An Optimal-Estimation-based Algorithm for Simultaneously Retrieving Aerosols and Ocean Parameters using Multi-angular Polarimetric Measurements

Aerosol and ocean remote sensing traditionally rely on top-of-atmosphere (TOA) radiance measurements to retrieve optical properties. However, there are ongoing challenges in retrieving chemical and physical parameters due to the limited information contents of the radiances. Polarization measurements, which are sensitive to several aerosol and ocean parameters, provide additional information on these parameters for more detailed characterization. The polarization measurements also give valuable insights into the interaction between aerosols and the atmosphere. Over the years, numerous efforts have been made to measure polarization, including the POLDER generations and the Glory mission. More recently, two passive multi-angular polarimeters, HARP-2 and SPEXone, were launched in February 2024 as part of the PACE mission. This study analyzed the benefit of measuring polarization for retrieving aerosols and oceans based on the optimal-estimation method (OEM). The retrieval sensitivities of two measurement types are compared: one using radiance data alone and another combining radiance with Degree of Linear Polarization (DoLP). We aim to develop an algorithm based on the optimal estimation method that can retrieve aerosols and oceans simultaneously with estimated uncertainties. The key products generated by this algorithm include Aerosol Optical Depth (AOD), the real and imaginary parts of the refractive indices, Fine Mode Fraction (FMF), particle size parameters, and ocean parameters (e.g., ocean surface roughness, chlorophyll-a concentration and CDOM). The improved retrieval capabilities may contribute to a better understanding of atmosphere-ocean interactions.

Acknowledgements

This research was supported by Particulate Matter Management Specialized Graduate Program through the Korea Environmental Industry & Technology Institute(KEITI) funded by the Ministry of Environment(MOE)