The Fall Speed Variability of Similarly Sized Ice Particle Aggregates

Schmitt, Carl G.; Sulia, Kara; Lebo, Zachary J.; Heymsfield, Andrew J.; Przybyo, Vanessa; Connolly, Paul

doi:10.1175/jamc-d-18-0291.1

Title: The Fall Speed Variability of Similarly Sized Ice Particle Aggregates

Journal Article · 23 July 2019 · Journal of Applied Meteorology and Climatology

DOI: https://doi.org/10.1175/jamc-d-18-0291.1 · OSTI ID:1544580

Schmitt, Carl G. ^[1]; Sulia, Kara ^[2]; Lebo, Zachary J. ^[3]; Heymsfield, Andrew J. ^[1]; Przybyo, Vanessa ^[2]; Connolly, Paul ^[4]

National Center for Atmospheric Research (NCAR), Boulder, CO (United States)
State Univ. of New York (SUNY), Albany, NY (United States)
Univ. of Wyoming, Laramie, WY (United States)
Univ. of Manchester (United Kingdom)

The terminal velocity (V_t) of ice hydrometeors is of high importance to atmospheric modeling. V_t is governed by the physical characteristics of a hydrometeor, including mass and projected area, as well as environmental conditions. When liquid hydrometeors coalesce to form larger hydrometeors, the resulting hydrometeor can readily be characterized by its spherical or near-spherical shape. For ice hydrometeors, it is more complicated because of the variability of ice shapes possible in the atmosphere as well as the inherent randomness in the aggregation process, which leads to highly variable characteristics. The abundance of atmospheric processes affecting ice particle dimensional characteristics creates potential for highly variable V_tfor ice particles that are predicted or measured to be of the “same size.” In this article we explore the variability of ice hydrometeor V_t both theoretically and through the use of experimental observations. Theoretically, the variability in V_t is investigated by analyzing the microphysical characteristics of randomly aggregated hexagonal shapes. The modeled dimensional characteristics are then compared to aircraft probe measurements to constrain the variability in atmospheric ice hydrometeor V_t. Results show that the spread in V_t can be represented with Gaussian distributions relative to a mean. Variability expressed as the full width at half maximum of the normalized Gaussian probability distribution function is around 20%, with somewhat higher values associated with larger particle sizes and warmer temperatures. Finally, field campaigns where mostly convective clouds were sampled displayed low variability, while Arctic and midlatitude winter campaigns showed broader V_t spectra.