Novel Decomposition Scheme for Characterizing Urban Air Quality with MODIS

Abstract

Urban air pollution is one of the most widespread global sustainability problems. Previous research has studied growth or fall of particulate matter (PM) levels using on-ground monitoring stations in urban regions. However, studying this worldwide is difficult because most cities do not have sufficient infrastructure to monitor air quality. Thus, satellite data is increasingly being employed to solve this limitation. In this paper, we use 16 years (2001–2016) of aerosol optical depth (AOD) and Angstrom exponent ( α ) datasets, retrieved from MODIS (Moderate Resolution Imaging Spectroradiometer) sensors on the National Aeronautics and Space Administration’s (NASA) Terra satellite to study air quality over 60 locations globally. We propose a novel technique, called AirRGB decomposition, to characterize urban air quality by decomposing AOD and α retrievals into ‘components’ of three distinct scenarios. In the AirRGB decomposition method, using AOD and α dataset three scenarios were investigated: ‘R’—high α and high AOD, ‘G’—high α and low AOD, and ‘B’—low α and low AOD values. These scenarios were mapped and quantified over a triangular red, green and blue color scale. This visualization easily segregates regions having a high concentration of industrial aerosol from only natural aerosols. Our analysis indicates that a sharp divide exists between North American and European cities and Asian cities in terms of baseline pollution and slopes of R and G trends. We found that while pollution in cities in China has started to decrease (e.g., since 2011 for Beijing), it continues to increase in South Asia and Southeast Asia. e.g., R offset of Beijing and New Delhi was 54.98 and 50.43 respectively but R slope was −0.04 and 0.08 respectively. High offset (≥45) and slope (≥0.025) of B for New York, Tokyo, Sydney and Sao Paolo shows that they have clean air, which is still getting better.