Fast Antenna Lightning Mapping Array (FALMA), a low‐frequency lightning mapping system comprising an array of fast antennas, was developed and established in Gifu, Japan, during the summer of 2017. Location results of two hybrid flashes and a cloud‐to‐ground flash comprising 11 return strokes (RSs) are described in detail in this paper. Results show that concurrent branches of stepped leaders can be readily resolved, and K changes and dart leaders with speeds up to 2.4 × 10⁷ m/s are also well imaged. These results demonstrate that FALMA can reconstruct three‐dimensional structures of lightning flashes with great details. Location accuracy of FALMA is estimated by comparing the located striking points of successive RSs in cloud‐to‐ground flashes. Results show that distances between successive RSs are mainly below 25 m, indicating exceptionally high location accuracy of FALMA.

We have been developing an LF sensor network called Broadband Observation network for Lightning and Thunderstorm (BOLT), to image the structure of lightning discharges in 3D using time of arrival (TOA) technique. This paper documents initial results and characteristics of BOLT source locations for further understanding of LF radiation associated with lightning. Theoretical reduced Chi‐square distribution fitted to BOLT observation data indicates a root mean square (rms) timing error of about 200 ns for each sensor. Monte Carlo simulations of BOLT indicate that at an altitude of 5000 m the standard deviations for horizontal differences between a known source and a location that the BOLT algorithm produces are less than 200 m, and vertical differences are less than 400 m in most of the network. Furthermore, comparison of BOLT and VHF source locations arriving at the same site within 5 µs indicates that the average difference for elevation direction is 0.73° with a standard deviation of 3.64°, and that for the azimuth direction is 0.58° with a standard deviation of 1.98°. Lightning flash processes of intracloud (IC) and cloud‐to‐ground (CG) flashes, including preliminary breakdown pulses, negative leaders, breakdown in the negative charge region (NCR), and an attempted leader, are well imaged by BOLT. Normalized amplitudes in E‐field change waveform of BOLT sources associated with negative leaders and breakdown occurred in the NCR do not have significant difference, implying that most BOLT sources in the NCR might be associated with negative recoil leaders.

Using a low‐frequency lightning location system comprising 11 sites, we located preliminary breakdown (PB) processes in 662 intracloud (IC) lightning flashes during the summer of 2013 in Osaka area of Japan. On the basis of three‐dimensional location results, we studied initiation altitude and upward propagation speed of PB processes. PB in most IC flashes has an initiation altitude that ranges from 5 to 10 km with an average of 7.8 km. Vertical speed ranges from 0.5 to 17.8 × 10⁵ m/s with an average of 4.0 × 10⁵ m/s. Vertical speed is closely related with initiation altitude, with IC flashes initiated at higher altitude having lower vertical speed during PB stage. Characteristics of PB pulse trains including pulse rate, pulse amplitude, and pulse width are also analyzed. The relationship between pulse rate and vertical speed has the strongest correlation, suggesting that each PB pulse corresponds to one step of the initial leader during the PB stage. Pulse rate, pulse amplitude, and pulse width all show decreasing trends with increasing initiation altitude and increasing trends with increasing vertical speed. Using a simple model, the step length of the initial leader during the PB stage is estimated. Most of initial leaders have step lengths that range from 40 to 140 m with an average of 113 m. Estimated step length has a strong correlation with initiation altitude, indicating that leaders initiated at higher altitude have longer steps. Based on the results of this study, we speculate that above certain altitude (~12 km), initial leaders in PB stages of IC flashes may only have horizontal propagations. PB processes at very high altitude may also have very weak radiation, so detecting and locating them would be relatively difficult.

The charge distribution of some cells in three winter thunderstorms in the Hokuriku region of Japan is investigated based on Lightning Mapping Array (LMA) flash data. The vertical arrangements of charge regions involved in lightning discharges suggest diverse charge patterns, including quad‐polar, tripole, positive dipole, inverted dipole, and inverted tripole. The riming electrification between graupel and ice crystals or their aggregations are thought to be responsible for the electrification of most cells. The charging process between snow/aggregates and ice crystals may be responsible for some inverted charge structure that occurred above 0 °C isotherm and accompanied with weak radar echoes. Convection indicated by the vertical development of radar reflectivity appears crucial to shaping the diverse charge distribution patterns by determining which charging mechanisms occur and where; it also influences changes in height or even the disappearance of the charge regions. The charged cores are distributed from 0.7‐ to 5.3‐km heights and 2‐ to –31‐ °C temperatures, while the distances between adjacent charged cores with opposite polarities change between 0.2 and 3.4 km, with a mean of 1.3 km. The mean flash duration and horizontal distance are 425.0 ms and 19.8 km, respectively. The average height, temperature, and power of flash initiations are 2.8 km, −11.9 °C, and 15.6 dBW, respectively.