The objective of this chapter is to demonstrate the application potential of data merging and information reconstruction techniques in supporting air quality monitoring. Specifically, satellite-based Aerosol Optical Depth (AOD) products derived from MODIS instruments on board Terra and Aqua are first merged together to increase the inherent spatial coverage. Next, the information reconstruction method is applied to the previous merged AOD data set for further spatial coverage improvement by filling in the remnant data gaps. Finally, such enhanced AOD product is used to estimate relevant PM2.5 concentrations for better air quality management and public health assessment therein. The study region is over the Atlanta metropolitan area in Georgia, United States of America. The essential techniques used in this chapter include:

data merging via modified quantile-quantile adjustment method,

gap filling via smart information reconstruction method,

statistical modeling of PM2.5 from AOD and other external factors by using linear (Multiple Linear Regression, MLR) and nonlinear (Artificial Neural Network, ANN) techniques.

Modeling performances of MLR and ANN models are finally evaluated by comparing with in situ PM2.5 concentration measurements.