Development of Dual-polarimetric QPE relations based on Disdrometer measurements in Metro Manila, Philippines

Abstract

AbstractQuantitative precipitation estimates (QPE) can be further improved using estimation algorithms derived from localized raindrop size distribution (DSD) observations. In this study, DSD measurements from two disdrometer stations within Metro Manila during the Southwest monsoon (SWM) period were used to investigate the microphysical properties of rainfall and develop localize dual-polarimetric relations for different radar bands and rainfall types. Observations show that the DSD in Metro Manila is more distributed to larger diameters compared to Southern Luzon and neighboring countries in the Western Pacific. This is reflected by the relatively higher mass-weighted mean diameter (Dm) and smaller shape (μ) and slope (Λ) parameters measured in the region. The average values ofDmand normalized intercept parameter (Nw) in convective rain samples also suggest that convective rains in Metro Manila are highly influenced by both continental and oceanic convective processes. Dual-polarimetric variables simulated using the T-matrix scattering method showed good agreement with disdrometer-derived reflectivity (ZH) values. The 0.5 dB and 0.3° km-1thresholds for the differential reflectivity (ZDR)and specific differential phase (KDP) based on the blended algorithm of Cifelli et al. (2011) and Thompson et al. (2017) are proven to be useful since the utility of the dual-polarimetric variables as rainfall estimators are shown to have dependencies on the radar band and rainfall type. Evaluation of the QPE products with respect to the C-band shows thatR (KDP, ZDR) has the best performance among the dual-pol relations and statistically outperformed the conventional Marshall & Palmer relation [R(ZMP)]. The results show that dual-polarimetric variables such as ZDRand KDPcan better represent the DSD properties compared to one-dimensionalZ, hence providing more accurate QPE products than the conventionalR(Z) relations.