A semianalytic radiance model of ocean color

Abstract

A semianalytical radiance model is developed which predicts the upwelled spectral radiance at the sea surface as a function of the phytoplankton pigment concentration for Morel Case 1 waters. The model is in good agreement with experimental measurements carried out in waters which were not included in the data base used to derive it. It suggests that the observed variability in the radiance is due to variations in the backscattering of plankton and the associated detrital material. The model is extended to include other material in the water, such as dissolved organic material, referred to as yellow substances, and detached coccoliths from coccolithophorids, e.g., Emiliana huxleyi. Potential applications include an improved bio‐optical algorithm for the retrieval of pigment concentrations from satellite imagery in the presence of interference from detached coccoliths and an improved atmospheric correction for satellite imagery. The model also serves to identify and to interpret deviations from Case 1 waters.

Estimation of chlorophyll-a concentration in turbid productive waters using airborne hyperspectral data

“…As a result, chl-a is not the sole factor dominating the optical properties of these waters. Hence, the standard open-ocean algorithms that rely on reflectances in the blue and green spectral regions (e.g., Gordon and Morel, 1983;Gordon et al, 1988;O'Reilly et al, 1998;O'Reilly et al, 2000) do not yield accurate estimates of chl-a concentration in these waters (e.g., Carder et al, 2004;Darecki and Stramski, 2004;.…”

Biological Oceanography by Remote Sensing

“…2,15 Theoretical modeling suggests that R l in turn depends on the absorption a l and backscattering b b l coefficients (note that the scattering coefficient 2,20 In addition, the absorption and scattering coefficients can be decomposed into the contributions due to the water itself and to each of the constituents in the water, and so related to the chlorophyll concentration. On this basis it is possible to develop what are known as semianalytical models of ocean color, where the behavior of a l and b b l is established through a combination of modeling and measurements.…”

“…In principle, the semianalytical approach should provide improved information as it should reduce some of the uncertainties associated with the empirical approach. 20 In either case accurate measurements of the water-leaving radiance in the sensor bands need to be obtained and this requires that the data be corrected for atmospheric effects, which are discussed next.…”

“…15 This is a simple measurement that compares the waterleaving radiance in the blue to that in the green part of the visible spectrum, the ratio decreasing with increasing concentration of pigments. 20 For future missions both empirical and semianalytical approaches to estimating pigment concentration will be used (see, for example, Esaias et al 9 ). In principle, the semianalytical approach should provide improved information as it should reduce some of the uncertainties associated with the empirical approach.…”

“…On this basis it is possible to develop what are known as semianalytical models of ocean color, where the behavior of a l and b b l is established through a combination of modeling and measurements. 20 For CZCS the standard algorithms were in fact based on an empirical approach, where the ratio of the water-leaving radiance in two bands was compared to values of the pigment concentration measured in situ (section 3.2). 15 This is a simple measurement that compares the waterleaving radiance in the blue to that in the green part of the visible spectrum, the ratio decreasing with increasing concentration of pigments.…”