Comparison of dual-polarization radar-derived and disdrometer drop-size distributions with S- and X- band radars

Abstract

There have been many studies on the evaluations of drop-size distributions and the parameters that affect these distributions, however, few, if any, have directly compared the relationship between the radar-derived parameters and those parameters that are disdrometer-derived. This study focuses on many different features of thunderstorms that changes the structure of the drop-size distribution (DSD) including: Horizontal reflectivity (ZH), differential reflectivity (ZDR), median drop diameter (D0), the shape parameter of the gamma-distributed DSD ([mu]), and the slope parameter of the gamma-distributed DSD (lambda). This work compares data collected by two disdrometers (OTT PARSIVEL and the Campbell Scientific Present Weather Sensor 100) against DSD parameters derived from dual-polarization radar observations. Using the Warning Decision Support System-Integrated Information (WDSS-II), radar data was merged at 1-km resolution to account for the movement of the precipitation systems before comparing to the 10-minute disdrometer data intervals. It was found that to accurately estimate DSDs from the perspective of using a weather radar, a larger precipitation event is needed. At the beginning and end of a precipitation event the difference between the radar retrieved values of D0, [mu], and [lambda] and those sampled by the disdrometer were much greater than during the middle of the event. Throughout the majority of the cases, the radar-derived reflectivity values were consistently lower than those collected by the disdrometers.