites in order to calculate the annual number of HARD cases. Figure 5
depicts the cumulative annual number of HARD cases due to PM10
per interval of 10 μg/m3
per 100,000 inhabitants within the GAA. More specifically, the y
axis represents the number of HARD cases and the x
axis represents the mean daily PM10
concentration per interval of 10 μg/m3
Four of the examined sites (ARI, MAR, LYK and PIR) present almost the same pattern. Furthermore, the two other sites (AGP and THR) show protective conditions with a lower number of HARD cases per year.
In order to compare the annual number of HARD cases between the different areas within the GAA, the annual number of HARD cases was normalized per 100,000 inhabitants with the application of the AirQ2.2.3 model. In Figure 6
, the percentage (%) of the annual number of HARD cases due to PM10
exposure during the examined period in each studied area per 100,000 inhabitants is depicted. The highest rate of about 21% appears in ARI and LYK, followed by 19% in PIR and 17% in MAR, while the lowest rate of about 11% appears in THR and AGP.
The mean annual number of HARD cases per 100,000 inhabitants for each studied area during the examined period 2001–2013 reveals that LYK, ARI, PIR and MAR are the most vulnerable areas (35 ≤ HARD ≤ 40) followed by AGP and THR (20 ≤ HARD ≤ 22).
In order to provide quantitative relations for the temporal variability of the annual number of HARD cases per 100,000 inhabitants and the annual means of PM10
concentrations during the examined 13-year period, scatter diagrams were constructed (Figure 7
a,b). Figure 7
a depicts that the linear model interprets 91.2% of the variance (R2
= 0.912) of the annual HARD cases. Moreover, the linear model explains 90.1% of the variance (R2
= 0.901) of the mean annual PM10
concentrations (Figure 7
b). According to the performed analysis, during the 13-year period, strong decreasing trend patterns appear in both annual HARD cases per 100,000 inhabitants (Figure 7
a) and mean annual PM10
concentration values (Figure 7
b) all over the GAA.
presents the spatial distribution of the mean annual HARD cases within the GAA in relation to the total population and the activities in each examined area. The center of Athens (ARI) is the most densely populated and air polluted area, presenting an annual number of HARD cases ranging from 60 to 262 cases, always based on the population of about 664,000 inhabitants. The port of the GAA (PIR) follows due to shipping activities presenting an annual number of HARD cases from 30 to 60 cases. Finally, all the other areas present lower HARD cases from 0 up to 30, always based on their population (see Table 1
Finally, the quantitative relation between the annual number of HARD cases and the annual number of days where PM10
concentrations exceed the European Council (EC) threshold of 50 μg/m3
] was investigated. Figure 9
indicates that the exponential fitting model explains 77.7% (R2
= 0.777) of the variance of the annual HARD cases. Furthermore, when the number of exceedance days is greater than or equal to 35, then the number of HARD cases due to PM10
exposure increases abruptly. This seems to be in full agreement with the limit number of 35 exceedance days per year, which has been established by the EU for the protection of public health.
presents a comparative analysis of HARD cases per 100,000 inhabitants due to PM10
exposure for different sites worldwide. It seems that only Volos city, Greece, and Suwon city, Seoul, South Korea, can be compared with the six examined sites within the GAA and in general within Athens city, the capital city of Greece. In all these cities, the mean annual PM10
concentrations range from 30 μg/m3
to 50 μg/m3
and the annual number of HARD cases per 100,000 inhabitants varies between 20 and 40 cases. The capital city of Tehran, Iran, follows with a mean annual PM10
concentration of about 90.6 μg/m3
and a mean annual number of HARD cases of around 77 cases per year. It is worth noting that the holy city of Makkah, Saudi Arabia, and the capital city of Cairo, Egypt, face significant PM10
exceedances, having a great impact on HARD cases per year.