Abstract. NASA's Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission, scheduled for launch in the timeframe of 2023, will carry a hyperspectral scanning radiometer named the Ocean Color Instrument (OCI) and two multi-angle polarimeters (MAPs): the UMBC Hyper-Angular Rainbow Polarimeter (HARP2) and the SRON Spectro-Polarimeter for Planetary EXploration one (SPEXone). The MAP measurements contain rich information on the microphysical properties of aerosols and hydrosols and therefore can be used to retrieve accurate aerosol properties for complex atmosphere and ocean systems. Most polarimetric aerosol retrieval algorithms utilize vector radiative transfer models iteratively in an optimization approach, which leads to high computational costs that limit their usage in the operational processing of large data volumes acquired by the MAP imagers. In this work, we propose a deep neural network (NN) forward model to represent the radiative transfer simulation of coupled atmosphere and ocean systems for applications to the HARP2 instrument and its predecessors. Through the evaluation of synthetic datasets for AirHARP (airborne version of HARP2), the NN model achieves a numerical accuracy smaller than the instrument uncertainties, with a running time of 0.01 s in a single CPU core or 1 ms in a GPU. Using the NN as a forward model, we built an efficient joint aerosol and ocean color retrieval algorithm called FastMAPOL, evolved from the well-validated Multi-Angular Polarimetric Ocean coLor (MAPOL) algorithm. Retrievals of aerosol properties and water-leaving signals were conducted on both the synthetic data and the AirHARP field measurements from the Aerosol Characterization from Polarimeter and Lidar (ACEPOL) campaign in 2017. From the validation with the synthetic data and the collocated High Spectral Resolution Lidar (HSRL) aerosol products, we demonstrated that the aerosol microphysical properties and water-leaving signals can be retrieved efficiently and within acceptable error. Comparing to the retrieval speed using a conventional radiative transfer forward model, the computational acceleration is 103 times faster with CPU or 104 times with GPU processors. The FastMAPOL algorithm can be used to operationally process the large volume of polarimetric data acquired by PACE and other future Earth-observing satellite missions with similar capabilities.
Passive ocean color images have provided a sustained synoptic view of the distribution of ocean optical properties and color and biogeochemical parameters for the past twenty-plus years. These images have revolutionized our view of the ocean. Remote sensing of ocean color has relied on measurements of the radiance emerging at the top of the atmosphere, thus neglecting the polarization and the vertical components. Ocean color remote sensing utilizes the intensity and spectral variation of visible light scattered upward from beneath the ocean surface to derive concentrations of biogeochemical constituents and inherent optical properties within the ocean surface layer. However these measurements have some limitations. Specifically, the measured property is a only weighted-integrated values over from a relatively shallow depth, the measurements provide no information during the night and retrieval are compromised by clouds, absorbing aerosols and low Sun zenithal angles. In addition, ocaen color data provide limited information on the morphology and size distribution of marine particles. Major advances in our understanding of global ocean ecosystems will require measurements from new technologies, specifically lidar and polarimetry. These new techniques have been widely used for atmospheric applications but have not had as much as interest from the ocean color community. This is due to many factors including limited access to in-situ instruments and/or space-borne sensors and lack of attention in university courses and ocean science summer schools curricula. However, lidar and polarimetry technology will provide an complement to standard ocean color products by providing depth-resolved values of attenuation and scattering parameters and additional information about particles morphology and chemical composition. This review aims at presenting the basics of these techniques, examples of applications and at advocating for the development of in-situ and space-borne sensors. Recommendations are provided on actions that would foster the embrace of lidar and polarimetry as powerful remote sensing tools by the ocean science community.
Abstract. Multi-angle polarimetric (MAP) measurements can enable detailed characterization of aerosol microphysical and optical properties and improve atmospheric correction in ocean color remote sensing. Advanced retrieval algorithms have been developed to obtain multiple geophysical parameters in the atmosphere–ocean system. Theoretical pixel-wise retrieval uncertainties based on error propagation have been used to quantify retrieval performance and determine the quality of data products. However, standard error propagation techniques in high-dimensional retrievals may not always represent true retrieval errors well due to issues such as local minima and the nonlinear dependence of the forward model on the retrieved parameters near the solution. In this work, we analyze these theoretical uncertainty estimates and validate them using a flexible Monte Carlo approach. The Fast Multi-Angular Polarimetric Ocean coLor (FastMAPOL) retrieval algorithm, based on efficient neural network forward models, is used to conduct the retrievals and uncertainty quantification on both synthetic HARP2 (Hyper-Angular Rainbow Polarimeter 2) and AirHARP (airborne version of HARP2) datasets. In addition, for practical application of the uncertainty evaluation technique in operational data processing, we use the automatic differentiation method to calculate derivatives analytically based on the neural network models. Both the speed and accuracy associated with uncertainty quantification for MAP retrievals are addressed in this study. Pixel-wise retrieval uncertainties are further evaluated for the real AirHARP field campaign data. The uncertainty quantification methods and results can be used to evaluate the quality of data products, as well as guide MAP algorithm development for current and future satellite systems such as NASA’s Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission.
Polarized light in the oceans carries intrinsic information that can be utilized to estimate the optical and microphysical properties of the oceanic hydrosols. It is especially sensitive to the scattering coefficient, which cannot be retrieved from the unpolarized light used in current ocean color remote sensing algorithms. Through the unpolarized remote sensing reflectance (Rrs), these classical algorithms can only estimate backscattering coefficients bb, but the total scattering coefficient b could be solely retrieved based on the characteristics of polarized light. The correlation is quantified in this paper. Based on extensive simulations using the vector radiative transfer program RayXP, the attenuation-to-absorption ratio (c/a), from which b is readily computed, is shown to be closely related to the degree of linear polarization (DoLP). The relationship is investigated for the upwelling polarized light for several wavelengths in the visible part of the spectrum, for a complete set of viewing geometries, and for varying concentrations of phytoplankton, non-algal particles, and color dissolved organic matter (CDOM) in the aquatic environment. It is shown that there is an excellent correlation between the DoLP and c/a for a wide range of viewing geometries. That correlation is investigated theoretically using fitting techniques, which show that it depends not only on the general composition of water but also on the particle size distribution (PSD) of the (mainly non-algal) particles. A large dataset of Stokes components for various water compositions, measured in the field with a hyperspectral and multi-angular polarimeter, then provides the opportunity to validate the parameterized relationship between DOLP and c/a. This study opens the possibility for the retrieval of additional inherent optical properties (IOPs) from air- or space-borne DoLP measurements of the ocean.
The attenuation coefficient of the water body is not directly retrievable from measurements of unpolarized water-leaving radiance.Based on extensive radiative transfer simulations using the vector radiative transfer code RayXP, it is demonstrated that the underwater degree of linear polarization (DoLP) is closely related to the attenuation-to-absorption ratio (c/a) of the water body, a finding that enables retrieval of the attenuation coefficient from measurements of the Stokes components of the upwelling underwater polarized light field.The relationship between DoLP and the c/a ratio is investigated for the upwelling polarized light field for a complete set of viewing geometries, at several wavelengths in the visible part of the spectrum; for varying compositions of the aquatic environment, whose constituents include phytoplankton, non-algal particles, and color dissolved organic matter (CDOM); and for varying microphysical properties such as the refractive index and the slope of the Junge-type particle size distribution (PSD).Consequently, this study reveals the possibility for retrieval of additional inherent optical properties (IOPs) from air-or space-borne DoLP measurements of the water-leaving radiation.
Abstract. The University of Maryland, Baltimore County (UMBC) Hyper-Angular Rainbow Polarimeter (HARP2) will be on board NASA's Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission, scheduled for launch in January 2024. In this study we systematically evaluate the retrievability and uncertainty of aerosol and ocean parameters from HARP2 multi-angle polarimeter (MAP) measurements. To reduce the computational demand of MAP-based retrievals and maximize data processing throughput, we developed improved neural network (NN) forward models for spaceborne HARP2 measurements over a coupled atmosphere and ocean system within the FastMAPOL retrieval algorithm. To this end, a cascading retrieval scheme is implemented in FastMAPOL, which leverages a series of NN models of varying size, speed, and accuracy to optimize performance. Two sets of NN models are used for reflectance and polarization, respectively. A full day of global synthetic HARP2 data was generated and used to test various retrieval parameters including aerosol microphysical and optical properties, aerosol layer height, ocean surface wind speed, and ocean chlorophyll a concentration. To assess retrieval quality, pixel-wise retrieval uncertainties were derived from error propagation and evaluated against the difference between the retrieval parameters and truth based on a Monte Carlo method. We found that the fine-mode aerosol properties can be retrieved well from the HARP2 data, though the coarse-mode aerosol properties are more uncertain. Larger uncertainties are associated with a reduced number of available viewing angles, which typically occur near the scan edge of the HARP2 instrument. Results of the performance assessment demonstrate that the algorithm is a viable approach for operational application to HARP2 data after the PACE launch.
Total and polarized radiances from above the ocean surface are measured by a state-of-the-art snapshot hyperspectral imager. A computer-controlled filter wheel is installed in front of the imager allowing for recording of division-of-time Stokes vector images from the ocean surface. This system, to the best of our knowledge, for the first time provided a capability of hyperspectral polarimetric multi-angular measurements of radiances from above the water surface. Several sets of measurements used in the analysis were acquired from ocean platforms and from shipborne observations. Measurements made by the imager are compared with simulations using a vector radiative transfer (VRT) code showing reasonable agreement. Analysis of pixel-to-pixel variability of the total and polarized above-water radiance for the viewing angles of 20°-60° in different wind conditions enable the estimation of uncertainties in measurements of these radiances in the polarized mode for the spectral range of 450-750 nm, thus setting requirements for the quality of polarized measurements. It is shown that there is a noticeable increase of above-water degree of linear polarization (DoLP) as a function of the viewing angle, which is due both to the larger DoLP of the light from the water body and the light reflected from the ocean surface. Results of measurements and VRT simulations are applied for the multi-angular retrieval of the ratio of beam attenuation coefficient (ctot) to absorption coefficient (atot) in addition to the other parameters such as absorption and backscattering coefficients retrieved from traditional unpolarized methods.
Abstract. Since early 2000, NASA's Multi-angle Imaging SpectroRadiometer (MISR) instrument has been performing remote sensing retrievals of aerosol optical properties from the polar-orbiting Terra spacecraft. A noteworthy aspect of MISR observations over the ocean is that, for much of the Earth, some of the multi-angle views have contributions from solar reflection by the ocean surface (glint, or glitter), while others do not. Aerosol retrieval algorithms often discard these glint-influenced observations because they can overwhelm the signal and are difficult to predict without knowledge of the (wind-speed-driven) ocean surface roughness. However, theoretical studies have shown that multi-angle observations of a location at geometries with and without reflected sun glint can be a rich source of information, sufficient to support simultaneous retrieval of both the aerosol state and the wind speed at the ocean surface. We are in the early stages of creating such an algorithm. In this paper, we describe our assessment of the appropriate level of parameterization for simultaneous aerosol and ocean surface property retrievals using sun glint. For this purpose, we use generalized nonlinear retrieval analysis (GENRA), an information content assessment (ICA) technique employing Bayesian inference, and simulations from the Ahmad–Fraser iterative radiative transfer code. We find that four parameters are suitable: aerosol optical depth (τ), particle size distribution (expressed as the fine mode fraction f of small particles in a bimodal size distribution), surface wind speed (w), and relative humidity (r, to define the aerosol water content and complex refractive index). None of these parameters define ocean optical properties, as we found that the aerosol state could be retrieved with the nine MISR near-infrared views alone, where the ocean body is strongly absorbing in the open ocean. We also found that retrieval capability varies with observation geometry and that as τ increases so does the ability to determine aerosol intensive optical properties (r and f, while it decreases for w). Increases in w decrease the ability to determine the true value of that parameter but have minimal impact on retrieval of aerosol properties. We explored the benefit of excluding the two most extreme MISR view angles for which radiative transfer with the plane-parallel approximation is less certain, but we found no advantage in doing so. Finally, the impact of treating wind speed as a scalar parameter, rather than as a two-parameter directional wind, was tested. While the simpler scalar model does contribute to overall aerosol uncertainty, it is not sufficiently large to justify the addition of another dimension to parameter space. An algorithm designed upon these principles is in development. It will be used to perform an atmospheric correction with MISR for coincident ocean color (OC) observations by the Moderate Resolution Imaging Spectroradiometer (MODIS) instrument, also on the NASA Terra spacecraft. Unlike MISR, MODIS is a single-view-angle instrument, but it has a more complete set of spectral channels ideal for determination of optical ocean properties. The atmospheric correction of MODIS OC data can therefore benefit from MISR aerosol retrievals. Furthermore, higher-spatial-resolution data from coincident MISR observations may also improve glint screening.
Ocean color remote sensing is an important tool to monitor water quality and biogeochemical conditions of ocean.Atmospheric correction, which obtains water-leaving radiance from the total radiance measured by satellite-borne or airborne sensors, remains a challenging task for coastal waters due to the complex optical properties of aerosols and ocean waters.In this paper, we report a research algorithm on aerosol and ocean color retrieval with emphasis on coastal waters, which uses coupled atmosphere and ocean radiative transfer model to fit polarized radiance measurements at multiple viewing angles and multiple wavelengths.Ocean optical properties are characterized by a generalized bio-optical model with direct accounting for the absorption and scattering of phytoplankton, colored dissolved organic matter (CDOM) and non-algal particles (NAP).Our retrieval algorithm can accurately determine the water-leaving radiance and aerosol properties for coastal waters, and may be used to improve the atmospheric correction when apply to a hyperspectral ocean color instrument.
