Joint retrieval of aerosol and water-leaving radiance from multispectral, multiangular and polarimetric measurements over ocean

Abstract

Abstract. An optimization approach has been developed for simultaneous retrieval of aerosol properties and normalized water-leaving radiance (nLw) from multispectral, multiangular, and polarimetric observations over ocean. The main features of the method are (1) use of a simplified bio-optical model to estimate nLw, followed by an empirical refinement within a specified range to improve its accuracy; (2) improved algorithm convergence and stability by applying constraints on the spatial smoothness of aerosol loading and Chlorophyll a (Chl a) concentration across neighboring image patches and spectral constraints on aerosol optical properties and nLw across relevant bands; and (3) enhanced Jacobian calculation by modeling and storing the radiative transfer (RT) in aerosol/Rayleigh mixed layer, pure Rayleighscattering layers, and ocean medium separately, then coupling them to calculate the field at the sensor. This approach avoids unnecessary and time-consuming recalculations of RT in unperturbed layers in Jacobian evaluations. The Markov chain method is used to model RT in the aerosol/Rayleigh mixed layer and the doubling method is used for the uniform layers of the atmosphere-ocean system. Our optimization approach has been tested using radiance and polarization measurements acquired by the Airborne Multiangle SpectroPolarimetric Imager (AirMSPI) over the AERONET USC_SeaPRISM ocean site (6 February 2013) and near the AERONET La Jolla site (14 January 2013), which, respectively, reported relatively high and low aerosol loadings. Validation of the results is achieved through comparisons to AERONET aerosol and ocean color products. For comparison, the USC_SeaPRISM retrieval is also performed by use of the Generalized Retrieval of Aerosol and Surface Properties algorithm (Dubovik et al., 2011). Uncertainties of aerosol and nLw retrievals due to random and systematic instrument errors are analyzed by truth-in/truth-out tests with three Chl a concentrations, five aerosol loadings, three different types of aerosols, and nine combinations of solar incidence and viewing geometries.

GOCI Yonsei aerosol retrieval version 2 products: an improved algorithm and error analysis with uncertainty estimation from 5-year validation over East Asia

“…Vector radiative transfer calculations (i.e., consideration of polarization) and spherical-shell atmosphere approximations can calculate Rayleigh scattering at high accuracy and may improve the accuracy of the GOCI YAER algorithm. Also, recent statistically optimized aerosol retrieval algorithms utilizing the characteristics of spatial and temporal smoothness constraints for aerosols result in improved accuracy by increasing the aerosol signal (Dubovik et al, 2011;Xu et al, 2016). They also enable the simultaneous retrieval of multiple geophysical variables, such as aerosol and surface reflectance over land and aerosol and chlorophyll concentrations over the ocean, which can reduce the remaining error due to the predefined surface reflectance over ocean and land surfaces in the GOCI YAER algorithm.…”

“…The V1 land AOD had negative biases of −0.1 for low SD and −0.05 for high SD and was persistently affected by surface reflectance issues and/or cloud contamination. Note that some recent aerosol retrieval algorithms have adopted a statistical spatial smoothness constraint for AOPs in the inversion procedure to improve accuracy (Dubovik et al, 2011;Xu et al, 2016).…”

Retrievals of aerosol optical and microphysical properties from Imaging Polar Nephelometer scattering measurements

“…Moreover, while all of the above features have never been used in one single application, they often provide important potential for evolution of each application, for example via implementing synergy retrievals using a combination of different observations. The GRASP algorithm has previously been successfully applied to both satellite and ground-based upward-looking sky radiance measurements (Dubovik et al, 2011(Dubovik et al, , 2014Xu et al, 2016), while this paper represents the first application of GRASP to polar nephelometer data.…”

Simultaneous determination of aerosol optical thickness and water leaving radiance from multispectral measurements in coastal waters

“…20 In future, optimization of the bio-optical model should be conducted using more extensive in situ observation data, such as the study of Maritorena et al (2002), to have a better modeling of radiation process in the ocean system, since small variations of inherent optical properties of ocean substance may translate into large nL w retrieval errors. Moreover, improvement of algorithm adopting multi-pixel constrain technique (Dubovik et al, 2011;Xu et al, 2016;Hashimoto and Nakajima, 2017) and acceleration scheme (Takenaka et al, 2011) is also part of our ongoing work. (1984), 105, 20673-20696, 2000 Statistically optimized inversion algorithm for enhanced retrieval of aerosol properties from spectral multi-angle polarimetric satellite observations, Atmos.…”

“…These one-step retrieval approaches involve a direct inversion scheme that simultaneously determines aerosols and oceanic substances with 10 minimal assumption, which have an advantage in dealing flexibly with the inversion problems in the absorption aerosol loading conditions and increasing the available measurements information content. Most of these approaches adopt the radiative transfer (RT) model to enable simulated radiance to converge with observation by defining the appropriate retrieved parameters and estimation schemes or neutral network method, to make full use of the satellite observed data, as well as multiple geometry, pixel or polarization information (Dubovik et al, 2011;Hasekamp et al, 2011;Knobelspiesse et 15 al., 2012;Xu et al, 2016).…”