Use of a Neurovariational Inversion for Retrieving Oceanic and Atmospheric Constituents from Ocean Color Imagery: A Feasibility Study

C. Jamet LODyC, Paris, and ACRIst, Sophia-Antipolis, France

Search for other papers by C. Jamet in
Current site
Google Scholar
PubMed
Close
,
S. Thiria LODyC, Paris, France

Search for other papers by S. Thiria in
Current site
Google Scholar
PubMed
Close
,
C. Moulin Laboratoire des Sciences du Climat et de l’Environnement, Gif-sur-Yvette, France

Search for other papers by C. Moulin in
Current site
Google Scholar
PubMed
Close
, and
M. Crepon LODyC, Paris, France

Search for other papers by M. Crepon in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

This paper presents a neurovariational method for inverting satellite ocean-color signals. The method is based on a combination of neural networks and classical variational inversion. The radiative transfer equations are modeled by neural networks whose inputs are the oceanic and atmospheric parameters, and outputs the top of the atmosphere reflectance at several wavelengths. The procedure consists in minimizing a quadratic cost function that is the distance between the satellite-observed reflectance and the computed neural-network reflectance, the control parameters being the oceanic and atmospheric parameters.

First, a feasibility experiment using synthetic data is presented to show that chlorophyll-a can be retrieved with an error of 19.7% when the atmospheric parameters are known exactly. Then both atmospheric and oceanic parameters are relaxed. A first guess for the atmospheric parameters was provided by a direct inverse neural network whose inputs are at near-infrared wavelengths. Sensitivity experiments showed that these parameters can be retrieved with an adequate accuracy.

An inversion of a composite SeaWiFS image is presented. Optical thickness and chlorophyll-a both give coherent spatial structures when a background term is added to the cost function. Finally, chlorophyll-a retrievals are compared with SeaWiFS product through in situ data. It shows a better estimation of the chlorophyll-a with the neurovariational inversion for the oligotrophic regions.

Corresponding author address: Prof. Sylvie Thiria, LODyC, tour 45-55, 5eme étage, case courrier 100, Université Pierre et Marie Curie, 4, place Jussieu, 75252 Paris, Cedex 05, France. Email: thiria@lodyc.jussieu.fr

Abstract

This paper presents a neurovariational method for inverting satellite ocean-color signals. The method is based on a combination of neural networks and classical variational inversion. The radiative transfer equations are modeled by neural networks whose inputs are the oceanic and atmospheric parameters, and outputs the top of the atmosphere reflectance at several wavelengths. The procedure consists in minimizing a quadratic cost function that is the distance between the satellite-observed reflectance and the computed neural-network reflectance, the control parameters being the oceanic and atmospheric parameters.

First, a feasibility experiment using synthetic data is presented to show that chlorophyll-a can be retrieved with an error of 19.7% when the atmospheric parameters are known exactly. Then both atmospheric and oceanic parameters are relaxed. A first guess for the atmospheric parameters was provided by a direct inverse neural network whose inputs are at near-infrared wavelengths. Sensitivity experiments showed that these parameters can be retrieved with an adequate accuracy.

An inversion of a composite SeaWiFS image is presented. Optical thickness and chlorophyll-a both give coherent spatial structures when a background term is added to the cost function. Finally, chlorophyll-a retrievals are compared with SeaWiFS product through in situ data. It shows a better estimation of the chlorophyll-a with the neurovariational inversion for the oligotrophic regions.

Corresponding author address: Prof. Sylvie Thiria, LODyC, tour 45-55, 5eme étage, case courrier 100, Université Pierre et Marie Curie, 4, place Jussieu, 75252 Paris, Cedex 05, France. Email: thiria@lodyc.jussieu.fr

Save
  • Badran, F., Thiria S. , and Crepon M. , 1991: Wind ambiguity removal by the use of neural networks techniques. J. Geophys. Res., 96C , 2052120529.

    • Search Google Scholar
    • Export Citation
  • Bishop, C. M., 1995: Neural Networks for Pattern Recognition. Oxford University Press, 482 pp.

  • Bricaud, A., Bosc E. , and Antoine D. , 2002: Algal biomass and sea surface temperature in the Mediterranean basin intercomparison of data from various satellite sensors, and implications for primary production estimates. Remote Sens. Environ., 81 , 163178.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chomko, R., and Gordon H. R. , 1998: Atmospheric correction of ocean color imagery: Use of a Junge power-law size distribution with variable refractive index to handle aerosol absorption. Appl. Opt., 37 , 55605572.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Claustre, H., and Coauthors, 2002: Is desert dust making oligotrophic water greener? Geophys. Res. Lett., 29 .1469, doi:10.1029/2001GL014056.

    • Search Google Scholar
    • Export Citation
  • Cybenko, G., 1989: Approximation by superposition of a sigmoidal function. Math. Control Signal Syst., 2 , 303313.

  • Deschamps, P. Y., Herman M. , and Tanre D. , 1983: Modeling of the atmospheric effects and its application of the remote sensing of ocean color. Appl. Opt., 22 , 37513758.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • D’Ortenzio, F., Marullo S. , Ragni M. , Ribera d’Alcalã M. , and Santoleri R. , 2002: Validation of empirical SeaWiFS algorithms for chlorophyll-a retrieval in the Mediterranean Sea: A case study for oligotrophic seas. Remote Sens. Environ., 82 , 7994.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gordon, H. R., 1997: Atmospheric correction of ocean color imagery in the Earth Observing System era. J. Geophys. Res., 102 , 1708117106.

  • Gordon, H. R., and Wang M. , 1994a: Retrieval of water-leaving radiance and aerosol optical thickness over the oceans with SeaWiFS: A preliminary algorithm. Appl. Opt., 33 , 443452.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gordon, H. R., and Wang M. , 1994b: Influence of oceanic whitecaps on atmospheric correction of ocean-color sensors. Appl. Opt., 33 , 77547763.

  • Gordon, H. R., Clark D. K. , Brown J. W. , Brown O. B. , Evans R. H. , and Broenkow W. W. , 1983: Phytoplankton pigment concentrations in the Middle Atlantic Bight: Comparison between ship determinations and Coastal Zone Color Scanner. Appl. Opt., 22 , 2036.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gordon, H. R., Brown J. W. , and Evans R. H. , 1988a: Exact Rayleigh scattering calculations for use of the Nimbus 7 coastal zone color scanner. Appl. Opt., 26 , 21112122.

    • Search Google Scholar
    • Export Citation
  • Gordon, H. R., Brown O. B. , Evans R. H. , Brown J. W. , Smith R. C. , Baker K. S. , and Clark D. K. , 1988b: A semi-analytic radiance model of ocean color. J. Geophys. Res., 93 , 1090910924.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gordon, H. R., Du T. , and Zhang T. , 1997: Remote sensing of ocean color imagery and aerosol properties: Resolving the issue of aerosol absorption. Appl. Opt., 36 , 86708684.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hornik, K., Stinchcomb M. , and Muller-Karger F. E. , 1989: Multi-layer feedforward networks are universal approximators. Neural Networks, 2 , 359366.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Junge, C. E., 1958: Atmospheric Chemistry Advances in Geophysics. Vol. 4, Academic Press, 1–108.

  • McClain, C. R., Ainsworth E. J. , Barnes R. A. , Eplee R. E. Jr., Patt F. S. , Robinson W. D. , Wang M. , and Bailey S. W. , 2000: SeaWiFS postlaunch calibration and validation analyses, Part 1. NASA Tech. Memo 206982, 85 pp.

  • O’Reilly, J. R., and Coauthors, 1998: Ocean color chlorophyll algorithms for SeaWiFS. J. Geophys. Res., 103 , 2493724953.

  • Pinkus, A., 1999: Approximation theory of the MLP model in neural networks. Acta Numer., 143–195.

  • Shettle, E. P., and Fenn R. W. , 1979: Models of the atmospheric aerosols and their optical properties. Air Force Geophysical Laboratory (AFGL), Rep. AFGL-TR-79-0214, Hanscomb Airforce Base, MA.

  • Thiria, S., Mejia C. , Badran F. , and Crepon M. , 1993: A neural network approach for modelling transfer functions: Application for wind retrieval from spaceborne scatterometer data. J. Geophys. Res., 98 , 2282722841.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yang, H., and Gordon H. R. , 1997: Remote sensing of ocean color: Assessment of the water-leaving radiance bidirectional effects on the atmospheric diffuse transmittance. Appl. Opt., 36 , 78877897.

    • Crossref
    • Search Google Scholar
    • Export Citation
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 432 141 7
PDF Downloads 120 39 1