3.1. Basic Statistical Characteristics of the Measuring Series
The following are tables with basic statistical characteristics for health resorts such as Ciechocinek, Rabka Zdrój, Sopot, and Ustroń in 2015–2019. The analyzed meteorological parameters are: average air temperature, relative humidity, speed, and concentrations of air pollutants (PM10, SO2, NO2,) and benzo(a)pyrene content in PM10 suspended particulate matter recorded in 2015–2019.
According to the Minister’s Regulation of 24 August 2012 on the levels of certain substances in the air, during the calendar year, the level of PM
10 particulate matter should not exceed the standard of 40 μg/m
3 [
28]. On the basis of
Table 2, it was shown that in Ciechocinek in 2015–2019, the average PM
10 particulate matter level standard was not exceeded. In recent years, due to unsatisfactory air quality, air protection programs have been carried out in Poland. In many cities such as Ciechocinek, the practical implementation of these programs results in lower levels of harmful pollutants. The aim of the establishment of air protection programs is to improve air quality and to comply with the standards set out in the Regulation in areas where exceedances occur. The practical implementation of air protection programs results in a reduction in the levels of harmful pollutants. A favorable trend has been noted, as from year to year, the average level of harmful pollutants gradually decreases. The same has been shown in Rabka Zdrój, where average annual concentrations of particulate matter have been reduced from 32.98 μg/m
3 in 2015 to 25.01 μg/m
3 in 2019. However, particularly untapped situations are in Sopot and Ustroń, where average annual particulate matter pollution is at a similar level and not decreasing. Previous studies have shown that in conditions of long-term effects, there is most likely no safe level of air pollution below which adverse health effects are no longer observed [
29]. The highest concentrations of SO
2 (
Table 3) were observed in 2017 and 2018 in the health of Rabka Zdrój. Although the exceedance levels were insignificant, the air quality in the health resorts of Rabka Zdrój compared to Ciechocinek or Ustroń looks untapped. For example, in 2018, the concentration of SO
2 air pollution in Ciechocinek (2.63 μg/m
3) was 8 times lower than in Rabka Zdrój (20.46 μg/m
3). The most favorable situation was observed in the health resorts of Sopot, where the average annual concentrations of SO
2 in 2015–2019 did not exceed 2.00 μg/m
3, and a further decrease in these values was observed. Winds by the sea blow more often and harder than in other regions Polish, which effectively contributes to the reduction of concentrations of harmful pollutants. Analyses of the average annual concentration values for NO
2 sulfur dioxide (
Table 4) indicate that, within five years, the annual standard value has not been exceeded which, in accordance with the Regulation of the Minister of 24 August 2012, is 40 μg/m
3. The most unfavorable situation—the highest annual average concentration occurred in Ustroń (
Table 4) in 2017, followed by a sudden decrease in 2018 and gradually decreasing from 2019. B
(a)P in PM
10 was found to decrease level in 2015–2019. Despite the steady decrease in B
(a)P concentrations in the air since the 1990s, exceedances of its permitted level (1 ng/m
3) [
28] are still recorded in the predominant area of the country. In 2005, under Directive 2004/107/EC of the European Parliament and of the Council of 15 December 2004 on arsenic, cadmium, mercury, nickel, and polycyclic aromatic hydrocarbons, an obligation was introduced in Poland to measure the benzo(a)pyrene content in particulate matter [
30]. From the above data (
Table 5), it appears that environmental pollution by PAHs compounds in Poland may pose a serious health risk. It therefore appears necessary to take all measures to reduce the exposure of the population to these compounds. According to data from the European Environment Agency, more than 46,000 people die every year from poisoned air in Poland. Achieving the required levels of reduction of particulate matter and benzo(a)pyrene emissions from the municipal and living sectors, at the current rate of action, can take 24 to almost 100 years on a per-state scale [
31].
The following are linear figures that show the average air temperature, relative humidity, wind speed, and visibility in 2015–2019 in four health resorts analyzed, such as Ciechocinek, Sopot, Rabka Zdrój, and Ustroń.
In 2015–2019, the increase in annual air temperature occurred in all health resorts. In the health resorts of Sopot, the average air temperature rises by as much as 0.95 °C in five years. Sopot is located beneath the surface of the ocean air flow from the west, and their impact from the east is low. The proximity of the sea, and especially the depths of Gdansk, influenced the thermal and humid phenomena of the Middle Ages, as well as the diverse geographical environment, especially the diverse topography. The lowest average annual temperatures were recorded in Rabka Zdrój. In 2015, the average annual currency price was 7.24 °C (
Figure 2), and as much as 7.50 °C in 2019. In the years 2015–2019, the air temperature is 0.26 °C. Climate change is important for mountainous areas. Mountain ecosystems are highly vulnerable and vulnerable to climate change [
32]. The highest average temperature in 2015–2019 in Ciechocinek is 10.53 °C. In the health resorts of Ustroń, in 2015, the average annual temperature was 10.26 °C, in 2016 and 2017, the temperature dropped by about 1 C, and in 2018, there was an increase in the average annual air temperature.
The highest average wind speed(
Figure 3) of the five years analyzed occurs in the health resorts of Ustroń and is 10.84 km/h, then in Sopot where the average wind speed in the health resorts during the period considered was 9.92 km/h consecutively, in Ciechocinek, 8.67 km/h and Rabka Zdrój, 5.54 km/h. Wind speed has a stimulating effect on mixing processes, and in addition, in the case of large urban agglomerations, which are a cluster of age emitters, it has an impact on the displacement of pollutants outside the city. Therefore, there is usually a decrease in the height of concentrations of pollutants with an increase in wind speed [
33].
The highest average relative humidity (
Figure 4) of 2015–2019 during the analysis period is 77.30% and occurs in Sopot. In Rabka Zdrój the average relative humidity is 76.61%, in Ciechocinek, 75.64%, In the health resort of Ustroń the average relative air humidity is 73.50% The development of basic statistics on meteorological elements is very important in order to determine their impact on concentration levels of harmful air pollutants. This is confirmed by numerous research works carried out by other researchers [
34,
35]. The relative humidity and air temperature [
34] have the greatest impact on the con-centration of sulfur dioxide and particulate matter among meteorological elements. The level of particulate matter pollution of the atmosphere largely depends on meteor-ological conditions, primarily the direction and speed of wind, relative humidity of air, the intensity of solar radiation [
35].
The average visibility during the period considered in the health resorts of Ciechocinek was 20.96 km, and in Rabka Zdrój, 20.35 km (
Figure 5). In 2018, these health resorts saw a positive growing trend of visibility, which may indicate a decrease in concentrations of harmful air pollutants. A sudden decrease in visibility was noticed in Sopot; in 2017, visibility was 17.92 km, and in 2018, 8.60 km. Such a rapid change in the range of visibility is a very unfavorable situation, which requires determining the reason for such sudden changes, identifying the source of the problem in order to eliminate it. In the health resorts of Ustroń, the average annual visibility in 2015 is 9.82 km, in 2016 and 2017, there is a gradual decrease in the range of visibility, and in 2018, you can see a slow beginning of a growing trend, which can testify, among other things, to the reduction of levels of harmful pollutants. Visibility is a very complex issue—it is directly due to the level of anthropogenic air pollution, but it is also shaped by meteorological conditions. The impact of anthropogenic air pollution on human health and visibility has been studied for decades. A number of studies have been carried out, not only to assess the benefits to human health of reducing air pollutant emissions [
36], but also to understand how air pollution negatively affects visibility. In general, visibility is a good indicator of the degree of air pollution, and can also be used as a substitute for an assessment of the impact on human health [
37].
3.2. Analysis of the Occurrence of Episodes of Elevated PM10 Concentrations in Health Resorts in Poland
Table 6, attached below, shows the thresholds for exceeding the limit values for concentrations of pollutants. Within 24 h, the overrun value in accordance with the Regulation may not exceed 50 μg/m
3, and the allowable frequency of exceeding the allowable level in a calendar year shall be 35 days [
28]. The second threshold > 100 μg/m
3 is an alert threshold that calls on state institutions to inform the public about an episode of elevated concentrations of pollutants.
Episodes of elevated concentrations are an atmospheric phenomenon resulting from the primary emission of particulate matter and gaseous pollutants into the air and the formation of secondary particulate matter as a result of chemical reactions occurring in the atmosphere, under meteorological conditions conducive to the accumulation of pollutants. Children, the elderly, and people already suffering from respiratory or blood diseases are particularly vulnerable to diseases associated with exposure to harmful airborne agents. Inhalation of air with a high content of harmful substances causes health problems, and in an increasing number of cases in the long term even contributes to death. The occurrence of episodes of high concentrations of particulate matter results in an increase in disease symptoms, mainly respiratory and vascular-blood diseases [
1]. An episode of elevated concentrations of pollutants is a situation of one- or several days of elevated concentrations of PM
10 particulate matter, in which there is a significant exceedance of the daily limit for PM
10 (50 μg/m
3). The most common episodes of elevated concentrations of PM
10 particulate matter occur in Rabka Zdrój, a mountain health resorts located in the south of the Lesser Poland province. In accordance with the Regulation of the Minister of 24 August 2012 on the levels of certain substances in the air, the permissible frequency of exceeding the limit in a calendar year should not exceed 35 days. In 2015, 67 days out of 365, in 2016—39 days, in 2017—41, and in 2018—51 days is characterized by exceeding the average value of impurities for PM
10 particulate matter (
Table 6). In 2016 (15 days) and 2017 (14 days), there were numerous exceedances above 100 μg/m
3. In 2018, the analysis showed a decrease in the frequency of crossings, below the standard set out in the Regulation. Short-term episodes of elevated concentrations of pollutants are mainly caused by low emissions. The problem of low emissions affects many communities, negatively affecting health. At the same time, this factor generates an increase in external costs for the health care of people with respiratory diseases. The increased impact of air pollution can be observed in urban areas, where increasing and compact housing makes it difficult to exchange air in cities and exacerbates the phenomenon of smog. Locally, this is also associated with bad weather and topographic conditions of cities [
11]. Air pollution also affects flora, fauna, and cultural heritage. Monuments located in contaminated areas are destroyed and degraded at a much faster rate. A side effect of low emissions is acid rains, which cause, i. a., acidification of the soil, disruption of reproduction of living organisms, as well as damage to needles and leaves of trees [
38]. However, it is promising to reduce the frequency of exceeding the daily limit value. In 2019, 27 days were characterized by exceeding the daily average value and let us say that this declining trend will continue. The situation is also unfavorable in Ciechocinek, a health resort located in central Poland. In 2015, the concentration levels of harmful pollutants exceeded the daily norm in 42 days, and 38 days in 2018. In 2017, although 24-h concentrations of particulate matter were exceeded 20 times as many as 7 days out of 20, the levels of harmful pollutants were more than 100 μg/m
3. In Ustroń, the highest number of days during which values exceeded the concentrations of pollutants allowed in the Regulation is 26 in 2018. In 2019, only 5 days were found in which levels of harmful pollutants exceeded the threshold of 50 μg/m
3. The best of all analyzed health resort is in the seaside town of Sopot. In 2015–2019, the permissible frequency of exceeding the permissible level did not exceed the 35-day standard, during which the achievable level was 50 μg/m
3.
3.3. Analysis of the Correlation between Harmful Air Pollutants Together with Benzo(a)pyrene and Meteorological Conditions
We calculated Pearson’s correlation coefficient between meteorological conditions and air quality parameters: PM
10, SO
2, NO
2, and B
(a)P in PM
10 [μg/m
3]. The correlation was determined between daily average values of the parameter. All the analyses were performed at the significance level
p < 0.05.
Table 7 shows all the data to demonstrate the significant as well as non-significant correlation of the variables. The red digits in the tables indicate the statistical significance of the correlation at 0.05.
Table 7 of the correlation clearly indicates the relationship of average air temperature (T
av) to levels of harmful pollutants such as PM
10, NO
2, SO
2, and to the B
(a)P content of PM
10. Correlation analyses confirm the influence of air temperature on the formation of concentrations of pollutants. Kobus and his co-authors came to similar conclusions in 2020. When comparing the correlation between average daily PM
10 concentrations and the daily values of selected meteorological indicators in the health resorts and agglomerations studied in 2019, the highest negative correlation with the minimum daily temperature is clearly observed. It is the strongest in the areas of health resorts and agglomerations located in the southern part of the country [
22]. The strongest air temperature correlates with air pollution in Rabka Zdrój, which shows the lowest average annual air temperatures (
Figure 2). In Ciechocinek, where air temperatures reached their highest values, it has been proven that as air temperatures rise, the impact on pollution decreases. With each of the municipalities with health resorts status, the impact of relative air humidity on air quality has been proven. The analysis showed the greatest effect on B
(a)P levels in PM
10 has relative air humidity. In addition, the increase in the amount of PAHs in a given urban area is facilitated by an increase in relative humidity and strong winds blowing from directions where transport routes with significant vehicle traffic intensity are located [
39] Low wind speeds increase levels of harmful pollutants. According to a 2013 Zaho study, wind speed was a major meteorological factor affecting both the visibility and concentration of particle mass. Low wind speeds combined with temperature inversions indicated stable meteorological conditions during the pollution process, which reduced the dispersion of pollutants and may have caused higher PM concentrations and lower visibility [
40]. A negative correlation was noted between the wind speed in Sopot (
Figure 3) and the level of harmful particulate matter PM
10 and SO
2. A negative correlation was observed in the case wind speed, when the conditions of dispersion of pollutants are improved [
41]. Based on the analysis, visibility has been shown to be strongly correlated with air pollution. Emissions can cause impaired visibility, which makes visibility an important indicator of particulate matter pollution [
42,
43]. Therefore, the long-term trend in visibility may indicate a change in the state of air pollution [
44,
45]. Due to the rapid development of in particulate matter and economic growth, people have incurred huge environmental costs for serious air pollution problems that can seriously harm public health [
46]. In Sopot, on the basis of
Table 2,
Table 3,
Table 4 and
Table 5, the lowest concentrations of impurities were shown, compared to the rest of the health resorts studied. The correlation analysis shows that we cannot reject hypothesis that air pollutants affect visibility at
. In Sopot, where air quality was best and episodes of elevated concentrations of pollutants were the rarest (
Table 6), correlations despite the fact that they showed relevance are the weakest among the health resorts analyzed.
3.4. Regression Analysis
We decided to propose multilinear regression model to determine the visibility as a linear combination of the meteorological and air quality data.
Table 8 presents the basic statistical characteristics of the fitted models together with regression equations for health resorts: Ciechocinek, Rabka Zdrój, Ustroń, and Sopot.
All the meteorological parameters were tested for the normality according to [
26]. Eight independent variables were used for regression analysis. All of the analyzed variables were statistically significant (
p < 0.05). The analysis showed that the multilinear model fits the best the data collected in Ciechocinek (the Pearson coefficient of correlation is R = 0.838). The result obtained indicates the significant relationship between visibility and meteorological parameters and concentrations of harmful pollutants in Ciechocinek. Ustroń found a relationship at R = 0.404, in Rabka Zdrój had R = 0.707, and Sopot, R = 0.428. Step-by-step regression analysis allows us, at the initial analysis stage, to remove from the model those variables that are not affected by the model form, facilitating further analysis. Five of the eight variables got into the model. From harmful concentrations of pollutants, the main influence on the form of the model in Ciechocinek was a relative humidity of air (0.496
Table 8) and level B
(a)P (−0.317). Among the health resorts analyzed, B
(a)P, only in Ciechocinek, had a limiting effect on visibility, and nowhere else appeared in the analyses. In the health resorts of Ustroń, visibility is influenced by only two variables such as average air temperature and relative humidity. In Rabka Zdrój, the visibility of four variables out of the eight analyzed was found to affect the visibility. Visibility is affected by PM
10 harmful pollutants
, average air temperature, relative air humidity, and wind speed. In the other stations where the data were analyzed, a link was also found between the level of harmful PM
10 pollutants and visibility, but it is not that large. The developed model for Rabka Zdrój shows a positive relationship of wind speed on horizontal visibility, increasing it through more efficient circulation of polluted air. In Sopot, in the seaside health resorts, SO
2 and NO
2 are the most affected by visibility. The number of cases, i.e., days where the data met the requirements for analysis, is 1681 days for Ciechocinek, 1339 days for Ustroń, 1781 days for Rabka Zdrój, and 1582 days in Sopot. The more days–number of cases, the more reliable the model form you get. Humidity has a negative impact on the form of the model, but with regard to the model form of the other stations, it is negligible. The best-fit empirical regression models have shown a negative impact on the visibility of PM
10, SO
2, NO
2, and in particular PM
10 [
47]. Statistical analysis showed that regression models were well matched to the observed data. Regression analysis confirmed the hypothesis that there is a relationship between visibility and concentration levels, as well as meteorological parameters.