An assessment of PM₁ levels based on indicative PM₁ measurements and relationships with PM₁₀ and PM_2.5 concentrations, for the analysis of hospital admissions and mortality in the Moravian region

Particulate matter (PM) air pollution is a serious concern in the city of Ostrava. Thus, in 2018, a project entitled “Validation of the relationships between PM10, PM2.5 and PM1 concentrations, and morbidity and mortality, in the heavily polluted region in the Czech Republic,” was launched. The relationship between hospital admissions and mortality in the said region is based primarily on short-term PM10 and PM2.5 concentrations and indicative PM1 measurement. The analysis of spatiotemporal variations and the relationship between PM10, PM2.5 and PM1 data from 3 measurement sites within the city of Ostrava is presented.Material and MethodsThe analysis was based on the daily average PM concentrations for 5 and 6 months at 2 sites, and on the annual average values (2018–2019) at the baseline station. The correlations of and variability between PM fractions, seasonal differences and explanation of the differences found were the objectives of a detailed analysis. Especially, the potential PM1 variability and its causes were analyzed with respect to the location of the site.ResultsThe study findings confirmed good correlations between the PM fractions. Compared to PM10, PM2.5 concentrations were more predictive for PM1 concentrations. The annual means of PM10, PM2.5 and PM1 reached 37.5, 29.9 and 27.1 μg/m3 in 2018, respectively, and 25.8, 19.9 and 17.9 μg/m3 in 2019, respectively. The concentration levels in the non-heating season were significantly lower than in the heating season in the 2 years under consideration. The levels of PM10, PM2.5 and PM1 were significantly correlated (the correlation coefficient, r > 0.96). The levels of PM2.5 represented about 0.82–0.86 of PM10, and the levels of PM1 about 0.92–0.93 of PM2.5. These ratios were found to differ in the heating and non-heating seasons, with the PM2.5–PM10 ratio ranging 0.61–0.63 in the non-heating seasons.ConclusionsThe correlations found will be used for indicative PM1 measurements in other areas of the region. Seasonal variability should be taken into account as well. Med Pr. 2021;72(3):249–5