Electric field soundings and Lightning Mapping Arrays have confirmed the existence of thunderstorms with vertical charge structure that is inverted from the usual polarity. This inverted charge structure can be described grossly as a large upper-level negative charge at roughly the -40°C level, which lies immediately above a large midlevel positive charge, at roughly the -20°C level. This charge structure is often accompanied by a third charge, a smaller negative charge, closer to the freezing level. Cloud-to-ground (CG) flashes lowering positive charge to ground (+CG flashes) instead of the usual negative charge (-CG flashes) make up an unusually large fraction of CG flash activity in these anomalous storms. In this study, we gridded CG flashes from 2004-2014 in order to identify storm cells with high flash rates and having either ≥80% +CG flashes, or having ≥90% -CG flashes. Those with at least 80% +CG flashes were assumed to have an inverted-polarity charge structure, and those with at least 90% -CG flashes were assumed to have a normal-polarity charge structure. We then partitioned the contiguous United States into seven regions, and in each region, we compared the environmental conditions of the inverted-polarity storm cells to those of the normal-polarity storm cells.

We analyzed 17 different environmental parameters, which we divided into 3 categories: dynamic parameters, thermodynamic parameters, and moisture parameters. The dynamic parameters are: 0-3 km shear, 0-6 km shear, 0-3 km storm-relative helicity, and storm-relative wind speed at the equilibrium level. The thermodynamic parameters are: surface equivalent potential temperature, convective available potential energy (CAPE) and normalized CAPE from the level of free convection (LFC) to the equilibrium level (EL), from the LFC to -20°C, and from 0°C to -20°C, convective inhibition, and EL. The moisture parameters are: dew point depression 2 m above ground level, cloud base height, warm cloud depth, and precipitable water. Hypotheses for the mechanism behind the formation of inverted-polarity thunderstorms are based on conditions causing high supercooled liquid water content in the updraft, thereby favoring positive charging of graupel during rebounding collisions with small ice crystals throughout the depth of the mixed-phase region. Therefore, the environmental parameters we studied are those thought to influence liquid water content in the mixed-phase region of the storm.

Our results show that a storm cell’s polarity is determined by no single environmental parameter, but rather by a combination of parameters. Furthermore, different combinations of parameters appear to affect supercooled liquid water content, and hence storm polarity, from region to region. In every region, at least one parameter that was expected to enhance supercooled liquid water content instead had a more favorable median value for normal-polarity cells than for inverted-polarity cells. However, in every region at least four parameters expected to enhance supercooled liquid water content had more favorable values for inverted-polarity cells than for normal-polarity cells. This suggests compensating effects in each region, whereby environmental parameters with values that are unfavorable for maximizing supercooled liquid water content are sufficiently offset by enough environmental parameters with values that are favorable, to produce inverted-polarity storms.