An Atmospheric Correction Algorithm for Remote Sensing of Bright Coastal Waters Using MODIS Land and Ocean Channels in the Solar Spectral Region

The present operational atmospheric correction algorithm for multichannel remote sensing of ocean color using imaging data acquired with the Moderate Resolution Imaging Spectroradiometer (MODIS) works well over clear ocean but can give incorrect results over brighter coastal waters. This is because: 1) the turbid waters are not dark for the two atmospheric correction channels centered near 0.75 and 0.86 mum; and 2) the ocean color channels (0.488, 0.531, and 0.551 mum) often saturate over bright coastal waters. Here, we describe an atmospheric correction algorithm for multichannel remote sensing of coastal waters. This algorithm is a modification of our previously developed atmospheric correction algorithm for hyperspectral data that uses lookup tables generated with a vector radiative transfer code and multilayered atmospheric models. Aerosol models and optical depths are determined by a spectrum-matching technique utilizing channels located at wavelengths longer than 0.86 mum, where the ocean surface is dark. The aerosol information in the visible spectral region is estimated based on the derived aerosol models and optical depths. Water-leaving radiances in the visible spectral region are obtained by subtracting out the atmospheric path radiances from the satellite-measured total radiances. Applications of the algorithm to two MODIS data sets are presented and compared to field measurements. The water-leaving reflectances retrieved with this algorithm over brighter shallow coastal waters compare closely with those from field measurements. In addition, the retrieved water-leaving reflectances over deeper ocean waters compare well with those derived with the MODIS operational algorithm