Evaluation of a technique for radar identification of large hail across the upper midwest and central plains of the United States

Radar data were analyzed for severe thunderstorms that produced severe hail (>19 mm diameter) across the central and northern plains of the United States during the 2001–04 convective seasons. Results showed a strongly linear relationship between the 50-dBZ echo height and the height of the melting level—so strong that a severe hail warning methodology was successfully deployed at the National Weather Service Warning and Forecast Offices in North Dakota and Iowa. Specifically, for each of 183 severe hailstorms, the 50-dBZ echo height near the hail event time was plotted against the depth of the environmental melting level. Linear regression revealed a coefficient of determination of 0.86, which suggested a strong linear relationship between the 50-dBZ echo height and the melting-level depth for the severe hail producing storms. As the height of the melting level increased, the expected 50-dBZ echo height increased. A severe warning criterion for large hail was based on the 10th percentile from the linear regression, producing a probability of detection of 90% and a false alarm rate of 22%. Additional analysis found that the 50-dBZ echo-height technique performs very well for weakly to moderately sheared thunderstorm environments. However, for strongly sheared, supercell-type environments, signatures such as weak-echo regions and three-body scatter spikes led to more rapid severe thunderstorm detection in many cases.