Reevaluation of a technique for radar identification of large hail in western and central Wyoming

Donavon and Jungbluth (2007) analyzed radar data associated with severe thunderstorms that produced severe hail (>19 mm or 3/4 inch diameter) across portions of the central and northern plains for four convective seasons from 2001 through 2004. Their results showed a strong relationship between the 50-dBZ echo height and the height of the melting level; subsequently, a severe hail warning methodology was employed at National Weather Service (NWS) offices in North Dakota and Iowa. A similar study was conducted for severe thunderstorms observed in the NWS Riverton, WY county warning area (CWA) for five convective seasons from 2004 through 2009. The study was conducted to determine if the same evaluation conducted in the central portion of the country would be valid in areas of complex terrain, as was questioned by the original authors. In addition, a methodology was desired for the new criteria of severe hail (>25 mm or 1 inch diameter). A total of 153 severe hail reports (>19 mm or 3/4 inch diameter) were evaluated during this period. Radar data was interrogated to identify the approximate 50-dBZ echo height, the 60-dBZ echo height, as well as the storm top divergence associated with the hail-producing storms. Proximity sounding data was obtained from either RUC or NAM BUFR soundings in BUFKIT; the melting level (0°C) and -20°C level were obtained for each report from the nearest sounding location.

A scatterplot (melting level vs. 50-dBZ echo height) of all 153 hail reports from the RIW CWA yields similar results to Figure 6 from Donavon and Jungbluth (2007), although there appears to be less correlation with the data over complex terrain. As suggested previously, the melting levels over complex terrain are lower than those observed across the Midwest US. In general, the higher the 50-dBZ echo height, the larger the hail size observed. The greater variability may in part be explained by the potential observing density in the regions. According to the latest population estimate obtained from the Internet, the state of Iowa has a population density of 53.4 per square mile, whereas Wyoming (the least populated state in the country and the 10th largest) has a population density of 5.6 per square mile – an order of magnitude difference! Quite a few of the reports in the RIW CWA came from observers who were “grazed” by the storm rather than having the core pass over them. Splitting the data by season (melting level either above or equal to 8000 feet AGL compared to melting level below 8000 feet AGL) indicates better correlation during the warm season. Specifically, if the melting level is at or above 8000 feet AGL and the 50-dBZ echo height is at or above 22000 feet AGL, there is a high probability of severe hail.