1. Introduction
Air pollution represents health risks and an overall decrease in quality of life, particularly in densely populated megacities. Pollutants impact human health, but the source-receptor pathway is intermediated by atmospheric conditions and population exposure. The World Health Organization (WHO) has been playing a vital role in defining air quality standards for criteria air pollutants, based on several international epidemiological studies, defined as concentration thresholds above which adverse health effects are expected to occur to a significant parcel of the population [
1,
2]. Currently, the ozone (O
3) limits in the state of São Paulo (SP) are higher than the WHO guidelines (
Table 1).
In 2017, the state ozone standard was exceeded on 28 days in the metropolitan area of São Paulo (MASP), in 22 of 23 monitoring sites. Under these conditions, high population exposure to air pollution is expected, which is not correctly assessed by the lenient environmental laws in the MASP.
Vehicles are the greatest source of air pollution in this region [
3]. However, pollutants can be transported and affect areas tens of kilometers downwind. This is important for ozone, which is formed by photochemical reactions 1 or 2 hours after the emission of its precursors (NOx and VOCs). These reactions are triggered by sunlight and can be potentialized by higher air temperatures, above 20 °C [
4]. In the MASP, atmospheric conditions favorable for ozone concentrations have been observed during many periods in the previous years [
5,
6]. According to a broad scientific literature, the frequency and intensity of extreme climatic events are likely to increase in the following decades, so, high ozone episodes are also likely to increase in this region [
7].
As other megacities in the developing world, the MASP is a dense urban conglomeration of 39 cities, 23 million inhabitants, concentrating nearly 10% of the national GDP and marking socioeconomic contrasts. The spatial variation of life expectancy in the municipality of São Paulo alone ranges from 58 to 81 years of age [
8]. Several studies worldwide investigated air pollution exposure according to socioeconomic conditions, showing that the lower-income population, elderly and children are more vulnerable to the effects of air pollution [
9,
10,
11,
12], similarly to São Paulo [
13,
14,
15]. Recent studies on vulnerability to high temperature and air pollution demonstrated how the intraurban spatial variation of environmental conditions are associated to exposure and health effects, but additional studies are needed for a better assessment of these relationships [
16]. During high ozone episodes in São Paulo, most of the population experiences very warm and sunny conditions. The lower income population, besides living in poor housing conditions with worse wall insulation, tends to keep their doors and windows open to promote better air circulation and improve thermal comfort. However, this increases outdoor to indoor air exchanges, and so, population exposure to ambient pollution increase inside their homes [
9].
According to previous studies, increasing ozone concentrations lead to an increased risk of respiratory problems [
2]: for every 10 µgm
−3 increase in ozone, the relative risk of overall mortality increases by 0.1% [
17]. In areas such as the MASP, the population density reaches more than 50,000 persons/km
2 in the most crowded districts (
Figure S1). Therefore, defining limits which are higher than the guidelines of the WHO (
Table 1) poses a serious risk to millions of people, impairs proper environmental and health planning and indicates a lack of responsibility by the decision makers. High ozone concentrations potentially worsen asthma conditions and lead to ER visits and hospitalization [
1,
17,
18].
In the US, ozone events have been estimated to lead to billions of dollars of economic loss yearly [
18]. In Brazil, direct hospital asthma costs alone have been estimated to be more than US
$500.00 per patient yearly, not to mention the more systemic economic impacts of medication needs and absenteeism [
19]. A study comparing health and economic impacts using actual ozone averages and those recommended by the WHO shows that, by reaching the WHO levels, 50 yearly hospitalizations and 152 premature deaths would be avoided in São Paulo, leading to an overall economic gain of US
$24.9 million [
20]. Most studies on health costs use average exposure conditions. However, extreme air pollution events certainly correspond to a larger percentage of yearly average exposure. For example, events of bushfire smoke have been investigated to produce health effects on more than 30,500 people in the municipality of Albury, Australia, over the course of only 30 days [
21]. The Fuzzy Technique for Order Preference by Similarity to Ideal Situation (TOPSIS) has been used to evaluate vulnerability to air pollution and climate change in South Korea, but not using air quality modelling [
22]. Other integrative works using air quality modelling have been performed for other regions, such as the Economic Valuation of Air Pollution (EVA) model for Denmark and Europe [
23]. However, it focused on a continental scale, and few studies have been performed for the metropolitan scale. Policymakers must be aware of the real risks faced by the population in order to design and implement health and environmental management strategies which are both efficient and cost-effective.
The objective of this work is to show how outdated air quality standards lead to incorrect assessments of the impacts of an extreme ozone event. We show the number of exceedances of the air quality standards and attention levels, in the area size, the difference in the number of people affected, including those in vulnerable situations and cost analysis for the public health system. Results are shown using the current air quality standards and attention levels from the SP state, compared to the guidelines recommended by the WHO. We discuss the consequences of the lenient air quality limits of the SP State for environmental and health management and planning.
4. Discussion
Firstly, regional authorities must consider ozone a priority pollutant, as it still exceeds even the outdated SP state ozone standards. In addition, when considering the WHO limits, the number of exceedances increased much more in our simulation (from 37 to 55 exceedances). For the attention level exceedances, the number increased from five to 26 exceedances. This is probably due to the fact that the ozone attention level from the WHO (160 µgm
−3) is not much higher than what is the current standard in the SP state (140 µgm
−3). It becomes evident that, currently, the so-called “air quality standards” of SP state are close to very dangerous levels according to the international recommendations. Such pollutant levels should be treated as concentration limits to be avoided at all cost, due to their potential risks, not as a minimum acceptable standard of air quality. Concerning the difference in the size of the areas affected, caution must be taken when extrapolating model results, but our findings bring an insight into the severity of this problem. When applying stricter air quality standards, such as the WHO guidelines, the size of the area affected increases, and this should be carefully considered due to the extreme population concentration in megacities. In these settings, an increase of a few km
2 corresponds to a potential threat to thousands or even millions of people (
Table 4 and
Table 5).
The current air quality standards and attention levels must be updated in the SP state, particularly given the likelihood of increased extreme climatic events. A recent study showed that ozone exposure in Europe increased from 9% (2014) to 30% (2015), due to the strong positive temperature anomalies observed in that year over the European continent [
36], implying that exposure to ozone can increase along with increasing air temperature in other regions characterized by high ozone levels, even more so in developing countries. However, another study points out that the ozone increase in urban areas is likely observed due to the decrease of the titration effect by the decrease in urban NOx concentrations [
37]. This shows the nature of ozone pollution, influenced both by global and regional factors and calls for multiple-scale studies able to cope with this complexity.
If the WHO air pollution limits are to be attained, several improvements must be made. Policies involving the massive use of biofuels are constantly presented as more environmental-friendly options compared to fossil fuels, due to their renewability and lower emission factors of certain pollutants, such as CO. However, the use of biofuels may lead to higher VOC emissions. Many VOCs are ozone precursors, so, increased VOCs emission might result in an increase in ozone concentrations, according to their reactivity and the NOx/VOCs ratio [
37,
38,
39,
40]. Since ozone concentrations are still a major concern in the MASP and many other megacities in the world, the use of biofuels must be evaluated properly and implemented conjointly with policies that encourage the use of public transport systems, urban mobility improvements and electric vehicles, aimed to decrease vehicle activity and total emissions. In this context, public policies must be integrated in all levels of public management to provide the best conditions for mitigating emissions, concentrations and impacts of air pollution, according to their responsibilities—federal (fuel improvements, etc.), state-level (establishment of truly protective air quality limits, intercity transport) and municipal (proper healthcare management, urban transport). Policymakers must account for such events and guarantee proper investments in public health capable to deal with their consequences, such as medical procedure costs from respiratory conditions. Planning for cost must be a priority and using outdated air quality standards will only hurt long-term planning, therefore, the establishment of more rigid air quality standards is crucial.
Regarding the low-income population, whenever extreme events are forecasted or happen unexpectedly, allocating extra health agents in public hospitals and healthcare facilities near the exceedance areas will certainly help to deal with the higher number of patients seeking public health services. Public policies aimed to improve poor housing conditions, providing external coating, better isolation from outdoor air and overall cleaner indoor conditions, have the potential to decrease pollution and environmental exposure. For the asthmatics group, the results from this study provide an idea of how much cost can be avoided during extreme events if asthma is better controlled with medications, and fewer ER visits and hospitalizations are required.
5. Conclusions
In this study, we used the WRF-Chem model to simulate an extreme ozone event triggered by unusually sunny and warm conditions in the metropolitan area of Sao Paulo in 2014. We then demonstrated how different air pollution limits correspond to different scenarios of exposure for vulnerable groups and their impacts on public health cost planning during the episode. Such events are possibly more likely to occur in areas with an abundance of ozone precursors due to climate change because ozone is strongly associated with atmospheric conditions. The frequent local wind system of the sea breeze has the potential to transport air pollution to areas located downwind characterized by low income, inhabited by vulnerable population groups. It is clear from these results that policymakers should prioritize these areas for environmental management and public health in extreme ozone pollution events.
Global climate trends produce different local impacts according to the social and environmental characteristics of each area and it is paramount to understand such interactions in a fast-changing world. In this study, it became clear that the current air quality standards in the state of São Paulo severely misrepresent potential population exposure to harmful air pollution levels. When the recommended limits from the WHO were applied to the results of our simulation, the area affected by violations of the Air Quality Standards increased considerably (from 2931 to 4138 km2), along with the estimated number of people exposed. More than two million people in low-income conditions live in areas affected by ozone concentrations above the WHO attention levels, a risk not detected by applying the current SP limits. The exposure of the low-income population to environmental hazards, such as extreme atmospheric and air pollution events, has complex origins, generates systemic problems and must be confronted with equally integrated approaches. The combination of policies which ignore the WHO ozone attention levels with low-income population exposed to those limits brings about an alarming situation. The greatest difference in public health cost estimates for this event was from US$585,049 (considering SP state attention levels) to US$5,441,716 (considering WHO attention levels), nearly a tenfold increase, which suggests a mismatch in proper public health cost planning. This calls for an urgent update of the SP state air quality standards, representative of the full vulnerability spectrum of the population across a megacity in a developing nation.
Concerning future works, other pollutants also reach dangerous levels due to atmospheric stagnation in such extreme events, not to mention health effects from the extreme atmospheric conditions per se. This calls for integrative approaches, considering the vulnerability and health effects of air pollution and extreme climatic conditions conjointly. Other countries or regions not following the WHO air quality guidelines are encouraged to perform similar analyses in order to understand the importance of updating air quality guidelines for proper impact assessment, particularly considering the Global South.