Improving Remote Sensing of Aerosol Microphysical Properties by Near‐Infrared Polarimetric Measurements Over Vegetated Land: Information Content Analysis

While polarimetric measurements contain valuable information regarding aerosol microphysical properties, polarization data in the near-infrared (NIR) bands have not been widely utilized. This paper evaluates whether the aerosol property information contents from single-viewing satellite polarimetric measurements at 1,610 and 2,250 nm can be used to improve the retrieval of aerosol parameters over vegetated land, in combination with shorter-wavelength bands (490, 670, and 870 nm). The a priori information and errors for the analysis are derived by assuming that the surface reflectance at visible wavelengths can be derived from the top of atmosphere at 2,250 nm. The information content in the synthetic data set is investigated for 10 aerosol parameters characterizing the columnar aerosol volumes (V f 0 and V0), particle size distributions (r f eff, v f eff, r c eff, and v c eff), and refractive indices (m f r , m f i , m c r , and m c i ) for the fineand coarse-mode aerosol models, respectively, and one parameter C characterizing the surface polarization. The results indicate that the degrees of freedom for signal can be increased by at least 2 with the addition of NIR measurements and that one to three additional parameters could be further retrieved with significantly decreased uncertainties. In addition, the 1,610 nm band is necessary for the simultaneous retrieval of V f 0, m f r , and r f eff for the fine mode dominated aerosols, while the 1,610 and 2,250 nm bands are both indispensable for retrieving V f 0, V c 0, m c r , r f eff , and r c eff in tandem for the coarse mode dominated aerosols. The analysis also reveals that C could be further retrieved by including scalar radiance and that measurement errors have significantly larger influences on the retrieval uncertainties than model errors.