• Radiative transfer modeling in the coupled atmosphere-ocean system and its application to the remote sensing of ocean color imagery

      Yan, Banghua; Stamnes, Knut; Nielsen, Hans; Watkins, Brenton; Olson, John (2001-08)
      Ocean color is the radiance emanating from the ocean due to scattering by chlorophyll pigments and particles of organic and inorganic origin. Thus, it contains information about chlorophyll concentrations which can be used to estimate primary productivity. Observations of ocean color from space can be used to monitor the variability in marine primary productivity, thereby permitting a quantum leap in our understanding of oceanographic processes from regional to global scales. Satellite remote sensing of ocean color requires accurate removal of the contribution by atmospheric molecules and aerosols to the radiance measured at the top of the atmosphere (TOA). This removal process is called 'atmospheric correction.' Since about 90% of the radiance received by the satellitee sensor comes from the atmosphere, accurate removal of this portion is very important. A prerequisite for accurate atmospheric correction is accurate and reliable simulation of the transport of radiation in the atmosphere-ocean system. This thesis focuses on this radiative transfer process, and investigates the impact of particles in the atmosphere (aerosols) and ocean (oceanic chlorophylls and air bubbles) on our ability to remove the atmospheric contribution from the received signal. To explore these issues, a comprehensive radiative transfer model for the coupled atmosphere-ocean system is used to simulate the radiative transfer process and provide a physically sound link between surface-based measurements of oceanic and atmospheric parameters and radiances observed by satellite-deployed ocean color sensors. This model has been upgraded to provide accurate radiances in arbitrary directions as required to analyze satellite data. The model is then applied to quantify the uncertainties associated with several commonly made assumptions invoked in atmospheric correction algorithms. Since Atmospheric aerosols consist of a mixture of absorbing and non-absorbing components that may or may not be soluble, it becomes a challenging task to model the radiative effects of these particles. It is shown that the contribution of these particles to the TOA radiance depends on the assumptions made concerning how these particles mix and grow in a humid environment. This makes atmospheric correction a very difficult undertaking. Air bubbles in the ocean created by breaking waves give rise to scattered light. Unless this contribution to the radiance leaving the ocean is correctly accounted for, it would be mistakenly attributed to chlorophyll pigments. Thus, the findings in this thesis make an important contribution to the development of an adequate radiative transfer model for the coupled atmosphere-ocean system required for development and assessment of algorithms for atmospheric correction of ocean color imagery.