Microphysical Characteristics of an Asymmetric Eyewall in Major Hurricane Harvey (2017)

Microphysical and kinematic structures of major Hurricane Harvey's (2017) asymmetric eyewall are analyzed from ground-based polarimetric and airborne Doppler radars. New polarimetric observations of differential reflectivity (Z(DR)) and specific differential phase (K-DP) show asymmetric wavenumber-1 patterns associated with vertical wind shear (VWS) but were shifted azimuthally with respect to the reflectivity (Z(H)) asymmetry. A Z(DR) column was found upwind of the Z(H) maximum in a region with strong updrafts estimated from multi-Doppler synthesis, with higher values of K-DP found cyclonically downwind. Retrieved raindrop size distributions show that azimuthal variations of size and number concentration were determined by both the VWS and the size sorting process. The diameter of raindrops decreases, while the number concentration increases cyclonically downwind of VWS-induced updrafts due to the differential terminal fall speed of raindrops and strong rotational flow at major hurricane wind speeds. Plain Language Summary Hurricane forecasts are highly sensitive to the representation of raindrop properties in numerical weather prediction models. Hurricane Harvey (2017) was the first major hurricane of category 4 intensity to make U.S. landfall since the recent upgrade of the U.S. weather radar network to dual-polarization technology that allows for better characterization of the shape, size, and number of raindrops in hurricanes. These new observations indicate substantial variation in the raindrop size distribution around Harvey's intense, asymmetric eyewall. Through additional analysis of data collected by the airborne Hurricane Hunters, we find that the largest raindrops are located where upward motion occurs due to interactions of environmental wind shear and strong rotational winds. The diameter of raindrops decreases, but the number of raindrops increases downwind of the updrafts around the eyewall. This new analysis can be used to evaluate and improve numerical models used in hurricane forecasting.