T. Ngin

Performance characteristics of the NLDN for return strokes and pulses superimposed on steady currents, based on rocket-triggered lightning data acquired in Florida in 2004–2012

We present a detailed evaluation of performance characteristics of the U.S. National Lightning Detection Network (NLDN) using, as ground truth, Florida rocket-triggered lightning data acquired in 2004–2012. The overall data set includes 78 flashes containing both the initial stage and leader/return-stroke sequences and 2 flashes composed of the initial stage only. In these 80 flashes, there are a total of 326 return strokes (directly measured channel-base currents are available for 290 of them) and 173 kiloampere-scale (≥1 kA) superimposed pulses, including 58 initial continuous current pulses and 115 M components. All these events transported negative charge to the ground. The NLDN detected 245 return strokes and 9 superimposed pulses. The resultant NLDN flash detection efficiency is 94%, return-stroke detection efficiency is 75%, and detection efficiency for superimposed pulses is 5% for peak currents ≥1 kA and 32% for peak currents ≥5 kA. For return strokes, the median location error is 334 m and the median value of absolute peak current estimation error is 14%. The percentage of misclassified events is 4%, all of them being return strokes. The median value of absolute event-time mismatch (the difference in times at which the event is reported to occur by the NLDN and recorded at the lightning triggering facility) for return strokes is 2.8 µs. For two out of the nine superimposed pulses detected by the NLDN, we found optical evidence of a reilluminated branch (recoil leader) coming in contact with the existing grounded channel at an altitude of a few hundred meters above ground.

Lightning attachment processes of an “anomalous” triggered lightning discharge

Using a high-speed optical imaging system specifically designed for observing the lightning attachment process, we have documented the process for stepped, dart, and dart-stepped leaders in an anomalous rocket-triggered lightning flash that terminated on a 10 m grounded utility pole. The initiation of the first return stroke was found to occur at a height of 23 ± 3 m above the top of the utility pole and was associated with three “slow front” dE/dt pulses. A time of 1.5 µs later, a fast rise in luminosity at 18 ± 2 m was associated with a “fast transition” dE/dt pulse. The first return stroke propagated bidirectionally from its initiation height, as did subsequent return strokes from their initiation heights of 8 ± 1 m to 16 ± 2 m above the top of the utility pole. The initial upward speed of the first return stroke was 1.4 × 108 m/s, while its initial downward speed was 2.2 × 107 m/s. The channel bottom luminosity of the first return stroke rose more slowly to a two or more times larger amplitude than that of the subsequent stroke luminosities. In contrast, the National Lightning Detection Network-derived first-return-stroke peak current is smaller than that of the second and the third strokes, and our electric field records at 45 km show similar behavior for the initial field peaks of the first and subsequent strokes.

Dart‐stepped‐leader step formation in triggered lightning

Dart-stepped-leader step formation in triggered lightning is documented with high-speed video recorded at 648 kiloframes per second (1.16 µs exposure time, 380 ns dead time) and linear streak film with a temporal resolution of about 1 µs. Locally luminous points and segments of channel both separate and below the main descending leader tip were recorded on the high-speed video. Bidirectional leaders were imaged initiating at the locally luminous points below the main channel tip, points that remain stationary during the interstep process. The average speed of five bidirectional leaders was 8.4 × 105 m/s upward and 4.8 × 105 m/s downward, assuming 1.5 µs between successive images. The main dart-stepped-leader channel tip moved downward between steps. Leader steps extended below the bottom of the previous bidirectional leader. Processes that can be seen between steps on high-speed video are generally below the noise threshold of the streak film, which shows primarily the newly formed steps.

Rocket‐triggered lightning propagation paths relative to preceding natural lightning activity and inferred cloud charge

Lightning Mapping Array (LMA) data are used to compare the propagation paths of seven rocket-triggered lightning flashes to the inferred charge structure of the thunderstorms in which they were triggered. This is the first LMA study of Florida thunderstorm charge structure. Three sequentially (within 16 min) triggered lightning flashes, whose initial stages were the subject of Hill et al. (2013), are reexamined by comparing the complete flashes to the preceding natural lightning to demonstrate that the three rocket-triggered flashes propagated through an inferred negative charge region that decreased from about 6.8 to about 4.4 km altitude as the thunderstorm dissipated. Two other flashes were also sequentially triggered (within 9 min) in a thunderstorm that contained a convectively intense region ahead of a stratiform region, with similar observed results. Finally, two unique cases of triggered lightning flashes are presented. In the first case, the in-cloud portion of the triggered lightning flash, after ascending to and turning horizontal at 5.3 km altitude, just above the 0°C level, was observed to very clearly resemble the geometry of the in-cloud portion of the preceding natural lightning discharges. In the second case, a flash was triggered relatively early in the storms lifecycle that did not turn horizontal near the 0°C level, as is usually the case for triggered lightning in dissipating storms, but ascended to nearly 7.5 km altitude before exhibiting extensive horizontal branching.

Evaluation of the GLD360 performance characteristics using rocket‐and‐wire triggered lightning data

We estimated the performance characteristics of the Global Lightning Dataset (GLD360) using rocket-and-wire triggered lightning data acquired at Camp Blanding, Florida, in 2011–2013. The data set consisted of 201 return strokes and 84 kiloampere-scale (≥1 kA) superimposed pulses (initial continuous current pulses and M components) in 43 flashes. All the events transported negative charge to ground. The GLD360 detected 75 strokes and 4 superimposed pulses in 29 flashes. The resultant detection efficiencies were 67% for flashes, 37% for strokes, and 4.8% for superimposed pulses. Out of 75 detected strokes, one (1.3%) was reported with incorrect polarity. The median location error was 2.0 km, and the median absolute current estimation error was 27%. This is the first comprehensive evaluation of GLD360 performance characteristics relative to absolute ground truth, with all previous evaluations being at least in part relative to other locating systems. The results presented in this work may be applicable to regions in and around Florida.

Correlation between the channel‐bottom light intensity and channel‐base current of a rocket‐triggered lightning flash

The correlations between channel-bottom light intensity and channel-base current of all discharge processes of a rocket-and-wire-triggered lightning flash, including initial continuous current (ICC) pulses, ICC pulse background continuing current (IBCC), return strokes, M components, and M component background continuing currents (MBCC), have been investigated. A rough linear correlation has been found between the current squared and the light intensity for ICC pulses (including peaks of different ICC pulses), IBCC, the initial rising stage (IRS) of return strokes (including current peaks of different strokes), and MBCC. The slopes of the correlation regression lines for the current squared versus light intensity of ICC pulses and IBCC are similar, but they are about 2–3 times smaller than the slopes of MBCC and 5–7 times smaller than the slopes of the IRS of return strokes. In contrast, a rough linear correlation has been found between the current and the light intensity for the later slow decay stage of return strokes and for the M components. The slopes of the correlation regression lines of the current versus the light intensity for these latter two processes are found to be similar. No simple correlation has been found between the current and the light intensity for the initial fast decay stage (IFDS) of return strokes. The duration of the IFDS of return strokes generally lasts from several microseconds to several tens of microseconds and is more or less directly proportional to the corresponding peak return stroke current squared. A time delay ranging from 12 µs to 300 µs has been found between the current and the light intensity of all ICC pulses and M components. The time delay decreases as the corresponding peak current increases.

Simultaneously measured lightning return stroke channel-base current and luminosity

The time delay between lightning return stroke current and the resultant luminosity was measured for 22 return strokes in eight lightning flashes triggered by the rocket-and-wire technique during the summer of 2014 in Florida. The current-to-luminosity delay measured at the channel base at the 20% amplitude level ranged from 30 to 200 ns with an average of 90 ns and at the 50% amplitude level ranged from 30 to 180 ns with an average of 94 ns. The delays are significantly shorter than that predicted by Liang et al. (2014) from theory. The current-to-luminosity delays increase with increasing current risetime, current risetime varying from 190 ns to 570 ns, but the delay appears not to depend on the peak current value.

Does the lightning current go to zero between ground strokes? Is there a current “cutoff”?

At the end of 120 prereturn stroke intervals in 27 lightning flashes triggered by rocket-and-wire in Florida, residual currents with an arithmetic mean of 5.3 mA (standard derivation 2.8 mA) were recorded. Average time constants of the current decay following return strokes were found to vary between 160 µs and 550 µs, increasing with decreasing current magnitude. These results represent the most sensitive measurements of interstroke lightning current to date, 2 to 3 orders of magnitude more sensitive than previously reported measurements, and contradict the common view found in the literature that there is a no current interval. Possible sources of the residual current are discussed.

Coordinated lightning, balloon-borne electric field, and radar observations of triggered lightning flashes in North Florida

This study examines coordinated storm and triggered lightning observations made in July–August 2013 at the International Center for Lightning Research and Testing to determine why triggered flashes in Florida typically transition from an upward vertical channel entering the cloud to horizontal structure near the storms melting level. Data from a balloon-borne electric field meter, a mobile 5 cm wavelength radar, and a small-baseline VHF Lightning Mapping Array acquired during a period in which three flashes were triggered on 1 August confirmed the hypothesis that the transition to horizontal lightning structure just above the melting level occurred in a layer of negative charge. This experiment was the first to provide vertical profiles of the electric field in Florida storms, from which their vertical charge distribution could be inferred. Three dissipating storms observed on different days all had negative charge near the melting level, but a growing mature storm had positive charge there.

Estimation of triggered‐lightning dart‐stepped‐leader currents from close multiple‐station dE/dt pulse measurements

The modified transmission line model is used to derive the vertically propagating leader-step currents necessary to radiate measured dart-stepped-leader dE/dt pulses from triggered lightning at close range (<400 m) and low altitude (<70 m). The model-predicted dE/dt pulses were compared with measured dE/dt pulses at nine locations ranging from 27 to 391 m from the channel base for four dE/dt pulses radiated from two triggered dart-stepped leaders. The dE/dt pulses at the closest station, 27 m, were unipolar, dominated by electrostatic and induction components of the radiated dE/dt, and of opposite polarity to the more distant initial dE/dt peaks. The other, more distant, eight stations exhibited bipolar dE/dt pulses, being more or less dominated by the dE/dt radiation component. The derived leader-step current has a slow front that precedes a fast transition to peak amplitude followed by a slow decay to zero after several microseconds. For the four modeled dE/dt pulses, the estimated causative leader-step current peak amplitudes varied from 0.9 to 1.8 kA, the half-peak widths ranged from 370 to 560 ns, the charge transfers were about 1 mC, and the peak current derivatives were about 10 kA/µs. The upward propagation speeds of the leader-step current were from 1.1 to 1.5 × 108 m/s with exponential spatial current decay constants from 13 to 27 m. One dE/dt pulse is analyzed in more detail by studying changes in model-predicted waveforms versus current initiation altitude and by examining the effect of varying model input parameters.