Elsevier

Atmospheric Research

Volume 276, 1 October 2022, 106260
Atmospheric Research

A study of cloud-to-ground lightning flashes initiated by fast positive breakdown

https://doi.org/10.1016/j.atmosres.2022.106260Get rights and content

Highlights

  • CG1 flash was initiated by a short VHF pulse followed by a VHF burst whileCG2 flash was initiated by a VHF burst alone.

  • The VHF burst was precursor for the initiation of NBE-like pulse and IECs.

  • Total lengths of VHF sources vertical propagation were 1.9–3.8 km for CG1 and 0.2–3.0 km for CG2.

  • Estimated velocities were 2.4–4.8 × 108 ms−1 for CG1 and 2.7 × 107 to 4.1 × 108 ms−1 for CG2.

  • The IECs process resulted from charge transfer during fast positive streamers breakdown.

Abstract

In this paper, we examine the initiation event of two tropical negative cloud to ground (CG) lightning flashes within reversal distance (<8 km) detected on 12 (CG1) and 24 (CG2) November 2019, respectively. The CG flashes were recorded using a fast antenna (FA), slow antenna (SA), magnetic field (B-field), and very high frequency (VHF) sensors located at two observation stations (ST1 and ST2) 13.26 km apart. The key finding is that CG1 flash was initiated by a short VHF pulse followed by a VHF burst (fast positive breakdown) while CG2 flash was initiated by a VHF burst (fast positive breakdown) not preceded by a short VHF pulse. The VHF burst was a precursor for initiation of a narrow bipolar event (NBE)-like pulse and initial electric field changes (IECs). Based on the VHF interferometer map, the VHF sources of NBE-like pulses have been propagated upward predominantly vertical. The initiation altitude, total length of the VHF sources propagation, estimated velocity of NBE-like pulse for CG1 flash were 4.5 ± 0.5 km, 1.9–3.8 km, and 2.4 × 108 to 4.8 × 108 ms−1, respectively, and for CG2 flash were 5.1 ± 0.8 km, 0.2–3.0 km, and 2.7 × 107 to 4.1 × 108 ms−1, respectively. It is evident that IECs process has been initiated during fast positive streamers propagation. It can be suggested that the IECs process resulted from charge transfer during a fast positive breakdown. The durations and magnitudes of IECs for CG1 flash were 248.5 μs and 27.97 V/m, respectively, and for CG2 flash were 265.8 μs and 2.31 V/m, respectively.

Introduction

Lightning flashes have been observed to be initiated by a series of bipolar events that has several names known as breakdown pulses, preliminary breakdown pulses (PBP), initial breakdown (IB) pulse and characteristic pulse (Appleton and Chapman, 1937; Schonland, 1938; Clarence and Malan, 1957; Kitagawa and Brook, 1960; Weidman and Krider, 1979; Beasley et al., 1982; Bils et al., 1988; Rakov et al., 1996; Gomes and Cooray, 2004; Sharma et al., 2008; Nag and Rakov, 2008; Nag and Rakov, 2009; Nag et al., 2009; Ahmad et al., 2010; Sharma et al., 2011; Stolzenburg et al., 2013; Esa et al., 2014a; Esa et al., 2014b). In this paper, we have chosen the term IB to refer to the bipolar pulses that initiate lightning flashes. There are two types of IB, namely narrow IB (NIB) and classical IB (CIB) pulses (Nag and Rakov, 2008; Nag et al., 2009). The classical IB pulse is larger than the narrow IB pulse with a typical pulse duration of around 10 μs and longer, and the main bipolar shape has two or three fast-rising pulses or superimposed on the main bipolar pulse. Narrow IB pulse was often detected during the initial stage of the IB process, and it was a precursor event to the classical IB pulses (Stolzenburg et al., 2014).

Recent studies regarding lightning initiation found that both cloud-to-ground (CG) and intra-cloud (IC) lightning flashes have been initiated by an electric field process known as initial electric field changes (IECs) (Marshall et al., 2014; Chapman et al., 2017; Marshall et al., 2019; Sabri et al., 2019). The studies suggested that IB was not the first process that initiated lightning. Instead, IEC was the first process that initiated lightning and then followed by the first IB pulse after sometimes (on the scale of milliseconds). The IEC is defined as a process of slow electric field changes that occur (moving upward for CGs and downward for ICs) and end just before the first classical IB pulse. This process is only possible to be observed when it happens very close to lightning sensors (within reversal distance, Marshall et al., 2014).

Based on Florida thunderstorms (Marshall et al., 2014), IECs were detected before the first classic IB pulse at the beginning of 36 flashes with a mean (range) value of 0.18 ms (0.08–0.33 ms) for 18 negative CG flashes and 1.53 ms (0.18–5.70 ms) for 18 IC flashes. Also, based on Florida thunderstorms, Chapman et al. (2017) found that IECs were detected before the first classic IB pulse at the beginning of 75 flashes with a mean (range) value of 0.23 ms (0.08–0.54 ms) for 17 negative CG flashes and 2.70 ms (0.04–9.80 ms) for 58 IC flashes. Furthermore, based on Mississippi thunderstorms, Marshall et al. (2019) studied the beginning of two IC and two CG lightning flashes. The IECs have been observed together with the detection of weak very high frequency (VHF) events. The flashes were initiated by VHF events with durations of 1 μs and 1–2 μs for IC and CG flashes, respectively. The peak VHF powers were less than one watt. Moreover, the duration of IECs for IC flashes was longer than CG flashes, between 0.62 and 1.790 ms compared to between 0.124 and 0.130 ms, respectively.

Marshall et al. (2014) proposed several hypotheses regarding IECs in lightning initiation events. It was suggested that the IEC process helped the cause of the first IB pulse. In other words, IEC was needed for an IB pulse to be started. This hypothesis was supported by independent studies in Chapman et al. (2017), Marshall et al. (2019) and Sabri et al. (2019). The second hypothesis relates VHF radiation with IEC process. The beginning of IECs was coincident with VHF radiation for most flashes where the VHF radiation was suggested to be emitted by an impulsive initiation event. Marshall et al. (2014) suggested these VHF radiations might be associated with a fast positive breakdown of narrow bipolar events (NBEs) discussed in Rison et al. (2016), Tilles et al. (2019) and Huang et al. (2021). The remaining hypotheses suggested these impulsive events produced ions, increasing electrical conductivity and the local electric field in the proximity of initiation location, thereby causing the first IB pulse to be emitted.

Sabri et al. (2019) have conducted similar experiment to measure IECs from tropical thunderstorms in Malacca, Malaysia. Based on the analysis of 80 flashes captured less than 10 km from observation station, it was found that most of CG and IC flashes have been initiated by IECs and the durations of IECs were longer compared to Florida storms. The mean (range) durations of IECs were 4.32 ms (0.22–13.47 ms) and 6.83 ms (0.08–64.71 ms) for CG and IC flashes, respectively. Moreover, the magnitudes of the IECs were smaller than the magnitudes of IECs in Florida thunderstorms. On the other hand, the authors reported that IECs of some flashes were preceded by several VHF radiation pulses. In other words, those VHF pulses were initiation events and not IECs. However, considering the measurement was conducted with a single observation station and the IECs might have been initiated at locations outside within reversal distance, it is possible the IECs were detected later than VHF pulses.

In this paper, we extend the study in Sabri et al. (2019) by analyzing lightning initiation events of two tropical negative CG flashes within reversal distance recorded on 12 (CG1) and 24 (CG2) November 2019 by using two observation stations placed 13.26 km apart. The locations of several IB pulses and first return strokes are obtained by using a magnetic direction finder (MDF) and VHF interferometer. The lightning waveforms were sampled with higher sampling rates of 125 MS/s and 250 MS/s than the sampling rate used in Sabri et al. (2019). The analysis focuses on the process before the onset of IECs and during the process of IECs. We found that CG1 flash was initiated by a short VHF pulse followed by a VHF burst (fast positive breakdown) while CG2 flash was initiated by a VHF burst (fast positive breakdown) not preceded by any VHF pulses.The VHF burst was a precursor for the initiation of NBE-like pulses and IECs. Interestingly, the characteristics of the first IB pulse for both CG flashes are very similar to NBEs except for the small electric field magnitude, and therefore we named it an NBE-like pulse.

Section snippets

Instrumentation and methods

The data presented here were recorded on 12 and 24 November 2019 at two observation stations located in Malacca, Malaysia, denoted as ST1 (UTeM, 2.314077°N, 102.318282°E) and ST2 (Paya Rumput, 2.3044°N, 102.1991°E). The observation stations were separated 13 km apart. The sensors that we deployed to study lightning initiation were wideband fast electric field change antenna (Fast Antenna or FA) and slow electric field change antenna (Slow Antenna or SA) systems (decay time constants of 13 ms

Results and analysis

The analysis presented here is based on two negative CG flashes (CG1 and CG2) detected within reversal distance from two tropical thunderstorms in Malacca, Malaysia. The flashes were detected simultaneously with slight time variation in microsecond scale at two stations (ST1 and ST2) placed 13.26 km apart, as shown in Fig. 1. The analysis focuses on the initiation event of the flashes with the detection of IECs and VHF pulses. Fig. 4 shows the records of FA, SA, B-fields, and VHF radiation

Discussions

We have analyzed lightning initiation event of two negative CG flashes (CG1 and CG2) detected at two observation stations located 13.26 km apart with higher sampling rate (125 MHz, 8 ns) and higher sensitivity of FA and SA antennas compared to previous work in Sabri et al. (2019). In both flashes, the initiation event started with the detection of a short VHF pulse/burst that initiated an NBE-like pulse and IECs. The CG1 flash was initiated by a short VHF pulse (0.31 μs) followed by a VHF burst

Conclusions

In this paper, we extended the studies in Sabri et al. (2019), Marshall et al. (2019), Chapman et al. (2017) and Marshall et al. (2014) by looking at the IECs in two tropical negative CG flashes detected within reversal distance at two observation stations. The first return stroke of both flashes was detected less than 5.1 km from the main station (ST1). The key finding is that CG1 flash was initiated by a short VHF pulse followed by a VHF burst (fast positive breakdown) while CG2 flash was

CRediT authorship contribution statement

Muhammad Haziq Mohammad Sabri: Investigation, Formal analysis, Writing – original draft, Writing – review & editing, Visualization. Mohd Riduan Ahmad: Conceptualization, Methodology, Formal analysis, Writing – original draft, Writing – review & editing, Supervision, Funding acquisition. Ammar Ahmed Nasser Al-Kahtani: Funding acquisition, Validation, Resources, Supervision. Mohd Zainal Abidin Ab Kadir: Validation, Resources, Supervision. Shamsul Ammar Shamsul Baharin: Investigation, Formal

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to acknowledge the support provided by Centre for Telecommunication Research and Innovation (CeTRI), Fakulti Kejuruteraan Elektronik dan Kejuruteraan Komputer, Universiti Teknikal Malaysia Melaka (UTeM), Institute of Power Engineering, College of Engineering, Universiti Tenaga Nasional (UNITEN), and UNITEN Bold Postgraduates Strategic Hires. This project is funded by CRG National Grant (S01707), research grant UNSRI International Grant (A00022), UNITEN Bold under project

References (39)

  • N.A. Ahmad et al.

    The first electric field pulse of cloud and cloud-to-ground lightning discharges

    J. Atmos. Sol. Terr. Phys.

    (2010)
  • A. Alammari et al.

    Kalman filter and wavelet cross-correlation for VHF broadband interferometer lightning mapping

    Appl. Sci.

    (2020)
  • A. Alammari et al.

    Lightning mapping: Techniques, challenges, and opportunities

    IEEE Access

    (2020)
  • E.V. Appleton et al.

    On the nature of atmospherics. IV

    Proc. Royal Soc. A: Math. Phys. Eng. Sci.

    (1937)
  • S.A.S. Baharin et al.

    Electric field waveforms of very close negative cloud to ground flashes

    J. Eng. Sci. Res.

    (2020)
  • W.H. Beasley et al.

    Electric fields preceding cloud-to-ground lightning flashes

    J. Geophys. Res.

    (1982)
  • J.R. Bils et al.

    Electric field pulses in close lightning cloud flashes

    J. Geophys. Res.-Atmos.

    (1988)
  • G.C. Carter

    Coherence and time delay estimation

    Proc. IEEE

    (1987)
  • R. Chapman et al.

    Initial electric field changes of lightning flashes in two thunderstorms

    J. Geophys. Res.

    (2017)
  • Cited by (4)

    • Microwave radiation associated with positive narrow bipolar events

      2023, Journal of Atmospheric and Solar-Terrestrial Physics
      Citation Excerpt :

      The lower and upper bounds of the propagation velocities are unreasonably high (faster than 1.0 × 108 ms−1) when compared to the velocities obtained in previous studies (Eack, 2004; Watson and Marshall, 2007; Rison et al., 2016; Tilles et al., 2019; Huang et al., 2021) where the acceptable upper bound should not exceed 1.0 × 108 ms−1. This is due to the high angular uncertainties because of the short baseline (1λ) used in the interferometry measurement (Sabri et al., 2022). The NBE3 that initiated an IC flash progressed to a single-stroke CG flash was located 13 km from the location of the first return stroke (RS1) as shown in Fig. 5A. Both NBE3 and RS1 were located within the convective core of the same thunderstorm on November 12, 2019.

    View full text