The history of particle scattering was recently reviewed in a special issue of Applied Optics [November 20, 1991].
One of the outstanding issues in atmospheric radiation is the need for a clearer understanding of the impact the properties of particle shape and, in an associated way, particle composition, on particle scattering and absorption. This is particularly relevant to studies of both ice crystal clouds and aerosol--radiation interactions [ Liou and Takano, 1993; Lacis and Mishchenko, 1994]. Analytic methods that permit solutions to Maxwell's equations are limited to a small class of simple geometries and these methods have been worked out long ago. A technique that extends the theory of scattering by spheres, that of scattering by clusters of spheres [ Fuller, 1991, 1994, 1995a] and spherical inclusions [ Fuller, 1995b] is reaching some level of maturity [e.g. Muinonen and Lumme, 1991; Ngo and Pinnick, 1994].
More recently, numerical methods have advanced and it is now possible to study complex large particles in the geometric optics limit via ray tracing [e.g Macke, 1993; Liou and Takan, 1993] and smaller particles using finite grid methods such as the finite--difference--time--domain approach introduced in the 1970's [e.g., Kunz and Luebbers, 1993] and the discrete dipole approximation (consult the text of Stephens [1994] for a brief description of this approximation) [e.g., Draine and Flatau, 1994]. The latter class of numerical method is generally confined to particles of size smaller or comparable to the wavelength of the radiation in question. Validation of these techniques is an ongoing line of enquiry and results have been favorably compared with solutions for particles other than spheres [e.g., Flatau et al., 1994]. The dipole method is popular and although it is limited in its application, such as to scattering of microwave radiation by cloud particles [ Evans and Stephens, 1994a] or small aerosol particles in the visible
[ West and Smith, 1991] or at infrared frequencies the method is beginning to contribute towards an understanding of how irregular particles scatter radiation. Examples of applications of this work are found in the radar backscattering study Dungey and Bohren [1993], and in the microwave radiative transfer study of Evans and Stephens [1994b]. The dipole method is also being extended to include contributions by magnetic dipoles [e.g., Mulholland et al., 1994] and applied to scattering of agglomerate particles.
These finite difference methods are also being tested using laboratory data and laboratory research on particle scattering may be on the upswing [ Kuik et al., 1991]. For example, the study of Fuller et al. [1994] presents results of laboratory microwave measurements of scattering by a model hexagonal ice crystal. The aim of these measurements is to test these numerical techniques. With this kind of innovative test of these methods, we can expect significant progress on these numerical methods with an application to even larger particles than considered at present. This research, together with basic laboratory scattering measurements will progress our understanding of scattering by irregular particles.