A generic dark-target aerosol retrieval algorithm for use on MODIS, VIIRS and other multi-spectral sensors

A current research topic concerns the role of aerosols within our climate system. To quantify aerosol impacts, we first need to characterize the global distribution and variability of aerosol properties, including aerosol optical depth (AOD). To this end, there are a number of algorithms to retrieve global aerosol properties from multi-spectral reflectance observations. Over the past decade, we have continually updated the dark-target aerosol retrieval from Moderate Imaging resolution Spectroradiometer (MODIS), and have reached a point where we can robustly characterize errors and uncertainties of the retrieved aerosol products. As MODIS will be replaced by VIIRS (on Suomi-NPP) and future sensors, we need a consistent retrieval algorithm to extend the data record started by MODIS. To this end, we have developed a 'generic' retrieval algorithm, based on the Collection 6 MODIS algorithm. This algorithm is generic in that the same procedures are followed to prepare lookup tables, derive gas absorption corrections, and other assumptions of radiative transfer. However, the specific characteristics of the instrument (e.g. spectral response functions, resolution, etc) are taken into account. At the same time, cloud masking and pixel selection procedures are homogenized. The beauty of this generic algorithm is that it is not limited only to MODIS and VIIRS. In addition to current and future satellite datasets, it can be applied to historical MODIS Airborne Simulator (MAS) data taken during field experiments since the mid 1990's. With this capability, we will be able to look at aerosol variability at very high resolution (e.g. 50 meters or less), and fine-tune our understanding of aerosols in cloud fields. We show examples of our generic dark-target algorithm applied to many datasets, and discuss remaining issues in generic dark-target algorithm development