Elsevier

Remote Sensing of Environment

Volume 118, 15 March 2012, Pages 284-308
Remote Sensing of Environment

Sensitivity of multiangle, multispectral polarimetric remote sensing over open oceans to water-leaving radiance: Analyses of RSP data acquired during the MILAGRO campaign

https://doi.org/10.1016/j.rse.2011.11.003Get rights and content

Abstract

For remote sensing of aerosol over the ocean, there is a contribution from light scattered under water. The brightness and spectrum of this light depends on the biomass content of the ocean, such that variations in the color of the ocean can be observed even from space. Rayleigh scattering by pure sea water, and Rayleigh–Gans type scattering by plankton, causes this light to be polarized with a distinctive angular distribution. To study the contribution of this underwater light polarization to multiangle, multispectral observations of polarized reflectance over ocean, we previously developed a hydrosol model for use in underwater light scattering computations that produces realistic variations of the ocean color and the underwater light polarization signature of pure sea water. In this work we review this hydrosol model, include a correction for the spectrum of the particulate scattering coefficient and backscattering efficiency, and discuss its sensitivity to variations in colored dissolved organic matter (CDOM) and in the scattering function of marine particulates. We then apply this model to measurements of total and polarized reflectance that were acquired over open ocean during the MILAGRO field campaign by the airborne Research Scanning Polarimeter (RSP). Analyses show that our hydrosol model faithfully reproduces the water-leaving contributions to RSP reflectance, and that the sensitivity of these contributions to Chlorophyll a concentration [Chl] in the ocean varies with the azimuth, height, and wavelength of observations. We also show that the impact of variations in CDOM on the polarized reflectance observed by the RSP at low altitude is comparable to or much less than the standard error of this reflectance whereas their effects in total reflectance may be substantial (i.e. up to > 30%). Finally, we extend our study of polarized reflectance variations with [Chl] and CDOM to include results for simulated spaceborne observations.

Highlights

►A hydrosol model for numerical studies of polarized radiance emerging from open oceans is updated. ►Studies show this radiance to vary with viewing angle, wavelength λ, and Chlorophyll a amount [Chl]. ►Waterleaving radiance computations are reproduced in airborne photopolarimetric measurements. ►Bio-optical variation of polarized waterleaving radiance with [Chl] cannot be ignored if λ  470 nm. ►Natural variation of polarized waterleaving radiance with CDOM is negligible for [Chl]  1 mg/m3.

Introduction

The polarized intensity of light scattered by particles exhibits features as a function of wavelength and scattering angle that are distinctly different from those of the total scattered intensity (Coulson, 1988, Ulaby and Elachi, 1990, Videen et al., 2004). The polarized and total intensity features also exhibit very different sensitivities to particle properties such as size relative to the wavelength, shape, and composition (Hansen and Travis, 1974, Mishchenko et al., 2002). Furthermore, both sensitivities vary with particle properties. For example, the polarized component of light scattered by particles having a refractive index close to that of the surrounding medium, such as particulates suspended in the ocean, shows fewer features with scattering angle and less variation with shape and size than the corresponding features of light scattered by particles that have a strong refractive-index contrast with the surrounding medium such as atmospheric aerosols. Finally, the features in polarized intensity of singly scattered light are much less likely to be washed out by those of multiply scattered light than is the case for the features in total intensity (Hansen and Travis, 1974, Hovenier et al., 2004, Mishchenko et al., 2006, van de Hulst, 1980). This is because the magnitude of polarized intensity of light scattered n times decreases rapidly with n as compared with the magnitude of total intensity. These differences cause the retrieval of aerosol properties from remotely sensed polarization to be much more accurate than the corresponding retrievals from remotely sensed intensity, as has been demonstrated in theoretical studies (Hasekamp and Landgraf, 2005, Mishchenko and Travis, 1997a, Mishchenko and Travis, 1997b). Analyses of actual polarimetric remote sensing data obtained by the Polarization and Directionality of the Earth's Reflectance (POLDER; Deschamps et al., 1994) instrument confirm these studies – whether these analyses focus exclusively on POLDER observations (e.g., Bréon and Goloub, 1998, Herman et al., 2005), or on a comparison of aerosol retrievals from POLDER and from intensity-only remote sensing such as from the moderate resolution imaging spectrometer (MODIS; Barnes et al., 1998) instrument (Gérard et al., 2005).

The recognition of the advantages of polarimetric remote sensing (Hansen and Travis, 1974, Hovenier et al., 2004, Mishchenko et al., 2004, Mishchenko et al., 2006, Ulaby and Elachi, 1990, Videen et al., 2004) has led to the development of a new generation of satellite scanning instruments, such as the Aerosol Polarimetry Sensor (APS) instrument (Mishchenko et al., 2004, Mishchenko et al., 2007; see specifications in Section 2), that combine, and in many respects surpass (except for cross ground-track swath), the polarimetric and multiangle measurement strengths of POLDER and the multispectral measurement strengths of MODIS. The launch of such instruments will lead to more accurate retrievals of aerosols as demonstrated by Chowdhary et al. (2001), but it also puts more stringent requirements on the modeling of remote sensing data. For example, Chowdhary et al. (2002, 2005) show that for remote sensing over oceans the polarization of light measured in the visible part of the spectrum yields valuable information about the complex refractive index of fine mode aerosols provided that the contributions of polarized water leaving radiance are appropriately dealt with. They demonstrate that such contributions become very small for principal plane observations and consequently limited their analyses to such viewing geometries. Similar arguments were made by Chami et al., 2001, Harmel and Chami, 2008, although it was argued there from analyses for a single wavelength in the visible that polarized water leaving radiance over open oceans can be ignored for most viewing geometries available from space borne observations.

To provide a lower boundary condition for aerosol retrievals from polarized reflectance observations outside the principal plane over open oceans, Chowdhary et al. (2006, henceforth referred to as C2006) developed a hydrosol model that reproduces empirically observed variations in ocean albedos as a function of biomass concentration in the ocean and of the wavelength of observation, and that provides the corresponding angular variation in intensity and polarization of water leaving radiance. The primary purpose of this hydrosol model is therefore not for retrieval of marine particulates but to account for oceanic contributions in polarimetric remote sensing of aerosols as in Hasekamp et al. (2011). Nevertheless, one can change the microphysical and bio-optical properties prescribed for such particles in this model to mimic local (natural) variations and anomalous cases. The objectives of this paper are 1) to revisit this model and assess changes caused by variations in colored dissolved organic matter (CDOM) and in the scattering matrix of marine particulates, 2) to validate the water-leaving total and polarized radiance computed with this model against actual data obtained from aircraft at various altitudes and azimuth angles, and 3) to assess the sensitivity of spaceborne polarimetry over open oceans to variations in polarized water-leaving radiance. The organization of this paper is as follows. Section 2 provides a description of the airborne polarization data analyzed in this work, and of the field campaign in which these data were obtained. In Section 3, we evaluate the hydrosol model used in the analysis of underwater-light contributions to these data. The multiple scattering computations that link the hydrosol model to the polarization data are briefly reviewed in Section 4. Section 5 applies these computations to sensitivity studies of underwater light scattering, to analyses of the airborne polarization data, and to simulations of spaceborne polarization data. Finally, we summarize our results in Section 6.

Section snippets

Measurements

The primary measurements used in this work are obtained by the Research Scanning Polarimeter (RSP) instrument (Cairns et al., 1999), which is an airborne version of the APS satellite instrument (Peralta et al., 2007). The objectives of APS-like measurements are discussed and detailed in Mishchenko et al. (2007), and can be summarized as the retrieval of the optical thickness, size distribution, complex refractive index, and shape information for fine and coarse mode aerosols, as well as of the

Bulk ocean

To compute the scattering matrix and single scattering albedo of ocean waters, we consider these waters to be a bulk mixture of seawater, with scattering matrix Fw and scattering coefficient bw, and a particulate component, with scattering matrix Fp and scattering coefficient bp (see Table 1 for notation, dimension, and definition of variables). We follow the convention that a scattering matrix F has its scattering function referred to using the italic font face, i.e. F  F11 (C2006), and the

Multiple-scattering computations

The total and polarized reflectance of radiation emerging from the top of an atmosphere–ocean system (AOS) are calculated as in C2006, where various numerical recipes used to increase the efficiency of computations are described. That is, we first use the doubling/adding method for radiative transfer computations of polarized light (de Haan et al., 1987, Hovenier et al., 2004) to obtain the reflection and transmission properties of the atmosphere and the ocean body, and the geometric-optics

Irradiance ratio sensitivity analysis

The graphs in Fig. 3a depict model variations of the bulk ocean irradiance ratio Ablk, and in Fig. 3b of the bulk ocean absorption coefficient ablk, with wavelength λ for various concentrations of Chlorophyll a. The spectral range evaluated is from 400 to 700 nm which covers those spectral bands used for aerosol remote sensing by RSP and APS-like instruments, and also the spectral bands typically used for remote sensing of ocean color (Esaias et al., 1998). Note however that our models do not

Summary

In a previous paper (Chowdhary et al., 2006), we developed a variable detritus-plankton (D–P) hydrosol model to compute the multi-angle and multi-spectral behavior of polarized light emerging from open oceans as a function of Chlorophyll a concentration [Chl]. The purpose of that model was to account for the polarized underwater light contribution to measurements made by spaceborne polarimeters. In this work, we reviewed the properties of this hydrosol model and made a correction for the

Acknowledgements

The MILAGRO/INTEX-B Campaign is a collaborative effort of a large number of participants with the support of multi-national agencies. We thank the governments of the Federal District, the States of Mexico, Hildalgo and Veracruz, and the Mexican Ministries of the Environment, Foreign Relations, Defense and Finance for their logistical support; IMP, U-Tecámac, and Rancho La Bisnega for hosting the supersites; and many other Mexican institutions for their support. We further extend our gratitude

References (90)

  • W.L. Barnes et al.

    Prelaunch characteristics of the Moderate Resolution Imaging Spectrometer (MODIS) on EOS-AM1

    IEEE Transactions on Geoscience and Remote Sensing

    (1998)
  • R.W. Bergstrom et al.

    Aerosol spectral absorption in the Mexico City area: Results from airborne measurements during the MILAGRO/INTEX B

    Atmospheric Chemistry and Physics

    (2010)
  • F.-M. Bréon et al.

    Cloud droplet effective radius from spaceborne polarization measurements

    Geophysical Research Letters

    (1998)
  • A. Bricaud et al.

    Variations of light absorption by suspended particles with chlorophyll a concentrations in oceanic (case 1) waters: Analysis and implications for bio-optical models

    Journal of Geophysical Research

    (1998)
  • A. Bricaud et al.

    Erratum: Variations of light absorption by suspended particles with chlorophyll a concentrations in oceanic (case 1) waters: Analysis and implications for bio-optical models

    Journal of Geophysical Research

    (1999)
  • C.J. Bruegge et al.

    MISR prelaunch instrument calibration and results

    IEEE Transactions on Geoscience and Remote Sensing

    (1998)
  • B. Cairns et al.

    The Research Scanning Polarimeter: Calibration and ground-based measurements

    Proceedings of SPIE

    (1999)
  • B. Cairns et al.

    Polarimetric remote sensing of aerosols over land surfaces

  • M. Chami et al.

    Radiative transfer model for the computation of radiance and polarization in an atmosphere–ocean system: Polarization properties of suspended matter for remote sensing

    Applied Optics

    (2001)
  • J. Chowdhary et al.

    Retrievals of aerosol scattering and absorption properties from photopolarimetric observations over the ocean during the CLAMS experiment

    Journal of the Atmospheric Sciences

    (2005)
  • J. Chowdhary et al.

    Retrieval of aerosol properties over the ocean using multispectral and multiangle photopolarimetric measurements from the Research Scanning Polarimeter

    Geophysical Research Letters

    (2001)
  • J. Chowdhary et al.

    Case studies of aerosol retrievals over the ocean from multiangle, multispectral photopolarimetric remote sensing data

    Journal of the Atmospheric Sciences

    (2002)
  • J. Chowdhary et al.

    Contribution of waterleaving radiances to multiangle, multispectral polarimetric observations over the open ocean: Bio-optical model results for case 1 waters

    Applied Optics

    (2006)
  • H. Claustre et al.

    Is desert dust making oligotrophic waters greener?

    Geophysical Research Letters

    (2002)
  • K.L. Coulson

    Polarization and Intensity of Light in the Atmosphere

    (1988)
  • C. Cox et al.

    Statistics of the sea-surface derived from sun-glitter

    Journal of Marine Research

    (1954)
  • J.F. de Haan et al.

    The adding method for multiple scattering computations of polarized light

    Astronomy and Astrophysics

    (1987)
  • P.-Y. Deschamps et al.

    The POLDER mission: Instrument characteristics and scientific objectives

    IEEE Transactions on Geoscience and Remote Sensing

    (1994)
  • O. Dubovik et al.

    Application of spheroid models to account for aerosol particle nonsphericity in remote sensing of desert dust

    Journal of Geophysical Research

    (2006)
  • R.E. Eplee et al.

    Calibration of SeaWiFS. II. Vicarious techniques

    Applied Optics

    (2001)
  • R.E. Eplee et al.

    Cross calibration of SeaWiFS and MODIS using on-orbit observations of the moon

    Applied Optics

    (2011)
  • W.E. Esaias et al.

    An overview of MODIS capabilities for ocean science observations

    IEEE Transactions on Geoscience and Remote Sensing

    (1998)
  • B. Fougnie et al.

    PARASOL in-flight calibration and performance

    Applied Optics

    (2007)
  • G. Fournier et al.

    Analytical phase function for ocean water

  • G. Fournier et al.

    Computer-based underwater imaging analyses

  • R. Frouin et al.

    Spectral reflectance of sea foam in the visible and near-infrared: In situ measurements and remote sensing applications

    Journal of Geophysical Research

    (1996)
  • B. Gérard et al.

    Comparisons between POLDER2 and MODIS/Terra aerosol retrievals over ocean

    Journal of Geophysical Research

    (2005)
  • H.R. Gordon

    Can the Lambert-Beer law be applied to the diffuse attenuation coefficient of ocean water?

    Limnology and Oceanography

    (1989)
  • H.R. Gordon et al.

    Computed relationships between the inherent and apparent optical properties of a flat homogeneous ocean

    Applied Optics

    (1975)
  • H.R. Gordon et al.

    Estimation of the depth of sun light penetration in the sea for remote sensing

    Applied Optics

    (1975)
  • V. Haltrin

    One-parameter two-term Henyey-Greenstein phase function for light scattering in seawater

    Applied Optics

    (2002)
  • J.E. Hansen et al.

    Light scattering in planetary atmospheres

    Space Science Reviews

    (1974)
  • T. Harmel et al.

    Invariance of polarized reflectance measured at the top of the atmosphere by PARASOL satellite instrument in the visible range with marine constituents in open ocean waters

    Optics Express

    (2008)
  • O.P. Hasekamp et al.

    Retrieval of aerosol properties over the ocean from multispectral single-viewing-angle measurements of intensity and polarization: Retrieval approach, information content, and sensitivity study

    Journal of Geophysical Research

    (2005)
  • O.P. Hasekamp et al.

    Aerosol properties over the ocean from PARASOL multi-angle photometric measurements

    Journal of Geophysical Research

    (2011)
  • Cited by (80)

    • Polarimetric coordinate validation for merging GRASP and AirMSPI FIREX-AQ observations

      2024, Journal of Quantitative Spectroscopy and Radiative Transfer
    • Scattering and absorption properties of spheroidal soot-sulfate aerosols

      2023, Journal of Quantitative Spectroscopy and Radiative Transfer
    View all citing articles on Scopus
    View full text