Concurrent measurements from the CSU-CHILL multiparameter Doppler radar, the ONERA VHF lightning interferometer, and the National Lightning Detection Network, obtained during Phase A of the Stratosphere-Troposphere Experiments: Radiation, Aerosols, Ozone (STERAO-A) field project, provided a unique data set with which to study the relationships between convective storm microphysics and associated lightning. Two events have been examined in detail: storms of 10 and 12 July 1996. Both storms underwent major organizational transitions during their lifetimes, identified by sharp changes in total lightning flash rates, dominant cloud-to-ground (CG) flash polarity, or dominant flash type (cloud-to-ground vs. intra-cloud). Both storms also featured relatively high intra-cloud (IC) flash rates.

The 10 July 1996 storm evolved from a multicellular line to an intense unicellular storm. The unicellular stage was marked by a sharp peak in IC flash rate as identified by the interferometer. Cloud-to-ground flash rates were low throughout the storm's lifetime. Small hail was produced during the entire observation period, suggesting storm updraft speeds were significant. The storm of 12 July evolved from an intense multicellular, hail- producing storm to a weaker rainstorm. Before this transition, hail was being produced and the CG flash rates were low. After the transition, hail was no longer being produced and negative CG flash rates were significantly larger. Storm updraft speeds likely weakened during the transition.

These observations are consistent with the elevated-dipole hypothesis to explain low CG production in convective storms, especially if the observed high IC flash rates mostly neutralized any charged core before it descended toward the ground. Alternatively, if significant charging does not occur during wet growth of hail and graupel, both these storms might have produced enough wet-growth ice to prevent the generation of a lower positive charge center that could act to stimulate CG production. However, the radar data, in particular the linear depolarization ratio (LDR) data, suggest that dry growth was more prevalent than wet growth.