Summary and Conclusions

Abstract

The initial problem addressed in ocean color remote sensing is the determination of the concentrations of various ocean constituents (phytoplankton, total seston, etc.) in surface waters through measurement of the spectral radiance L_T(λ) leaving the Earth’s atmosphere. The physics of this problem is well understood. That is, the constituents influence the ocean’s optical properties (a(λ) and β(γ,λ)), and given the optical properties of the ocean and the atmosphere (and their distribution, respectively, with depth and altitude), and other environmental factors such as surface wind velocity, the spectral radiance L_T(λ) at the sensor can be determined by solving the radiative transfer equation. Solving the inverse problem, the determination of the optical properties a(λ) and β(γ,λ) and hence the constituent concentrations, from measurements of L_T(λ) is impossible without the addition of some simplifying facts or assumptions. One considerable simplification is the decoupling of the atmospheric and oceanic parts of the problem, and the fact that the ocean’s contribution to L^T(λ) to a good approximation depends only on the irradiance ration R(λ).