Characterization of Ze and LDR of nonspherical and inhomogeneous ice particles for 95-GHz cloud radar:: Its implication to microphysical retrievals

Effect of density, shape, and orientation on radar reflectivity factor (Z(e)) and linear depolarization ratio (LDR) at 95 GHz are investigated by using the discrete dipole approximation (DDA) for ice cloud studies. We consider hexagonal plate, hollow hexagonal column, and hollow bullet rosette in horizontal (2-D) or three-dimensional (3-D) random orientation. We first validate a widely used method to take into account the density and shape effects by the combinational use of Mie theory with the Maxwell-Garnett mixing rule (the MG-Mie method). It is found that the MG-Mie method underestimates Ze and its applicability is limited to sizes smaller than 40 mm. On the basis of the DDA, it is possible to separately treat density, aspect ratio, orientation, and shape. Effect of density turns out to be minor. Orientation and shape are the major controlling factors for Ze especially at effective radius r(eff) > 100 mu m and LDR except for very large sizes where the effect of orientation in LDR diminishes. Comparison between the DDA results and the analytical solution for 3-D Rayleigh spheroids show that LDR in the small size range is characterized by the target boundary and aspect ratio. In the large size range, LDR reveals features of a single target element; for example, LDR of bullet rosette is similar to that of a single branch of the particle. Combinational use of Ze and LDR is effective in microphysics retrieval for LDR < -23 dB. For LDR > -23 dB, additional information such as Doppler velocity is required.