Investigation of DCS cloud-rain drop size distributions and precipitation through an integrative analysis of aircraft in situ, NEXRAD reflectivity and surface disdrometer measurements

 

Authors

Jingyu Wang —
Xiquan Dong — University of Arizona
Baike Xi — University of Arizona
Andrew Heymsfield — National Center for Atmospheric Research (NCAR)

Category

Deep convective clouds, including aerosol interactions

Description

Eight cases of deep convective systems (DCSs) with radar echo tops higher than 500 hPa during the Midlatitude Continental Convective Clouds Experiment (MC3E) were selected to investigate the rain drop size distribution (DSDs) of the liquid-phase layer (T > 3oC) over the ARM SGP site. A full spectrum of rain DSDs was constructed through a combination of 2DC and HVPS measurements. A total of 1126 five-second aircraft in situ measured DSDs, covering a horizontal trajectory of ~ 103 km, were fitted with Gamma size distribution to investigate the shape-slope (μ-λ) relationship of aloft DSDs. The fitted μ - λ relationship was found similar to the one previously constructed from surface disdrometer measurements, but with narrower μ and λ value ranges. The similar μ-λ relationships but different value ranges generated from two independent platforms (aircraft probes vs. disdrometer) at different elevations (aloft vs. surface) may represent the real nature of DSD shape information in clouds and at the surface. For the rain rate estimation, exponential function was adopted, and the new intercept-slope (N0-λ) was generated and parameterized as a function of radar reflectivity according to its theoretical definition. Contrary to the well-known power-law Z-R relationships where a constant N0 was assumed, the new scheme incorporated the dependencies between N0 and λ, and exhibited excellent agreement in rain rate estimation compared to the surface rain gauges measurements. However, the NEXRAD Q2 precipitation products based on power-law Z-R relationships demonstrated severe overestimation for heavy rain events, which revealed the necessity of incorporating N0-λ relationship in DSD parameterization.