Evaluation of Gamma Raindrop Size Distribution Assumption through Comparison of Rain Rates of Measured and Radar-Equivalent Gamma DSD

To date, one of the most widely used parametric forms for modeling raindrop size distribution (DSD) is the\udthree-parameter gamma. The aim of this paper is to analyze the error of assuming such parametric form to\udmodel the natural DSDs. To achieve this goal, a methodology is set up to compare the rain rate obtained from\uda disdrometer-measured drop size distribution with the rain rate of a gamma drop size distribution that\udproduces the same triplets of dual-polarization radar measurements, namely reflectivity factor, differential\udreflectivity, and specific differential phase shift. In such a way, any differences between the values of the two\udrain rates will provide information about how well the gamma distribution fits the measured precipitation. The\uddifference between rain rates is analyzed in terms of normalized standard error and normalized bias using\uddifferent radar frequencies, drop shape–size relations, and disdrometer integration time. The study is performed\udusing four datasets of DSDs collected by two-dimensional video disdrometers deployed in Huntsville (Alabama)\udand in three different prelaunch campaigns of the NASA–Japan Aerospace Exploration Agency\ud(JAXA) Global Precipitation Measurement (GPM) ground validation program including the Hydrological\udCycle inMediterraneanExperiment (HyMeX) special observation period (SOP) 1 field campaign inRome. The\udresults show that differences in rain rates of the disdrometer DSD and the gamma DSD determining the same\uddual-polarization radar measurements exist and exceed those related to the methodology itself and to the disdrometer\udsampling error, supporting the finding that there is an error associatedwith the gammaDSDassumption