Polarization impacts on the water-leaving radiance retrieval from above-water radiometric measurements

• Published in: Applied Optics
• Main Authors: Harmel, Tristan, Gilerson, Alexander, Tonizzo, Alberto, Chowdhary, Jacek, Weidemann, Alan, Arnone, Robert, Ahmed, Sam
• Other Authors: Laboratoire d'océanographie de Villefranche (LOV), Observatoire océanologique de Villefranche-sur-mer (OOVM), Centre National de la Recherche Scientifique (CNRS)-Institut national des sciences de l'Univers (INSU - CNRS)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS)-Institut national des sciences de l'Univers (INSU - CNRS)-Université Pierre et Marie Curie - Paris 6 (UPMC)-Centre National de la Recherche Scientifique (CNRS), NASA Goddard Institute for Space Studies (GISS), NASA Goddard Space Flight Center (GSFC)
• Format: Article in Journal/Newspaper
• Language: English
• Published: HAL CCSD 2012
• Subjects: [SDU.STU.OC]Sciences of the Universe [physics]/Earth Sciences/Oceanography; Long Island; Long Island Sound; Aerosol Robotic Network
• Online Access: https://hal.archives-ouvertes.fr/hal-03502652; https://doi.org/10.1364/AO.51.008324

Above-water measurements of water-leaving radiance are widely used for water-quality monitoring and ocean-color satellite data validation. Reflected skylight in above-water radiometry needs to be accurately estimated prior to derivation of water-leaving radiance. Up-to-date methods to estimate reflection of diffuse skylight on rough sea surfaces are based on radiative transfer simulations and sky radiance measurements. But these methods neglect the polarization state of the incident skylight, which is generally highly polarized. In this paper, the effects of polarization on the sea surface reflectance and the subsequent water-leaving radiance estimation are investigated. We show that knowledge of the polarization field of the diffuse skylight significantly improves above-water radiometry estimates, in particular in the blue part of the spectrum where the reflected skylight is dominant. A newly developed algorithm based on radiative transfer simulations including polarization is described. Its application to the standard Aerosol Robotic Network-Ocean Color and hyperspectral radiometric measurements of the 1.5-year dataset acquired at the Long Island Sound site demonstrates the noticeable importance of considering polarization for water-leaving radiance estimation. In particular it is shown, based on time series of collocated data acquired in coastal waters, that the azimuth range of measurements leading to good-quality data is significantly increased, and that these estimates are improved by more than 12% at 413 nm. Full consideration of polarization effects is expected to significantly improve the quality of the field data utilized for satellite data validation or potential vicarious calibration purposes. (c) 2012 Optical Society of America