Ambient Air Quality in Upper Silesia Region Pre-During, and Post-COVID-19 Periods

The study was carried out in Gliwice (50°17′37.1″N 18°40′54.9″E) which is the westernmost city of the Upper Silesia Region of Southern Poland. The city belongs to a conurbation of 2.0 million people, and is the third-largest city in this area, with 175,102 permanent residents as of 2021 (Fig. 1). Silesia is the most industrialized region of Poland. The main local industries are mining, iron, lead, zinc metallurgy, energy, mechanical engineering, automotive, chemicals, building materials, and textiles. There are currently 19 coal mines in operation. The region represents one of the largest industrially degraded areas in Poland.

Figure 1.

Regarding pollution, Silesia voivodeship accounts for 8.9% of PM_2.5, 8.5% of PM₁₀, 10.2% of NO₂ and 15.9% of SO₂ country's gas emissions. Industrial production, coal-based power, and heat generation are major contributors to air pollution by SO₂ and NO₂ (77% and 60%, respectively). While PM_2.5 and PM₁₀ in the region are mainly emitted from household sources (87% and 76%, respectively) [24]. Due to high levels of air pollution, Silesia has the lowest life expectancy in Poland and the highest incidence of premature birth and birth defects [25].

The outdoor air pollutants included bacterial aerosol (BA), PM_2.5, PM₁₀, NO₂, and SO₂ levels and meteorological parameters such as relative humidity (RH), air temperature (t), atmospheric pressure (P), wind speed (Ws), and wind direction (Wd) measurements. The samples were collected during the spring seasons (March from 2018 to 2023).

The data on PM_2.5, PM₁₀, NO₂, SO₂, and meteorological parameters (Table 1) were collected by the mobile air monitoring laboratory for air pollutant immission measurements located at Silesian University of Technology in Gliwice. The measuring equipment includes continuous automatic certificated monitors for PM₁₀/PM_2.5 (particulate matter Beta Attenuation Monitor BAM1020 Met One Instruments, Inc.), SO₂ (fluorescence analyzer – T100/API-Teledyne), NO₂ (chemiluminescence analyzer – T200/API Teledyne), and the meteorological station Meteo set WS 500 Lufft. Air pollutants data are 24 h averages.

During the study mobile laboratory was in one location, so the calibration of the NO₂ and SO₂ analyzers using certified standard gases was performed once a year, according to the producer's recommendations, while the flow checks were performed every 3 months. The calibration of the PM monitor was conducted through the factory calibration with leak and flow checks performed monthly and when the filter tape was changed.

Two sets of measurements of BA were performed at a height of about 1.5 m to simulate aspiration from the human breathing zone. Samples of microorganisms were collected between 11:00 and 12:00 local time. Sampling of culturable bioaerosols is generally performed periodically. Unfortunately, good methods for constant air sampling of culturable microorganisms are not available, and air sampling for a period of more than 15 min is often not possible, whereas air concentrations vary in time [26]. To check the variability of BA during the day the daily measurements were performed during the non-heating season. Through five randomly selected days four sets of BA measurements with meteorological parameters were performed at 9:00, 12:00, 15:00 and 18:00 o'clock. Figure 2 presents the obtained data of BA measurements with meteorological parameters. Despite the fact that the highest BA concentrations were obtained around noon, it can be seen that the standard deviation of hourly averages is from ±5 to ±12% in comparision to daily averages.

Figure 2.

BA concentrations were measured using an Air Ideal single-stage impactor with an airflow of 100 dm³ min⁻¹. The sampling time was 1 minute and tryptic soy agar (TSA) supplemented with cycloheximide to inhibit fungal growth was used for bacterial growth. Results were calculated as a total number of colonies and expressed as colony-forming units per cubic meter of air (CFU m⁻³).

Data were analyzed using Statistica software (TIBCO Software Inc.), version 13.3 for Windows, and a p-value < 0.05 was considered statistically significant. Due to the small data set for each season (n < 50), two tests were used: the Lilliefors test and the Shapiro-Wilk test to check the normal distribution of the data. Normality was demonstrated for meteorological parameters. None of the pollutants in none of the studied periods showed compliance with normal distribution was established for every of the studied periods hence Spearman's rank correlation was used, for the obtained variables generated from daily time series data, to see if there was a correspondence (strength and direction) between total bacteria levels, air pollutants, and meteorological parameters.

In 2016, the World Health Organization (WHO) and the International Agency for Research on Cancer (IARC) classified ambient particulate matter (PM) as Group 1 – compounds carcinogenic to humans [27]. The studies conducted in Poland obtained that the local emissions cause approximately 1600 attributable deaths and 29,000 disability-adjusted life-years (DALYs) per year and that about 80% of this health burden was due to exposure to PM_2.5 [28], and short-term exposure to PM_2.5 and PM₁₀ results in an increase of relative risk of premature death by 0.7% and 0.3% per 10 μg m⁻³, respectively [29].

In Gliwice, Upper Silesia Region we observed that there is a generally lower concentration trend for PM_2.5 and PM₁₀ throughout the lockdown phase (2020/2021) compared to the results of the concentration from the pre-COVID 19 phase (2018/2019) (Fig. 3 and 4). The average concentration for the pre-COVID-19 phase was 46.45 μg m⁻³ for PM₁₀, and 39.28 μg m⁻³ for PM_2.5 while the average concentration during the COVID-19 period was 40.53 μg m⁻³ and 28.35 μg m⁻³ for PM₁₀ and PM_2.5, respectively. For both PM fractions the highest concentrations were observed in the pre-COVID-19 period.

Figure 3.

Figure 4.

Although there was a significant PM_2.5 reduction in Gliwice (Table 2), the PM_2.5 concentrations exceeded 24h average concentrations recommended by WHO (15 μg m⁻³). In all examined periods (pre-during, post) average daily concentration exceeded WHO recommendations for 27, 27 and 28 days per 31 days of measurements.

Although the average concentration of PM₁₀, exceeded the level recommended by WHO (45 μg m⁻³) only during the pre-COVID period [30] the number of days in each period with WHO guidelines exceedances was 14, 14 and 8 for pre-during and post-COVID-19 periods, respectively. PM_2.5 in each period is significantly (p<0.05) correlated with PM₁₀, with Spearman correlation coefficient R_S>0.9 (Table 3). PM_2.5 and NO₂ are highly correlated at pre- and post-COVID periods (R_S>0.8), while with SO₂ the correlation was the highest at the pre-COVID period (R_S=0.9) and statistically significant but lower after- and during-COVID (R_S=0.5). The high correlation between PM_2.5 and NO₂ is not unusual, because both are characteristic pollutants for automobile traffic, additionally, PM_2.5 and NO₂ are derived from combustion processes, so the point sources have to be included. Moreover, 14% to 27% of measured secondary PM_2.5 is generated from NO_x set reactions and NO₂ is often treated as a surrogate for PM_2.5 [31, 32, 33].

According to the data of the Chief Inspectorate of Environmental Protection, the main source of air pollution in the Silesian Voivodeship is anthropogenic emissions from the municipal and household sectors (surface emissions), a smaller share is accounted for by emissions from transport (linear emissions) and industrial activity (point emissions). A noticeable share in the air concentrations of pollutants in the province is their inflow from the rest of Poland and Europe. The main local sources of pollution are chimneys of individually heated houses, while automobile transport influences pollutant concentrations, especially in areas directly adjacent to roads with heavy traffic. Industry located in the Silesian province, mainly commercial power generation, due to the high height of chimneys, largely exports pollution outside the province. Industrial plants with significant fugitive emissions or those emitted through low emitters can also directly affect air quality in the neighbourhood [24]. In the Upper Silesian agglomeration, especially in the large cities of the Silesian Voivodeship, a significant share of total emissions comes from fuel combustion, and in road areas, also from vehicle traffic. Traffic pollutants in the form of dust are mainly generated by the abrasion of brakes, tires and road surfaces, as well as the lift of pollutants from road surfaces, while nitrogen oxides are emitted from tailpipes. Figure 5 shows the distribution of emission balances for selected pollutants in the area of the Silesian province and the sources of emissions. The summaries were prepared by the Chief Inspectorate of Environmental Protection based on data provided by the National Center for Balancing and Emission Management (KOBIZE), operating within the structures of the Institute of Environmental Protection – National Research Institute (IOŚ-PIB). The emission inventory was performed, among other things, for the purpose of mathematical modelling of pollutant concentration distributions [24]. During the COVID-19 period, the structure of emissions changed. Point source emissions of PM, i.e. from industrial sources, declined markedly, while emissions from municipal and household sources increased.

Figure 5.

In our study, we observed a reduction of PM_2.5 at the level of 27.8%, and PM₁₀ at 12.7% during COVID-phase compared to the same period of the previous years (2018/2019). Larger decreases were obtained in different Polish cities, e.g. in Gdansk (Northern Poland) where during April and March 2020 PM₁₀ concentrations were reduced by 31.5% and 33.9%, while in Wroclaw (Southwestern Poland) by 20.1% and 26.9%, respectively as compared to pre-COVID-19 phase [4].

Cleaner air during COVID-19 in other Polish cities in comparison to Gliwice is caused by two significant factors. The first is meteorological conditions, and the second local emissions influencing background concentration. The climatogenic factor that most significantly shapes meteorological conditions is atmospheric circulation. An analysis of the atmospheric circulation index in Poland, carried out by the Institute of Meteorology and Water Management – National Research Institute (IMGW-PIB), indicates that in the northern and central parts of the country air masses coming from the west (W to SW sectors) dominate, while in the southern part of Poland from the southwest direction (SSW sector). The average annual wind vector from the 1991–2020 multi-year period indicates wind speeds in the 10–15 km/h range in the northern part of the country, while in the western part in the 5–10 km/h range. Figure 6 presents a wind rose diagram for March 2018–2023 derived from a mobile air quality lab located in Gliwice. The results of the study indicated that the wind in the studied area dominantly blew towards the southwest, south and southeast parts, and although the wind speed values were low, in a range from minimum of 0.42 to a maximum of 3.34 km h⁻¹ (Table 1) the difference between pre-, during and post-COVID-19 periods is statistically significant.

Figure 6.

The second, is lower concentrations of air pollutants. For example, in Gdansk during March 2018 and 2019, the average concentrations of PM_2.5 and PM₁₀ were 21.9 and 19.9 μg m⁻³, respectively, so approx. 50% lower than in Gliwice. In more polluted Wroclaw the concentrations of PM_2.5 and PM₁₀ were approx. 20% lower than in Gliwice. The significant difference between regions highlights the importance of both: meteorological conditions and pollution sources affecting analyzed periods.

During the lockdown in India PM concentrations were at the lowest recorded levels for the previous 20 years, and PM₁₀ and PM_2.5 were reduced by more than 50% compared to the pre-lockdown phase [34–35]. Due to the lockdown in 44 cities in China, the concentration of PM_2.5 and PM₁₀ decreased on average by 5.9% and 13.7%, respectively [36]. In Kazakhstan, PM_2.5 was reduced by 21% during the lockdown period compared to the average concentration in the pre-COVID-19 phase (2018–2019) [37]. Salé City, Morocco obtained a 75% reduction for PM₁₀ [38]. In the urban areas of Malaysia, during the period of lockdown, the concentrations of PM₁₀ and PM_2.5 were reduced by 26–31% and 23–32%, respectively, as compared to the corresponding periods in 2018 and 2019 [39].

NO₂ is formed as a result of nitrogen oxide (NO) oxidation in the atmosphere and is particularly dangerous for human health [14]. The toxicological evidence suggests that increased susceptibility to infection, functional deficits from effects on airways, and deterioration of the status of persons with chronic respiratory conditions, including asthmatics, are of potential concern [40]. In Poland, the highest air concentrations of NO₂ are observed in large cities (e.g. Cracow, Warsaw) as compared to other cities of the country, which is associated with high motor-vehicle traffic [4].

Due to lockdown actions, the European Environment Agency (EEA) has reported about 47% and 55% drops in NO₂ concentrations from the cities of Bergamo (Italy) and Barcelona (Spain), respectively [41]. Similarly, the United Kingdom observed a 92% reduction in NO₂ concentrations due to the lockdown phase compared to the pre-COVID-19 period [42]. In Gliwice, we observed that NO₂ concentrations (Fig. 7) were 10% lower during the lockdown (21.76 μg m⁻³) than in the pre-COVID-19 phase (24.05 μg m⁻³). In the corresponding period, a similar relative change of NO₂ concentrations between pre- and during- COVID period was observed in Wroclaw, 20% decrease in NO₂ concentrations was observed in larger cities (Cracow, Warsaw). It was associated with reduction of city traffic and the quantity of vehicles on the roads especially in larger cities. For example, in late March 2018, the General Directorate for National Roads and Motorways reported a reduction of motor-vehicles on the roads by 25–54% in comparsion to March 2019. Apple, based on the number of routing queries in Apple Maps during the COVID-19 pandemic, showed even higher reduction of transport activity in Polish cities. Pointing to more than 60% of reduction in vehicle transport, and more than 65% of walks [4]. In regions of Poland with smaller population densities, changes in NO₂ concentration were not as significant. For example, in Gdansk NO₂ concentration was 4% higher during the COVID-19 period in compare to before the COVID-19 period, however in both periods NO₂ concentrations were at the low level of approx. 16 μg m⁻³.

Figure 7.

Nevertheless, in spite of the improvement of the air quality in Gliwice during the lockdown period, it had a short-term effect. Since the gradual increase of human activity after the lockdown, there was a 27.8% increase in NO₂ ambient levels (27.82 μg m⁻³). NO₂ is highly correlated with all air pollutants during the pre-COVID period (R_S>0.8) while in the post-COVID-19 period, the correlation is statistically significant with SO₂ and BA.

Elevated SO₂ concentrations can have impacts on human health, climate, and ecosystems [23]. SO₂ contamination increases the risk of premature, chronic obstructive pulmonary disease (COPD) and acute diseases of the upper respiratory tract [43–44]. Although in Europe SO₂ concentrations are well below the EU limit values of 125 μg m⁻³ for an average period of 24h, the exceedance of the WHO daily mean guideline of 20 μg m⁻³ for the protection of human health persists [45]. The highest daily average concentrations monitored in Upper Silesia voivodeship were from 23 to 84 μg m⁻³ [23].

In Gliwice, we observed that SO₂ concentrations were significantly (p<0.05) reduced during the lock-down phase (Table 2), which is associated with a reduction of industrial activity. The highest average concentration we noted in the pre-COVID-19 phase (14.57 μg m⁻³) whereas during the lockdown the concentration of SO₂ decreased by 34.0% (9.62 μg m⁻³) (Fig. 6). Studies in selected Polish cities by Filonchyk et al. [4] point that in cities with heavy industry (Lodz and Cracow) during the lockdown phase the concentrations of SO₂ decreased on the comparable to Gliwice level (33 and 40%, respectively). However, in cities without heavy industry (Gdansk, Warsaw, Wroclaw) during the lockdown phase concentration of SO₂ increased (but remained at the low level of approx. 5 μg m⁻³), as compared to the previous years (2018/2019) with concentrations between 3 to 4 μg m⁻³. The observed increase in those cities may be related to the fact that some manufacturing facilities continued working during the quarantine period. On the other hand, the atmospheric chemistry of SO₂ shouldn't be neglected. SO₂ is highly soluble in water and air, with a residence time in the atmosphere of about two weeks. Therefore, SO₂ contamination is regionally significant, as gas clouds may be transported on longer distances, indicating the pollution sources may influence other regions according to environmental and meteorological factors. SO₂ is significantly correlated with PM₁₀, the correlation coefficients R_S are 0.9, 0.7, and 0.6 for pre-, during- and past-COVID periods, respectively.

Similar results to those obtained in Gliwice were noticed in Morocco, where in lockdown phase SO₂ levels were reduced by 49% [38], and in Vietnam where the reduction of SO₂ was at the level of 60% compared to 2018/2019 [46]. In turn, in the urban areas of Malaysia, the concentrations of SO₂ were reduced by 9–20% as compared to periods before COVID-19 (2018/2019) [39]. Bao and Zhang (2020) showed that the concentration of SO₂ decreased by 6.76% is an average of 44 cities in China during the quarantine period [36].

To our knowledge, no research has previously been carried out to evaluate and compare the outdoor air quality of BA in a different phase of the COVID-19 pandemic. We observed that the average concentration of BA significantly (p<0.05) decreased during the lockdown (602 CFU m⁻³) compared to the pre-COVID-19 phase (782 CFU m⁻³), but BA levels were higher during-COVID in comparison to the results from the post-COVID-19 phase (529 CFU m⁻³) (Fig. 9). BA is highly correlated with PM₁₀ (R_S = 0.8–0.9). The correlation with PM_2.5 is also significant but lower (R_S = 0.6–0.8), and the lowest with gaseous pollutants R_S = 0.3–0.6.

Figure 8.

Figure 9.

The question of whether the significant correlation between air pollutants and bacterial aerosol could be expected is open to debate. In fact, recently, researchers in Poland and Austria found that the concentrations of mesophilic bacteria are correlated with all particle fractions, especially with coarse particles [47–48]. Existing literature suggests that O₃, SO₂, NO₂ and CO could influence air-borne bacterial growth or their survival in the atmosphere. For example, Dong et al. [13] found that the concentration of microbes was positively related to SO₂ and NO₂. Zhen et al. [49] detected a significant positive correlation between airborne bacterial abundance and PM_2.5 concentration but only in the spring season, which was consistent with the hypotheses that particles in spring are more inclined to be derived from “natural sources” because a dry windy spring creates dust. On the other hand, Chi and Li [50] obtained only weak correlations between bioaerosol concentrations and air pollutants, including particulate matter.

In the analysis of the relationship between air pollutants and bioaerosols, the results of Zhen et al. [49] study should be considered. Zhen et al. [49] indicated that meteorological factors may play more important roles in shaping bacterial communities than air pollutants. Sampling outdoors presents its own set of challenges as temperature, relative humidity, atmospheric pressure, wind speed and direction may affect a sampler's aspiration and collection efficiencies [51]. The increased temperature would promote bacterial growth, and speed up convective air movements which increase bacterial dispersal in the atmosphere [52]. The barometric pressure increase could mirror the entry of cold air flow and an improvement in air diffusion conditions. Cold air may facilitate the release and transportation of bacteria, which might lead to a high bacterial abundance in the atmosphere. Cold air could also quickly dilute air pollution and bacterial concentrations, which would lead to a low bacterial abundance in the atmosphere. The increase in air humidity would prevent dust rising from wet surfaces and thus reduce bacterial entrance into the atmosphere. Higher wind strength could raise dust from various surfaces, aerosolizing bacteria from local sources. A strong wind could also bring in exogenous bacteria or dilute the concentration of local bacteria. Therefore, wind may increase or decrease airborne bacterial abundance, and this might be the reason for the small correlation between RH and bacterial abundance in spring [53].

This study showed significant positive correlations between gas or PM pollutants and BA as well as not significant correlations with meteorological parameters. However, a significantly correlated pollutant does not always mean that it is the dominant factor. The results could also have been caused through interference from other factors, e.g., collinearity with a more dominant factor. In this study, some air pollutants were significantly correlated with the dominant meteorological factors (Table 3), and these may cause joint effects. However, we cannot separate the contributions made by meteorological factors and air pollutants from their joint effects. The joint effects might be caused by meteorological factors and air pollutants only co-varied with meteorological factors. Alternatively, the opposite situation, whereby air pollutants were the determinants, but meteorological factors co-varied, might be true. The joint effects may also have been caused by interactions between some meteorological factors and air pollutants; or by a third class of factors that were not investigated, but upon which meteorological factors and air pollutants are dependent. Therefore, to further understand the relationship between airborne bacteria and air pollutants, it is necessary to investigate different categories of environmental factors with different spatial and temporal scales as well as the pollutant's sources, rather than the concentration.