Predicting hail size using model vertical velocities

Abstract

A simple test hail growth model is created in order to compare hailstone sizes from model vertical velocities and calculated updrafts from a simple cloud model using forecasted soundings. The models used MM5 model data coinciding with severe hail events collected from the Central and Southern Plains from March to May 2006 and 2007. In the test model, four different starting embryo sizes were interjected into four separate hail growth modes: dry growth and wet growth using model vertical velocities and dry and wet growths using calculated updrafts. These embryos were placed at four different beginning vertical levels resulting in 64 possible ending hailstone sizes. Examination of the 804 hail events revealed the potential usefulness of model vertical velocities in generating severe hailstones. In particular, using dry growth, the model vertical velocities produced 727 severe hailstones compare to 661 produced by dry growth using the thermodynamically calculated updraft. Model vertical velocities also proved more accurate than updrafts, resulting in an average error of 0.417 compare to 0.788 under dry growth conditions. Calculated updrafts were still required to generate the large severe hail that model vertical velocities could not produce.