Classification and measurement of precipitation using a dual-polarization radar

Both conventional R(Z) and polarimetric R(K/sub DP/) methods for rainfall measurement tend to overestimate actual rainfall if large raindrops dominate the drop size distribution (DSD). On the other hand, DSDs skewed towards small drops cause these two methods to underestimate rainfall. The R(K/sub DP/) relation, however, is less susceptible to DSD variations, especially in heavy rainfall where large drops are in abundance. But even the R(K/sub DR/) estimator occasionally underestimates actual rainfall in stratiform rain events observed during cold seasons (sometimes more than two times). In this paper the authors suggest a scheme to further reduce an uncertainty in polarimetric rainfall estimation caused by natural variations in drop size distributions.