Hail Growth in an Oklahoma Multicell Storm

Conrad L. Ziegler National Severe Storms Laboratory, Norman, OK 73069

Search for other papers by Conrad L. Ziegler in
Current site
Google Scholar
PubMed
Close
,
Peter S. Ray National Severe Storms Laboratory, Norman, OK 73069

Search for other papers by Peter S. Ray in
Current site
Google Scholar
PubMed
Close
, and
Nancy C. Knight National Severe Storms Laboratory, Norman, OK 73069

Search for other papers by Nancy C. Knight in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Hail growth in an Oklahoma multicellular storm is studied using a numerical model of hailstone growth and decay and dual-Doppler derived wind fields. Hail was collected at the time of the Doppler radar data collection which provided input for computation of the modeled trajectories. A unique feature of this investigation is the subsequent comparison of model hail trajectories with deuterium derived trajectories obtained from the hail samples. Formation of large hail is found to be almost entirely due to injection of embryos into the major storm updraft from the upwind side, with subsequent growth occurring during repeated vertical excursions through the prime growth layer between 7 and 8 km. Primary embryo source regions are a feeder cell and the precipitation debris region between the feeder and main cells. Qualitative comparisons between observed and modeled hailstones falling near the collection site reveal strong similarities, particularly with respect to ambient temperature during ice formation and layer structure. Horizontal advection of hail across the updraft during growth is typical, so that particle recirculation in a singe updraft is unimportant for hail growth. Observed hail size distributions are related to the distributions of modeled hailstones at the ground. Either a modal or “leveling-off” tendency is evident in each of the hail samples, whose shapes agree qualitatively with the distribution of numerically simulated large hail falling in the vicinity of the storm core at the surface. The gamma function is found to generally provide a better fit to the sample distributions than the Marshall-Palmer function.

Abstract

Hail growth in an Oklahoma multicellular storm is studied using a numerical model of hailstone growth and decay and dual-Doppler derived wind fields. Hail was collected at the time of the Doppler radar data collection which provided input for computation of the modeled trajectories. A unique feature of this investigation is the subsequent comparison of model hail trajectories with deuterium derived trajectories obtained from the hail samples. Formation of large hail is found to be almost entirely due to injection of embryos into the major storm updraft from the upwind side, with subsequent growth occurring during repeated vertical excursions through the prime growth layer between 7 and 8 km. Primary embryo source regions are a feeder cell and the precipitation debris region between the feeder and main cells. Qualitative comparisons between observed and modeled hailstones falling near the collection site reveal strong similarities, particularly with respect to ambient temperature during ice formation and layer structure. Horizontal advection of hail across the updraft during growth is typical, so that particle recirculation in a singe updraft is unimportant for hail growth. Observed hail size distributions are related to the distributions of modeled hailstones at the ground. Either a modal or “leveling-off” tendency is evident in each of the hail samples, whose shapes agree qualitatively with the distribution of numerically simulated large hail falling in the vicinity of the storm core at the surface. The gamma function is found to generally provide a better fit to the sample distributions than the Marshall-Palmer function.

Save