Analysis of Changes and Statistical Characteristics in PM-2.5 based on the Seasonal Management Data

Abstract

Objectives : This study aims to identify the changes and statistical characteristics of the observed concentration data of fine particles (PM-2.5), one of the air environment indicators, and to analyze the variability of observation data in South Korea based on the statistical significance of the observation data.Methods : We analyzed the changes and characteristics of daily average PM-2.5 observations based on descriptive statistics (the basic statistical and frequency analysis), and inference statistics (analysis of variance and post-hoc analysis) from December to March for the period 2015 to 2021.Results and Discussion : Through ANOVA analysis and post-hoc analysis, it was confirmed that there was a statistically significant difference in the observed PM-2.5 as of the winter of 2019. Before 2019, the frequency of observed data greater 30 µg/m;3 was relatively more than after 2019, and on the contrary, the frequency of observed data less than 20 µg/m;3 was smaller than after 2019. After 2019, the frequency of high-concentration observations greater than 50 µg/m;3 was insignificant, and on the contrary, the frequency of observations less than 20 µg/m;3 was more than before 2019. As a result, It’s an average difference of 7.5 µg/m;3 between before and after 2019, and it was reduced by about 24% compared to the average before 2019. Meanwhile, as a result of examining the weather influence (wind direction and wind speed) to analyze the causes of the observed data differences before and after 2019, no significant differences were identified between the two periods due to the influence of meteorological conditions.Conclusion : The observed concentration of PM-2.5 data set in South Korea showed variability as of the winter season of 2019 at the significance level of 5%. The variability could be more attributed to anthropogenic activities and socio-environmental changes such as COVID-19 rather than natural environmental factors such as meteorological factors. In order to more effectively manage air quality, it is necessary to identify social changes and their flows, and at the same time conduct continuous and systematic research such as data expansion and analysis technique development.