Machine learning and bias correction of MODIS aerosol optical depth

Abstract

Machine-learning approaches (neural networks and support vector machines) are used to explore the reasons for a persistent bias between aerosol optical depth (AOD) retrieved from the MODerate resolution Imaging Spectroradiometer (MODIS) and the accurate ground-based Aerosol Robotic Network. While this bias falls within the expected uncertainty of the MODIS algorithms, there is room for algorithm improvement. The results of the machine-learning approaches suggest a link between the MODIS AOD biases and surface type. MODIS-derived AOD may be showing dependence on the surface type either because of the link between surface type and surface reflectance or because of the covariance between aerosol properties and surface type. © 2009 IEEE.