Real-time Air Quality Monitoring and Personal Exposure Estimation

Prof. HUANG Bo

As a major environmental problem that threatens human health, air pollution led to around 6.4 million premature deaths all over the world in 2015. Public awareness of polluted air has been growing and there is an increasing demand for access to more accurate and timely air pollution data so as to reduce the population’s exposure to unhealthy air. However, due to high set-up and operation costs, ground monitoring stations are usually limited in number and often sparsely distributed, rendering them insufficient to capture the large spatial variation of urban air quality.

Under the support of Hong Kong RGC/GRF, Prof. Huang Bo and his research group have explored geospatial big data and developed a mobile application called TouchAir to address this problem. Touchair incorporates a sophisticated spatiotemporal statistical model, which is able to estimate six principle air pollutant concentrations (PM2.5, PM10, NO2, O3, SO2 and CO) in real-time and at any location, given the data crawled from the sparsely distributed environmental monitoring stations. The spatio-temporal statistical model also makes use of satellite imagery with areal coverage, aerosol network (AERONET) data, traffic conditions and meteorological data all trawled from the internet to improve the estimation accuracy up to 80%. Apart from showing real time data, Touchair could also predict air quality in the next 24 hours and record users’ daily exposure to fine suspended particles, PM2.5. These functions help citizens reduce exposure to air pollution in their daily living and traveling and thus make informed choices towards a healthy lifestyle.