Intra-community air quality monitoring in various urban microenvironments in South Korea: based on observations from highly dense cost-effective sensor network

 

As urbanization has spread, increased energy consumption, complicated built environments, and dense road networks cause spatiotemporal heterogeneity of air pollutant distributions even in an intra-community scale. High spatiotemporal heterogeneity of air pollutant distributions can affect pedestrian and/or traffic users’ exposure to air pollutants according to where and when they are, potentially forming air pollution hotspots. Thus, it is important to understand the characteristics of spatiotemporal distributions in air pollutants in various micro-built environments in populated urban areas. However, current air quality monitoring performed by the government cannot capture these highly heterogeneous distributions of air pollutants due to the limitations of financial and human resources. In this respect, cost-effective sensors have great potential to build highly spatially dense air quality monitoring networks to address the low spatial resolution issue of conventional air quality monitoring stations.

In this study, we built a highly dense air quality monitoring network consisting of 30 sets of sensor nodes in an 800 m ´ 800 m spatial domain to understand the characteristics of air pollutant distributions in various urban microenvironments. The domain includes urban street canyon with moderate traffic, a mixture of high and low buildings with high traffic, an open space with minimal traffic, and others. The sensor node consists of sensors (for CO, NO₂, O₃, PM_2.5, and PM₁₀, temperature, and humidity) and communication/data storage parts (wifi, interface for smartphone connection, and SD card). We also conducted inter-sensor comparison among sensor nodes and intercomparison tests between the sensor node and conventional reference instruments.

Intra-community air quality monitoring with a sensor network was conducted for a couple of weeks in two distinct weather conditions (humid and hot summer and dry and cold winter) in 2017 and 2018. During the observation periods, the concentration distribution analyses for air pollutants (except CO, PM) showed significant heterogeneity in their distributions in space. In addition, the correlation analysis with the meteorological factors showed that CO concentrations were affected by wind speed (winter, R²=0.22-0.25), but the other air pollutants were not directly correlated. We also examined the effects of land-use and building configuration on air pollution distributions. More details concerning these results are presented.

Keywords: Sensor network, low-cost sensor, spatial heterogeneity, micro-built environments