Using Dual-Polarized Radar and Dual-Frequency Profiler for DSD Characterization: A Case Study from Darwin, Australia

Abstract

Comparisons are made between the reflectivity Z, median volume diameter D0, and rain rate R from a dual-frequency profiler and the C-band polarimetric radar (C-POL), which are both located near Darwin, Australia. Examples from the premonsoon “buildup” regime and the monsoon (oceanic) regime are used to illustrate the excellent agreement between the dual-profiler retrievals and the polarimetric radar-based retrievals. This work builds on similar works that were limited in scope to shallow tropical showers and predominantly stratiform rain events. The dual-frequency profiler retrievals of D0 and R herein are based on ensemble statistics, whereas the polarimetric radar retrievals are based on algorithms derived by using one season of disdrometer data from Darwin along with scattering simulations. The latest drop shape versus D relation is used as well as the canting angle distribution results obtained from the 80-m fall bridge experiment in the scattering simulations. The scatterplot of D0 from dual-frequency profiler versus Zdr measurements from C-POL is shown to be consistent not only with the theoretical simulations and prior data but also within prior predicted error bars for both stratiform rain as well as convective rain. Based on dual-frequency profiler–retrieved gamma drop size distribution parameters, a new smoothly varying “separator” indexing scheme has been developed that classifies between stratiform and convective rain types, including a continuous “transition” region between the two. This indexing technique has been applied on a number of low-elevation-angle plan position indicator (PPI) sweeps with the C-POL from the two regime examples, to construct “unconditioned” histograms of D0 in stratiform and convective rain (to within the sensitivity of the radar). With reference to the latter, it is demonstrated that the distribution of D0 is different in the buildup example than in the monsoon example, because of the differences in both the microphysical and kinematic features between the two regimes. In particular, (i) the mean D0 is significantly larger in the buildup example than in the monsoon example, irrespective of rain type; (ii) the histogram width (or standard deviation) is much larger for the buildup example than the monsoon example, irrespective of rain type; and (iii) the histogram skewness is negative for the monsoon regime example because of a lack of larger D0 values, whereas the buildup histogram is positively skewed irrespective of rain type.

Comparison of different fittings of drop spectra for rainfall retrievals

“…This result is in agreement with the literature; in fact, (i) the three parameters of the gamma distribution (namely N w , D 0 , and μ) can vary within certain ranges, and (ii) their variation has been studied to characterize the precipitation event, such as convective or stratiform (e.g., [9,55]). Notwithstanding this, the mean models (whose parameter values are reported in Table 2) were used in a quantile-quantile plot (QQ plot) in order to evaluate the general performance of the fitting.…”

Tropical Rain Classification and Estimation of Rain From Z-R (Reflectivity-Rain Rate) Relationships

“…Figure 3 shows a very clear separation of convective and stratiform Fig. 3 represents the separator, defined by Bringi et al in [24], between the two rain types, given by log 10 (N w ) = −1.6D 0 + 6.3 (9) As can be seen from Fig. 3, most of the convective points lie above the separator line with high values of D 0 and log 10 (N w ).…”

Improvement of vertical profiles of raindrop size distribution from micro rain radar using 2D video disdrometer measurements

“…Just to mention an application, errors in precipitation estimation from satellite-borne or ground-based weather radar depend also on vertical gradients of RSD in rain Baldini, 2015). Investigation of RSD vertical profiles has been conducted with several instruments and methods (e.g., Bringi et al 2009, andGiangrande et al 2012). Profiler observations, although limited to the rain column above the instruments, have a higher vertical resolution than that achievable by scanning weather radar that varies depending on the distance between radar sample volume and radar antenna.…”