Assessing Air Quality Using Multivariate Statistical Approaches

The purpose of the current study was to evaluate air quality in Dong Thap province, Vietnam. The air quality data was collected during 2019–2020, representing the time of pre- and mid-COVID-19. Twenty-seven air quality samples (in the areas of urban, residential-administrative, hospital-schools, and industry-craft village areas) were used for the evaluation. Air quality was evaluated using national technical regulations on air quality, including QCVN 26:2010/BTNMT and QCVN 05:2013/BTNMT. The difference of mean air quality between the areas was examined using a one-way ANOVA followed by the Duncan test at a significant level of 5%. The relationship between air quality parameters and microclimate factors was tested using Pearson correlation. Principal component analysis (PCA) was utilized to identify critical variables and potential sources of air variation. Cluster analysis (CA) was applied to group similar air quality sites, thus recommending air monitoring site selection. The results show that the air quality in the study area is not polluted. The concentrations of noise, TSP, SO2, and NO2in the mid-COVID-19 pandemic were significantly lower than those in the pre-COVID-19 pandemic due to the social distancing policy. There was a close correlation among air quality parameters, except for air humidity. PCA identified two to four potential sources of air variation, explaining 84.3%, 100%, 100% and 89.7% of the total air quality variance at urban, residential–administrative, hospital-schools, and industry-craft villages, respectively. CA divided the 27 sampling sites into eight groups by the differences, mainly in humidity, wind speed noise, TSP, and CO. Eight sampling sites could be potentially reduced from the current monitoring program for representativeness and cost-effectiveness purposes. All air parameters in the current study are significant for monitoring, and the potential sources of air quality variation are traffic activities, industrial production, craft village activities, and daily life using fuels in residential areas. The results of the current study provide useful information for air quality monitoring and management. Future monitoring programs should include toxic air pollutants in air quality monitoring programs. Doi: 10.28991/CEJ-2024-010-02-012 Full Text: PDF