A Possible Origin of I.inear Depolarization Observed at Vertical incidence in Rain

. Recent observations by two different nadir pointing airborne radars with some linear polarization capabilities at times have detected surprisingly large linear clepolarization ratios in convective tropical rain. This depolarization cm be explained if the rain is considered to be a mixture of a group of apparent spheres and another group of drops that are distorted in the horizontal plane perpendicular to the direction of propagation of the incident wave. If confirmed in future observations, this suggests that at times the shapes of a significant portion of the larger raindrops are disturbed apparently beeause of collisions with smaller drops. Since many of the interpretations of radar polarization measurements in rain by ground-based radars presume that the raindrops shapes correspond to those of the well-knowll, so-called ‘equilibrium’ drops, the present observations may require adjustments to some radar polarization algorithms for estimating rainfall rate, for example, if the shape perturbations observed at nadir also apply to measurements along other axes as well. 1. In troduc t ion For many purposes including rainfall measurement using several recently developed radar polarization techniques, it is often assumed that raindrops are essentially quiescent, approximately oblate spheroids with their symmetry axes vertically oriented. Yet it is well known that drop size distributions evolve through the processes of coalescerw, collision and drop breakup, mechanisms likely to disturb such serenity. Presently, however, it is not known whether these perturbations are of sufficient magnitude to affect significantly the quantitative interpretation of radar polarization measurements. While an exploration c]f this latter question is beyond the scope of this brief note, as a first step we attempt to determine here whether possible are at least capable of producing effects detectable in the radar signals. drop perturbations There are already hints that they do at times. For example., on occasion ground-based radars detect surprisingly large vertically polarized signals backscattered from rain even when the polarization of the incident wave is horizontal. At frequencies typial of most weather radars, perfectly quiescent, horizontally which requires dipole moments ,’ incident radar wave. Moreover, aligned raindrops are incapable of producing such depolarization that are canted or tilted with respect to the polarization of the by using circular polarization radar measurements in rain McCormick and Hendry (1974) found the mean ‘canting’ angle (with respect to the plane of polarization) of the raindrops to be near zero but with a standard deviation of about 2°. This appears consistent with physical expectations (Beard and Jarneson 1983). However, although this deviation is small it is significant because the stanclard deviation of the instantaneous canting angle distribution should then be between 6° to 10° if the estimate of each mean is based upon between 9 to 25 independent radar samples. Calculations show that this would be su%cient to