Aerosol optical depth retrieval in the Arctic region using MODIS data over snow

The Arctic is vulnerable to the long-term transport of aerosols because they affect the surface albedo when particles are deposited on snow and ice. However, aerosol observations for this area are sparse and hence there is considerable uncertainty in the knowledge on the properties of the Arctic aerosol. Atmospheric remote sensing using satellite-based instruments offers an opportunity to obtain information on aerosol properties, in particular the aerosol optical depth (AOD) on an extended spatial scale as determined by the instrumentʹs swath width. However, AOD retrieval over a bright surface is a difficult task because it is difficult to separate and explicitly describe the contribution of the surface and that due to back-scattering by aerosols to the radiance observed by a satellite at the top of the atmosphere (TOA), especially at large Solar Zenith Angles (SZA). In this paper, an approach to achieve this is presented based on a synergetic approach using data from both Moderate Resolution Imaging Spectroradiometer (MODIS) instruments flying on the TERRA and AQUA satellites. The approached also uses prior knowledge for aerosol properties retrieval over snow as well as a Snow Bidirectional Reflectance Distribution Function (BRDF) model. The detailed analysis of the model results demonstrates that the Aerosol Properties Retrieval over Snow (APRS) algorithm is suitable for Arctic region Aerosol Optical Depth (AOD) retrieval. The study periods include April 2010 and April 2011, when the Arctic haze mostly occurs. Six AERONET stations at high latitude (Andenes, Barrow, Ittoqqortoormiit, OPAL, Thule, and PEARL) were used for comparison. The correlation coefficient between retrieved AODs and AERONET AODs was 0.8 and the relative error is between 10% and 20%, demonstrating the potential of the APRS method to retrieve AOD over the Arctic, with highly reflective snow/ice surfaces and large solar zenith angles.