A statistical model for the estimation of air pollutant concentrations in real time at any locations in Hong Kong and mainland China

Hong Kong and other cities in China are suffering from a serious air pollution problem. Public awareness of ambient air quality has been growing since poor air quality can cause serious health problems. Monitoring of air quality is probably a starting yet crucial step to mitigating air pollution. However, due to high operation cost, ground monitoring stations are usually limited and sparsely distributed, rendering them insufficient to capture large spatial variation of air quality over a city.

In this project, we have devised a spatio-temporal statistical model that utilizes MODIS data, AERONET data and meteorological parameters to improve the estimation accuracy. By using this new method, people can access real-time air pollutant concentrations, including PM2.5, PM10, NO2, Ozone, SO2 and CO, with higher accuracy at any locations.