SCATTERING AND DIFFRACTION OF LIGHT BY PARTICLES IN PLANETARY REGOLITHS

The nature of diffraction and scattering by particles in planetary regoliths is discussed. It is pointed out that the Percus–Yevick approximation to the particle radial distribution function is not valid for a regolith, so that structure factor models based on it are incorrect. A simple analytic expression, valid for general shape and size distributions, for the decrease in scattering cross section caused by the proximity of other particles is derived. It is shown that the classical diffraction pattern of an isolated particle does not exist in a regolith, and that the diffracted light is indistinguishable from the incident irradiance. Hence, to an excellent approximation diffracted light may be treated as unscattered. It is shown that back scattering particles are common in nature and that this conclusion is consistent with the theoretical models published in the paper by Mishchenko and Macke (JQSRT, 1997, 57, 767). It is demonstrated that when exact solutions of the radiative transfer equation are used to analyze observational data sets with a small range of phase angles they retrieve particle asymmetry parameters that are erroneously negative, but that if the range is sufficiently large, approximate reflectance models retrieve correctly positive asymmetries. Hence, assertions that approximate models retrieve erroneously negative asymmetries are false. However, most laboratory and planetary observations are poorly suited to constrain the particle scattering properties in the forward direction, so that many planetary regoliths may have particle phase functions with forward scattering lobes, in addition to back scattering ones. It is pointed out that the radiative transfer equation has inherent deficiencies that make it intrinsically incapable of calculating exact descriptions of scattering by a close-packed regolith of large particles and also that the accuracies of typical absolute reflectance measurements are not high. For both reasons it is pointless to be overly concerned with the relative accuracies of exact and approximate regolith scattering models based on the equation of radiative transfer.