The National Weather Service defines hail as severe when it has a diameter of 19 mm (3/4 in.) or larger. The Storm Prediction maintains a database that contains all reported severe hailstorms that were reported over the contiguous United States since 1955. These data include the latitude/longitude, date, time, size, and estimates of the damage if available. Although these data are not perfect, analysis of these data allows one to estimate the regional frequency and impacts of large hail.

The number of reported severe hail events has increased substantially over the past half century. This increase in the number of reports is the result of a number of non-meteorological factors. Potential causative factors include the implementation of the national warning verification program, the enhanced development of storm spotter networks, the deployment of the national NEXRAD radar network, and population increase that have resulted in the urbanization of previously rural areas all effect the number of hail storms reported. However, these effects have not been uniform across the nation and the resulting data base contains biases and inconsistencies that vary in both time and space.

The distribution of hail occurrences by size is examined. It is shown that hail is typically only reported for selected sizes related to the presumed size of common objects such as coins of various denominations. Charts of the occurrence frequency of severe thunderstorm winds in the United States are presented. The data is further subdivided by month and by size. Resulting geographic and seasonal patterns in the data are explained through meteorological considerations.

This analysis shows that even with the problems inherent in the data, it contains valuable information about the nature of large hail events across the country.