Monitoring of Urban Ambient Air Quality at Locations of Preschool and School Facilities in Serbia

Exposing children to environmental pollutants during periods significant for their physiological development can lead to long-lasting health effects, dysfunction, and disease. The aim of this paper was to focus on a relatively scarce ambient AQ monitoring practice at the locations of preschol and school institutions in Serbia, within the local network of AQ monitoring stations. This was a cross-sectional study covering period 2015 to 2020. Measuring points (SO2, NO2, soot, PM10) were in front of kindergardens, primary and secondary schools in urban settlements. Results originate from the continuous public health programme conducted by the Institute of Public Health of Serbia. Only PM10 and B(a)P values at some points have exceeded annual limit values in front of chosen facilities in 2020. In sum, both the absolute number of urban ambient AQ measuring points monitored by the local network of Institutes of Public Health and those in front of the educational facilities have increased in this period: from 73/17 to 172/26. In a percentual view, these facilities are chosen more seldom within this time frame, dropping from 23.29 to 15.12% of all AQ monitoring points. As such facilities are mainly located at important city intersections, in order to assess the exposure to air pollution of vulnerable population groups (children from preschool to adolescence age), there is a need to locate measuring points near them.


INTRODUCTION
According to their location, schools in the administrative area of large cities can be positioned in the central urban core of the city, in the broader urban area and in rural suburbs, which determines the potential risks from the environment [1]. Given the length of time children spend in school, the school environment represents a significant microclimatic environment, in terms of the impact of air pollution on their respiratory health, above all. According to International Review of Curriculum and Assessment Frameworks Internet Archive children in developed countries spend 175 to 220 days a year in school, five to eight hours a day [2].
It is not a negligible fact that schools in urban areas are often located near busy roads, which causes incre-ased exposure of children to pollutants, since children attend classes during the day, every day during the working week, which coincides with the maximum traffic flux [3]. If we take PM10 particles as an example of traffic induced pollutants, their transport from the ambient air of the school environment to the indoor air of classrooms has been proven, as well as harmful cosequences for students' respiratory health [4].
According to other studies, a link has been recognized between decreased lung function, inflammatory processes of the respiratory tract and asthma, in children who live and attend school near busy roads [5; 6]. The other two pollutants, which, if measured in indoor air, represent indicators of traffic intensity in the streets near the examined school, are nitrogen dioxide and carbon monoxide [7; 8].
Schools located in rural settlements represent a different type of air pollution source, and thus a risk to the respiratory health of students. First of all, unlike traffic air pollution, where mobile sources of pollution (motor vehicle exhaust) are involved, schools in rural areas are stationary sources of air pollution. Namely, they are most often not connected to the centralized system of heat energy supply, but they mainly use heating bodies on solid fossil fuels (coal, wood) to generate heat. In the case of this type of heating of the school space, children are exposed to polluted air inside the classrooms, as well as in the yard, due to the emission of coal / wood combustion products into the ambient air [9].
The use of solid fossil fuels for heating, in terms of energy, is at the bottom of the scale in terms of combustion efficiency and cleanliness [10]. Smoke from the combustion of solid fossil fuels produces a large number of hazardous pollutants, including respirable particles, carbon monoxide (CO), nitrogen oxides, formaldehyde, benzene, 1,3 butadiene, polycyclic aromatic hydrocarbons (PAHs), like benzo[a]pyrene, along with many other toxic compounds. In underdeveloped areas, a large part of households, including schools, most often use this type of fuel for cooking and heating, so the concentrations of the listed pollutants are highest in indoor air [11]. They typically burn in energy-inefficient radiators, which combined with poor room ventilation habits can generate dense smoke in such rooms. It is important to note that children are exposed to these pollutants only during their stay in school, because after their departure, the energy source is turned off, so they are much more exposed in the classrooms themselves [10; 12].
The aim of this paper was to focus on a relatively scarce ambient AQ monitoring practice at the locations of preschol and school institutions in Serbia, within the local network of AQ monitoring stations. The specific objective was to analyze the quantitative values of air quality monitoring, in order to assess the possible need to improve the monitoring process, in terms of increasing the number of sites or adding some parameters at defined measuring points.

METHODS AND DATA COLLECTION
The research tool used in this cross-sectional study is a data base formed and regularly updated by the Institute of Public Health of Serbia (IPHS), as a segment of a continuous AQ monitoring process implemented for the Ministry of Health, through regular reporting of local IPHs from the Network. Sampling, measuring and analyzing level of defined indicators are done by environmental health departments of local / regional IPHs. Data collation, processing and analysis are undertaken by the IPHS, whose assignment is to forward its remarks on possible public health risks to decision-makers. Due to the clear market conditions established in the field of operation of accredited laboratories, the data collected at the IPHS may vary from year to year [4].
Indicators that are monitored are grouped according to the nature of pollution sources, population coverage, traffic density and potential public health risks are: SO2, soot, NO2 and other specific pollutants, such as NH3, PM10, heavy metals, and traffic-originated organic chemicals). Together with the key stakeholders at the local level (local self-government, local IPH), each municipality/town produces a Local Action Plan for AQ monitoring, defining the time schedule of monitoring activities and spatial distribution of sampling points, sources of air pollution and choice of pollutants in need of monitoring. Sulfur dioxide and BC are measured daily due to the strong correlation among their ambient air concentrations and fossil fuel combustion, as Serbia's key energy source is lignite coal [13]. NO2 is the most frequently measured specific pollutant, on daily basis. Presence of particulate pollution in ambient air is monitored through sampling of coarse particles of total atmospheric deposition (mg/m 2 /day), PM10 and PM2.5, with different measurement frequency, according to local AQ monitoring plans and specific indication, depending on pollution sources and financial resources of the local administration and the polluter, per se [14]. Mapping school locations was done using GIS.
3. RESULTS Table 1 shows precise GIS location of all measuring points (PS -primary schools, KG-kindergartens and one secondary medical school, Sec.school; SOS Childrens Village) together with the sellected pollutants monitored at each site. The exception among these educational institutions is the SOS Children's Village (Sremska Kamenica), as the site is not only a teaching, but also a residential space, where complete care is taken of children without parental care in an organized and humane way. The fact that children are going through all of their daily routine at the same spot could be the reason why at this site the most number and type of pollutants are monitored simultaneously.
Although the total number of measuring points in the Local AQ monitoring network in the period 2015-2020 is, actually, increasing, the number of educational facilities used as sampling points is in decline ( Figures  1&2.).

Concentration of specific pollutants in 2020
As shown in the three following tables, pollutants measured in 2020 in front of educational facilities were distributed in the following frequency: SO2/13, soot/13, NO2/15 atmospheric deposition/12 and PM10/5. Analyzing values of SO2, NO2 and soot, being measured daily, no exceedances of the annual limit value were recorded for any of these parameters at the measuring points close to educational facilities ( Table  2). Atmospheric deposition in front of educational institutions was monitored at 11 out of 72 measuring points in 2020, which is slightly more than 15% (15.3%) of all measuring points at 9 out of 37 urban settlements within the local AQ network (Table 3).
Only at one point, in Kragujevac, the mean annual concentration has exceeded the annual limit value of 200 mg/m 2 /day. At neither of the points mean annual values for heavy metals in atmospheric deposition (Pb, Cd, Zn, Ni) were of any significance.
Less than 3% (2.9) of all measuring points in the survey had monitoring of PM10 particles in front of educational facilities, together with heavy metals contents (Table 4). In front of 2 out of 5 schools, PM10 concentration values were higher than the average anual limit value of 40 µg/m 3 . As for the adherent specific pollutants, only values for B(a)P at all 3 points had mean ALV beyound the reccomened 1.0 ng/m 3 .

DISCUSSION
In most cases educational facilities are located at important city intersections, which is justified by choosing traffic-induced pollutants, such as SO2, NO2 and soot to be measured on daily basis at 13 out of 27 facilities [15,16,17]. On the other hand there is a significant disparity between the need to monitor the exposure of children to extremely high values of SO2 [18,19] and unnecessary monitoring of atmospheric deposition, in which increased values of any parameter are regularly not recorded [14].
Through more rational allocation of resources, some better distribution of measuring stations and determining of parameters to monitor according to continuous air pollution sources a more valuable assessment of the exposure to air pollution of vulnerable population groups (children from preschool to adolescence age), could be made. For example, the town of Bor and Krivelj village have a long history of environmental pollution due to copper mining and smelting activities, and AQ monitoring is covered with a substantial number of stations and parameters [20,21].
But, besides being an industrial town, we have children living there, attending institutions of all levels of education (5 primary schools, kindergartens, a high school and University of Technology in Bor) in which they are exposed to pollutans entering from outdoor to indoor space, being well documented through literature [22,23]. Results of the APHEKOM network study indicate that near-road-related pollution may be responsible for a significant portion of the burden of asthma in children, as well as chronic heart diseases, in adults [23].
Particularly in Bor, more emphasis should be put on arsenic and cadmium concentraions adherent to PM10 particles, due to high mobility of airborne particles from outdoor to indoor environment [24,25].

CONCLUSIONS
After having an insight into the frequency and choice of monitoring air quality at the local level in Serbia, a conclusion is imposed on an existing strong need to make indicator corrections in AQ monitoring in industrially contaminated urban settlements in Serbia and allocate financial and technical resources for continuous monitoring in front of more schools and kindergartens in Bor, Smederevo, Pančevo, Šabac, as well as determining parameters depending on the source of emissions [14, [26][27][28][29].