Abstract
This paper uses Moderate Resolution Imaging Spectroradiometer (MODIS) data to investigate the spatial and temporal variations of aerosol optical thickness (AOT) over Guangdong, the most developed province in China, during 2010-2012. Linear regression and self-organizing maps (SOM) are used to investigate the relationship between AOT and its affecting factors, including Normalized Difference Vegetation Index (NDVI), elevation, urbanized land fraction, and several socio-economic variables. Results show that the highest values of τ 0.55 mainly occur over the rapidly-developing Pearl River Delta (PRD) region and the eastern coast. Seasonal averaged AOT is highest in summer (0.416), followed by spring (0.351), winter (0.292), and autumn (0.254). From unary linear regression and SOM analysis, AOT is shown to be strongly negatively correlated to NDVI (R(2) = 0.782) and elevation (R(2) = 0.731), and positively correlated with socio-economic factors, especially GDP, industry and vehicle density (R(2) above 0.73), but not primary industry. Multiple linear regression between AOT and the contributing factors shows much higher R(2) values (>0.8), indicative of the clear relationships between AOT and variables. This study illustrates that human activities have strong impacts on aerosols distribution in Guangdong Province. Economic and industrial developments, as well as vehicle density, are the main controlling factors on aerosol distribution.