This bb/b value describes the probability of scattering into the backward direction during a single scattering process. It would seem that, because the backscattering coefficient is used explicitly in the RSR approximation (1), the angular shape of the phase function is already accounted for. However, there are an infinite number of possible phase function shapes that correspond to the same backscattering ratio. Of course, only a limited subset of them selleck are actually relevant to oceanic radiative transfer calculations, but it is important

to check how much variability in the calculated RSR value may result from the choice of a phase function even with a fixed bb/b value. This possible source of the radiative transfer calculation error of RSR was studied by Chami et al. (2006) (this study is henceforth referred to as CMLK06), who compared the water leaving radiance for experimentally derived and Fournier-Forand (FF) parameterized phase functions

with identical backscattering ratios using the Mobley et al. (2002) parameterization (and building on the results of that paper, which also discussed the effect of phase function shape on computed light-field quantities). However, because there is more than one way to parameterize FF phase functions for identical Sotrastaurin in vivo scattering and absorption coefficients (including the backscattering ratio) ( Freda & Piskozub 2007), we decided to compare the effect of choosing a different FF function for a given bb/b value on calculated remote sensing reflectance. In addition to that, we also included the average Petzold function and Henyey-Greenstein functions,

as they are often used in radiative transfer modelling. This approach means that any discrepancies in calculated RSR values found in our study are independent of the ones previously reported by Chami et al. (2006), broadening the range of potential scattering phase functions for a given bb/b. The RSR was calculated with a 3D Monte Carlo radiative transfer algorithm, originally created to study self-shading instrumentation measurement this website artifacts (Piskozub, 1994 and Piskozub et al., 2000) but subsequently used in ocean radiative transfer studies (Flatau et al., 1999 and Piskozub et al., 2008). The algorithm makes it possible to calculate the RSR separately for photons leaving the marine environment and for photons, which as a result of reflection from a roughened sea surface, increase the value of the reflectance. These two parts of the RSR will be called the water leaving radiance reflectance and the reflective part of the RSR.