Spatial and Temporal Analysis of Daily Measurement Big Data on PM2.5 Air Pollution in Beijing, China

Keywords: PM2.5 Air Pollution, Spatial and Temporal Patterns, Spatial Statistical Analysis
Abstract Type: Virtual Paper Abstract

Abstract

Five-year hourly measured PM2.5 air pollution concentrations from 2014 to 2018 was obtained from the Beijing Municipal Environmental Monitoring Center. Datasets of 27 observation stations across the municipality were used. Both linear regressions based fundamental data analysis and principal component analysis (PCA) were conducted for temporal variations and temporal patterns of five-year seasonal PM2.5 pollution concentrations at each observation station. Geographic Information Systems (GIS) was utilized to analyze the spatial pattern of air pollution distribution during entire time period. Spatial statistical IDW models were applied to interpolate the atlas of results on 5-year annual average spatial distribution patterns and on average distribution patterns of different seasonal sample periods. The results are: 1) PM2.5 pollutions are most severe in winter in each of the natural years. 2) PM2.5 pollution concentrations in Beijing were gradually decreasing from 2014 to 2018. 3) Improvements of air quality and reduction of PM2.5 pollution appeared in all the seasons during five-year time. 4) PM2.5 pollution concentrations in summer are significantly less than that in other seasons. 5) The least PM2.5 pollutant concentrations occur in north and northwest regions, and the most PM2.5 pollutant concentrations occur in south and southeast regions. 6) Vehicle traffic congestions are not the significant contributing factor of PM2.5 pollution. 7) Heating supply of buildings and houses generated great contributions to the PM2.5 pollution concentrations. 8) Human activity contributions are limited, while type and quantity of coal or fossil fuel consumptions might contribute a large amount of PM2.5 air pollutions in Beijing.