Similarities in the Initiation of Upward Positive and Downward Negative Lightning Flashes

TL;DR

This study addresses the problem of lightning initiation by testing whether upward negative stepped leader pulses (UNL) share the same physical processes as downward negative preliminary breakdown pulses (PBPs). It analyzes simultaneous tower-base current and remote electric-field measurements from the Säntis facility, complemented by high-speed imagery, to classify UPF pulses into Category A (longer, ICC-correlated) and Category B (very short, weakly current-coupled) and compare them with downward PBPs (Classical and Narrow). The results show strong waveform and timing similarities: Category A pulses resemble Classical PBPs, while Category B pulses resemble Narrow PBPs, with at least one case where a Category B pulse is produced by a downward recoil leader; field–current analysis yields a robust linear relationship for Classical PBPs ($R^2 \\approx 0.95$) after topographic correction ($k \\approx 1.8$), and a measured $A/I_ ext{PBP}$ around $0.23 \,\pm\,0.05$ V m$^{-1}$ kA$^{-1}$, close to model predictions ($0.163 \,\pm\,0.025$ V m$^{-1}$ kA$^{-1}$). These findings support a shared initiation physics across upward and downward negative leaders and suggest that detailed observations of upward lightning can constrain and inform downward initiation models, with practical implications for lightning safety and modeling; future work should expand the dataset, incorporate VHF interferometry, and pursue 3D reconstructions to further validate these parallels.

Abstract

This study examines the relationship between upward negative stepped leader pulses in upward positive lightning and preliminary breakdown pulses (PBPs) in downward negative lightning discharges. Through analysis of simultaneous channel-base current and electric field data from the Säntis tower in Switzerland, we found notable similarities between the "Category A" and "B" pulses associated with the initial continuous current of upward negative leaders, and the "Classical" and "Narrow" PBPs observed in downward negative flashes. Statistical comparisons reveal correlations between electric field and current parameters for Category A pulses, supporting the field-current relationship for preliminary breakdown proposed in previous studies. These results suggest that similar physical processes may be involved in the early stages of negative leader development in both upward and downward lightning, providing valuable insights into lightning initiation that could not be obtained from conventional field measurements alone. Furthermore, high-speed camera footage revealed that Category B pulses can be produced by a downward-propagating recoil leader. As a whole, these findings demonstrate that detailed observations of upward lightning can offer valuable insight into the complex processes underlying lightning initiation and propagation.

Figures (10)

• Figure 1: Depiction of the Mt. Säntis research stations relevant to this study.
• Figure 2: Current and electric field waveforms associated with the upward positive flash #8 listed in Table \ref{['tab:flashes']}. (a) Overall waveforms. (b) A bipolar Category A pulse train occurring in the early stage of the ICC. (c) Current and E-field waveforms corresponding to an individual Category A pulse. (d) Two typical Category B pulses observed at the end of the ICC phase. The current in (a) has been wavelet-filtered to remove high-frequency electronic noise. The E-field waveforms follow the physics sign convention. Time is measured relative to the onset of the ICC.
• Figure 3: Characteristic measurements for the different pulses investigated in this study. Mean values have been selected for each characteristic.
• Figure 4: Plots depicting the relationships between the primary half-cycle characteristics of bipolar Category A pulses.
• Figure 5: A visual illustration comparing "Classical" and "Narrow" PBP with Category A and B pulse E-field waveforms, to emphasise their similarities. Plots (b) and (c) are adapted from azadifar_similarity_2018 (their Figures 2 & 3, respectively).