- Email: [email protected]
Characteristics of leader pulses in positive ground flashes in Sweden
G Model
ARTICLE IN PRESS
EPSR-4971; No. of Pages 7
Electric Power Systems Research xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Electric Power Systems Research journal homepage: www.elsevier.com/locate/epsr
Characteristics of leader pulses in positive ground flashes in Sweden Dalina Johari a,b,∗ , Vernon Cooray a , Mahbubur Rahman a , Pasan Hettiarachchi a , Mohd Muzafar Ismail a,c a
Ångström Laboratory, Division of Electricity, Department of Engineering Sciences, Uppsala University, Box 534, 75121 Uppsala, Sweden Faculty of Electrical Engineering, Centre for Electrical Power Engineering Studies, Universiti Teknologi Mara, 40450 Shah Alam, Selangor, Malaysia Faculty of Electronics and Computer Engineering, Telecommunication Engineering Department, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, 76100 Durian Tunggal, Malacca, Malaysia b c
a r t i c l e
i n f o
Article history: Received 19 April 2016 Received in revised form 25 October 2016 Accepted 28 November 2016 Available online xxx Keywords: Positive ground flash Positive leader Negative leader Stepped-leader Bi-directional leader Negatively-charged leader branch
a b s t r a c t This paper presents the characteristics of the electric field pulses observed during leader propagation in positive ground flashes. We analysed in detail the electric field changes occurring just before the first return stroke in 51 positive ground flashes during 2014 summer thunderstorms in Uppsala, Sweden. Pronounced leader pulses (having the same polarity as the return stroke) were observed in 22% of the cases. They were observed to occur within 1.4 ms before the first return stroke. Interpulse duration ranged from 13.3 to 50.3 s with a mean value of 24.7 s. The peak amplitude of the leader pulses relative to the return stroke peak ranged from 2.7 to 17.8%. The presence of these pulses shows that the leaders propagate in a stepped manner. Based on the leader pulses’ time of initiation and average speed of the leader, the distance travelled by the leader was also estimated. One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke is also reported. To the best of our knowledge, this is the first time that such a case in positive ground flashes is reported. We suggest that these opposite polarity leader pulses are due to the negatively-charged leader branch of a bi-directional leader inside the cloud that propagates towards observation point. © 2016 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Positive ground flash Cloud-to-ground lightning can be defined as a transient, highcurrent discharge that transports charges from thunderclouds to the ground. The overall discharge, termed a flash, is composed of a number of processes such as preliminary breakdown, stepped leaders, connecting leaders, return strokes, dart leaders and subsequent return strokes [1]. Depending on the charges being transported to ground, a ground flash can be categorized into positive and negative. Positive ground flash is therefore a ground flash that brings positive charges down to earth. Of the two, positive ground flashes are less dominant and account for about 10% of global cloud-toground lightning [2].
∗ Corresponding author at: Ångström Laboratory, Lägerhyddsvägen 1, Box 534, Uppsala University, 75121 Uppsala, Sweden. Fax: +46 18 471 5810. E-mail addresses: [email protected], [email protected] (D. Johari), [email protected] (V. Cooray), [email protected] (M. Rahman), [email protected] (P. Hettiarachchi), [email protected] (M.M. Ismail).
According to [2], among the five observed properties that are thought to be associated with positive ground flashes is the propagation of the leaders that appears to move either continuously or in a stepped manner. In contrast, negative leaders always propagate in a stepped manner when they travel towards the ground. This means that during negative leader formation, the leader steps appear bright while the leader channel remains dark in between steps formation [1]. Positive leaders, on the other hand, exhibit a continuously luminous channel image either without such steps or with superimposed steps in the forms of luminosity enhancements [2].
1.2. Review on previous studies As part of the process in cloud-to-ground lightning flashes (CG), many studies have been conducted to characterize the leader process in both positive (e.g. [3–7]) and negative ground flashes (e.g. [8–11]). Plenty of information have been discovered for negative leaders [1] but the same cannot be said about positive leaders. Characteristics of leaders in positive ground flashes are still not well understood [7] and very little is known about the downward positive leaders in positive ground flashes compared to leaders in
http://dx.doi.org/10.1016/j.epsr.2016.11.026 0378-7796/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7 2
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
negative ground flashes [12]. Based on the information found in the literature, past observations made on leader process in positive ground flashes, particularly on the stepping behaviour of the pulses, can be briefly summarized as follows:
1.2.1. Electric field measurements Using electric field measurements, several researchers have observed the pulses preceding the first return stroke in their electric field waveforms (essentially radiation). Cooray and Lundquist [13], for example, observed that some positive waveforms were preceded by small amplitudes pulses which they attributed to the leader process. Hojo et al. [14] observed that 26–30% of the waveforms had discernable pulses preceding the positive return stroke. Schumann et al. [5] reported a rather large percentage where they observed leader pulses in 74% of the cases. Nag and Rakov [15] reported that 27% of the first return stroke waveforms were preceded by pronounced step pulses. According to Cooray and Lundquist [13], their electric field observations indicated that in some cases, positive leader behaves in a stepped manner during the last few hundred microseconds of its journey towards the earth. On the other hand, they said that the pulses they observed could also be due to the negative connecting leader that may develop upward from the ground under the influence of the downward positive leader. Based on their results and observations by Berger [16] and Les Renardières Group [17], Hojo et al. [14] inferred that most positive return strokes are preceded by leaders that propagate downwards either continuously or in a stepped manner. Berger [16] had previously obtained a streak-camera image of a positive leader propagating continuously downward. The observations showed that positively charged leaders display a very weak luminosity and less clear, or no stepping as compared to negative leaders that exhibit a very distinct and bright stepping. In contrast, Les Renardières Group [17] observed that a positive leader in a long laboratory spark exhibits stepping depending on the humidity, with higher humidity conducive for stepping. As for Schumann et al. [5], they said that it is difficult to say whether the pulses are due to downward leader or not since both downward positive leader and upward connecting negative leader would produce the same polarity pulses in the waveforms. Nag and Rakov [15] stated that the reason for the occurrence of the field pulses indicative of stepping prior to the return stroke pulse in some positive cloud-to-ground discharges is not known. According to them, the pulses could be associated with a descending positive leader, an upward connecting negative leader, which may be launched in response to the non-stepped positive downward leader, or both.
1.2.2. High-speed video recordings From high-speed video recordings, previous studies showed that downward-moving positive leaders could produce step-like pulses. Wang and Takagi [4] recorded a downward positive leader that radiated optical pulses like a negative stepped leader over the height from 299 m to 21 m above the ground but with a much larger rise time. Kong et al. [6] observed a downward-moving positive leader whose optical images showed a stepped-like development characteristics with a high intensity in the leader tip, and the pulses in the fast electric field 0.5 ms prior to the return stroke also suggested a stepped-like development. Saba et al. [7], however, did not observe any discrete steps that might indicate stepping. Instead, they observed that all downward positive leaders showed continuous progression towards the ground. An interesting finding from a recent study by Saba et al. [3] also showed that the pulses observed just before the return stroke were solely due to the upward connecting leader. Based on this, they suggested that if positive leaders step, their steps do not produce any significant electromagnetic
radiation and are much weaker than those produced by the stepping of the connecting negative leader. 1.3. Objectives This paper presents the characteristics of the leader pulses occurring just before the first return stroke in positive ground flashes during 2014 summer thunderstorms in Uppsala, Sweden (59.837◦ N, 17.646◦ E). The study is motivated by the fact that lightning processes in positive ground flashes are less well investigated than those in negative ground flashes [12]. Since there is still much discussion about the characteristics of positive leaders in positive ground flashes, the findings would add to existing knowledge. We present a comprehensive study of the leader pulses based on electric field measurements. We determined: (1) the percentage of detection of pronounced leader pulses, (2) the time of initiation of pronounced leader pulses before the first return stroke, (3) the peak amplitude of pronounced leader pulses relative to the return stroke peak and (4) the interpulse duration between successive pulses just before the first return stroke. One case of positive ground flashes preceded by opposite polarity leader pulses just before the return stroke is also reported. 1.4. Sign convention The atmospheric sign convention is used throughout the paper according to which a downward directed electric field is considered to be positive. According to this notation, a negative charge in a cloud produces a negative field at ground level and a negative return stroke will produce a positive field change [18]. A positive charge in a cloud will then produce a positive field at ground level and a positive return stroke will produce a negative field change. 2. Methodology 2.1. Measurement setup Electric field measurements were carried out in Uppsala, Sweden (59.837◦ N, 17.646◦ E) from June to August 2014 during summer season. The measurements were conducted using broadband antenna system (up to 100 MHz) consisting of a parallel plate antenna and a vertical whip antenna, a 12-bit Yokogawa transient recorder and a Meinberg M400 GPS system. The parallel plate antenna was used to detect the fast variation of the electric field (i.e. fast field) while the vertical whip antenna was used to detect the slow variation (i.e. slow field). Each antenna has a buffer amplifier circuit with decay time constants of 15 ms for fast field and 1 s for slow field. The sampling rate for the measurement system was 100 Msample/s with 10 ns interval and the measurements were conducted for a record length of 1 s with 200 ms pre-trigger time. The transient recorder was triggered automatically based on the voltage amplitude of the incoming signal from the parallel plate antenna and the trigger level was set above noise level at 50 mV. Detailed description on the antenna system and the buffer electronic circuits can be found in [19]. Further readings can also be found in [20–23] since the measurement setups used were identical. 2.2. Swedish Lightning Location System Information obtained from the Swedish Lightning Location System (LLS) were used to estimate the location and peak currents of the positive return strokes. The overall lightning flash detection efficiency of the Swedish LLS is about 85% and varies from point to point within Sweden [24]. Around the measuring site, the flash detection efficiency of the LLS is approximately 90% [25]. The
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
3
Fig. 1. Observed number of leader pulses (sample size, N = 51).
detection efficiency of the positive ground flashes, however, is not known. Further information about the Swedish LLS and its performance can be found in [24–26]. 2.3. Sample size The measurements were carried out continuously throughout the entire summer during which a total of 3025 lightning data were recorded. The final analysis was restricted to 51 positive ground flashes located 6–130 km from the measuring station. Since the signatures of the electromagnetic fields depend on the physical processes of the lightning flash, they can be used to identify the different processes associated with the ground flash. The return stroke is indicated by an abrupt and considerable electric field change pulse in the fast field record with coincident field change in the slow field record [27]. For leader process, broadband measurements of the fields indicate that small field pulses appear in the electric field preceding the return stroke [1]. We only considered pulses with peak amplitudes equal to or exceeding twice that of the average noise level since this condition would result in readable pulses for the analysis. Pronounced leader pulses have the same polarity as the positive return stroke. The leader pulses would be considered as having opposite polarity if they had polarity opposite to that of the positive return stroke. 3. Results In the study, electric field changes occurring just before the first return stroke in 51 positive ground flashes were analysed. The analysis was done based on the radiation field pulses associated with the leader process in positive ground flashes. 3.1. Percentage of detection Out of 51 waveforms, pronounced leader pulses were observed in 11 (22%) of the cases. In 25 cases, these pulses were not detected while in eight cases, they have very small amplitudes barely visible over the noise level. In six cases, only one or two pulses were visible just before the return stroke. One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke was also observed. Fig. 1 shows the overview of the observation while Fig. 2 shows typical pronounced leader pulses observed just before a positive return stroke. For the 11 positive ground flashes with pronounced leader pulses, the ratio of the positive ground flashes with pronounced leader pulses to the number of positive ground flashes detected is about 8% when the return stroke distance from the measuring station is close (<20 km), 29% for distance in the 20–50 km range,
36% for the 50–100 km range and 11% for distance greater than 100 km. The ratio increases as the return stroke distance increases but then it decreases when the distance is greater than 100 km. In principle, the radiation field is inversely proportional to the distance [28]. Therefore, more leader pulses occurrences should be observed when the return stroke distance is close. However, when distance is close, static field is dominant and the fine features of the radiation field might be hidden inside the large static field (overwhelmed by the static contribution). On the other hand, when the distance is very large (say, greater than 100 km), the noise level might be comparable to the small radiation field pulses (since radiation field becomes smaller when the distance is larger). These could be the reasons why we only observed a small number of leader pulses occurrence when the distance was close and when the distance was greater than 100 km. The sample size is rather small though (N = 11) and the effect of propagation to the electric fields is not known. Therefore, we could not confirm any clear dependence of the leader pulses occurrences with respect to the return stroke distance at this moment.
3.2. Characteristics of pronounced leader pulses In order to determine the characteristics of the pronounced leader pulses, we obtained the time of initiation of the leader pulses before the first return stroke, the peak amplitude of the leader pulses relative to the return stroke peak and the interpulse duration between successive pulses just before the first return stroke. The statistics and comparison with previous studies are given in Table 1. From Table 1, we observed the leader pulses to start 0.26–1.40 ms before the following first return stroke. This is comparable to 0.14–1.32 ms found by Schumann et al. [5]. On the other hand, Nag and Rakov [15] observed the pulses to appear 0.07–0.63 ms before the first return stroke. We also found that the peak amplitude of the leader pulses relative to the return stroke peak ranged between 2.7 and 17.8% of the return stroke peak with an average of 6.4%. In comparison, Nag and Rakov [15] found the largest amplitude of the step pulses ranged from 8 to 19% of the following return stroke pulse with a mean value of 14%. Schumann et al. [5] found the peak amplitude of the leader pulses ranged between 2.1 and 3.6% of the return stroke peak. We also observed that the interpulse duration between successive pulses just before the first return stroke ranged between 13.3 and 50.3 s with a mean value of 24.7 s. Our mean interpulse duration was slightly higher than those found by Hojo et al. [14] (17.4 s) and Kong et al. [6] (17 s, ranging from 3–31 s) but comparable to that found by Nag and Rakov [15] (20 s, ranging
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model
ARTICLE IN PRESS
EPSR-4971; No. of Pages 7
D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
4
Fig. 2. Typical pronounced leader pulses observed just before a positive return stroke (top) and a larger scale (zoom in) of the waveform (bottom) (from fast field records).
Table 1 Time of initiation, relative peak amplitude and interpulse duration between successive pulses of pronounced leader pulses. Researcher
Present study Schumann et al. [5] Nag and Rakov [15] Cooray and Lundquist [13] Hojo et al. [14] Kong et al. [6] a b
Location
Uppsala, Sweden Brazil, US, Austria Florida, US Uppsala, Sweden Niigata, Japan Shandong, China
Sample size, N
11 14 15 96 7 –
Time of initiation (ms)
Ratio of leader peak to first RS peak (%)
Interpulse duration (s)
Max
Min
AMa
Max
Min
AMa
Max
Min
AMa
1.4 1.32 0.63 – – –
0.26 0.14 0.07 – – –
0.95 0.45 – – – –
17.8 – 19 – – –
2.7 – 8 – – –
6.4 2.1–3.6b 14 – – –
50.3 – 37 48 – 31
13.3 – 5.8 12 – 3
24.7 20.2–22.5 20 26 17.4 17
AM = arithmetic mean. Sample size = 7.
Fig. 3. Time of initiation of the pronounced leader pulses versus the LLS-reported peak current of the positive return strokes (sample size, N = 11).
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
from 5.8–37 s), Schumann et al. [5] (20.2–22.5 s) and Cooray and Lundquist [13] (26 s). Next, we plotted the time of initiation of the pronounced leader pulses against the LLS-reported peak current of the positive return strokes, given in Fig. 3. As seen from Fig. 3, there is a tendency for the time of initiation of the leader pulses to increase with respect to the peak current of the return stroke. It should be noted that the LLSestimated peak current values for the positive return strokes may not be accurate since the peak currents estimated by the lightning detection networks is calibrated against direct current measurements for negative triggered-lightning strokes [29].
3.3. Opposite polarity leader pulses One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke was also observed. The waveform is shown in Fig. 4 while the data for these pulses are given in Table 2. The distance of this positive ground flash, however, could not be estimated since the Swedish LLS did not record the occurrence. From Table 2, the pulses were observed to appear 0.02 ms before the following first return stroke. Compared to that of the pronounced leader pulses (0.26–1.40 ms), the time of initiation of the negative pulses was much closer to the return stroke. Interpulse duration was found to be 1.78 s (also smaller than the minimum of the pronounced leader pulses of 13.3 s) while the peak amplitude was 3.77% of the return stroke peak.
5
Table 2 Data for opposite polarity leader pulses. Sample size, N
Time of initiation (ms)
Interpulse duration (s)
Ratio of leader peak to first RS peak (%)
1
0.02
1.78
3.77
4. Discussion 4.1. Pronounced leader pulses The presence of these pronounced leader pulses (having the same polarity as the return stroke) suggests that the leaders can propagate in a stepped manner. In a recent study, Saba et al. [3] found that the pulses observed just before the return stroke were solely due to the upward connecting negative leader. They observed that these pulses in the electric field waveforms occur simultaneously with the initiation of the upward connecting negative leader in their video recordings. According to the authors, the initiation of the upward connecting negative leader and the pulses occur at approximately 1 ms before the return stroke, and the time interval between the pulses was approximately 23 s. In the literature concerning negative ground flashes, Krider et al. [30] found that the average time interval between the pulses of the radiation field produced by the negative stepped leaders is about 16 s [30]. Similarly, Cooray and Lundquist [13] found that the time interval between successive leader pulses immediately preceding the radiation fields
Fig. 4. A positive ground flash preceded by opposite polarity leader pulses (i.e. opposite to that of the positive return stroke) (top) and a larger scale (zoom in) of the waveform (bottom) (from fast field records).
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7 6
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
from negative first strokes is about 14 s (ranging between 6 and 30 s) while Ismail et al. [23] found the time interval to be larger than 10 s. Considering that the pulses we observed appear within 1.4 ms before the return stroke (mean 0.95 ms), the mean time interval between the pulses was 24.7 s, the time interval ranged between 13.3 and 50.3 s (within the negative stepped leader values), and the fact that negative leaders always appear to advance in steps [1,2], it is possible that the pulses we observed were due to the upward connecting negative leader. However, since we only have observations from the electric field measurements, we cannot rule out the possibility that it might also come from the downward positive leaders (as well as other possibilities).
4.2. Estimation of the height of the leader from the ground Previous researchers have reported the 2-D (two-dimensional) average speed of the downward-moving positive leaders [4,6,7,31–33]. Saba et al. [7] found that the 2-D average speeds for nine positive leaders ranged from 0.33 to 6.0 × 105 ms−1 (mean of 2.7 × 105 ms−1 ), and the speeds have the tendency to increase by a factor of 1.1–6.5 as the leaders approach the ground. They also made four other measurements and obtained average speeds that ranged between 1.6 and 7.8 × 105 ms−1 [31]. Campos et al. [34] determined the 2-D average speeds for 29 positive leaders and found that the values ranged between 0.24 and 11.80 × 105 ms−1 with an average of 2.76 × 105 ms−1 (also reported in [32]). Wang & Takagi [4] observed that one downward positive leader propagated at a speed of about 1.0 × 106 ms−1 at a height of 272–93 m above the ground before accelerating to a speed of 2.5 × 106 ms−1 at a height of about 45 m. Kong et al. [6] found that the 2-D speed of one stepped-like leader increased from 0.1 × 105 ms−1 to 3.8 × 105 ms−1 as it propagated towards the ground. In a recent study, Kong et al. [33] found that the average 2-D speeds for six positive leaders ranged from 0.3 × 105 ms−1 to 2.0 × 105 ms−1 and they increased as the leaders approached the ground. Based on their findings and our results on the pulses’ time of initiation, we can estimate the height above the ground when the leader pulses begin to appear. Assuming that the leader pulses were due to the downward positive leader, and using an average 2-D speed of 2.7 × 105 ms−1 for the downward leader [7] and the average time of initiation of 0.95 ms, we calculated that the distance travelled by the leader was approximately 256.5 m. Now, as the downward positive leaders approach the ground, they will induce upward-moving connecting leaders that will travel towards the positive leaders. The length of the connecting leader can be several tens of metres [35]. Electrical breakdown will occur when the separation between the two leader tips reaches a certain distance, in which case the two will meet resulting in a return stroke. Depending on the length of the connecting leader, several scenarios can be considered. If the length of the connecting leader is very small compared to the downward leader (i.e. negligible), then the distance that we calculated (i.e. the distance travelled by the downward positive leader) is the height above the ground that the leader pulses begin to appear. The height would increase if the connecting leader has a certain length and originated from a structure. In a recent study, Saba et al. [3] captured images of a positive downward leader and a negative upward connecting leader from a 163 m tower preceding a positive ground flash. The estimated peak current of the positive ground flash was 124 kA. The connecting leader travelled a distance of 270 m and the calculated average 2-D speed of the connecting leader was 3.0 × 105 ms−1 . If we consider these (assuming that the downward leader is attached to a connecting leader initiated from a tall structure), then the leader is estimated
to begin to appear at approximately 689.5 m above the ground (i.e. 256.5 + 270 + 163 = 689.5 m). The connecting leader would be longer if the charge on the leader channel and the height of the structure increase as these conditions promote long connecting leaders [35]. So, if the connecting leader were initiated from a shorter structure, its length might be less than 270 m. It then follows that if the connecting leader were initiated from the ground level, the height above the ground that the leader begin to appear should be less than 526.5 m. If however, we considered that the pulses were due to the upward connecting leader itself, then using the average 2-D speed of 3.0 × 105 ms−1 calculated by [3], we estimated that the connecting leader travelled a distance of approximately 285 m from the ground before attachment with the downward positive leader. If the downward positive leader has a higher current, the connecting leader may be initiated when the downward positive leader is still considerably far from the ground. This will result in longer connecting leader length and greater height of the leader pulses when they begin to appear above the ground. If, however, the downward positive leader has a smaller current, the length of the connecting leader may also be smaller because it will be initiated when the downward leader was closer to ground level. This distance that the connecting leader travelled would then represent an estimate of the striking distance i.e. the distance between the tip of the downward moving leader and the point of origin of the connecting leader when they meet [35]. 4.3. Opposite polarity leader pulses For the opposite polarity leader pulses, they were observed to start 0.02 ms before the following positive return stroke. The interpulse duration was 1.78 s and the peak amplitude was 3.77% of the return stroke peak. Both downward positive leader and upward connecting leader in positive ground flashes produce pulses with the same polarity as the positive return stroke in the electric field waveforms. However, the pulses we observed in this case were of the opposite polarity to that of the positive return stroke. Considering that they occurred during the leader process in positive ground flashes, these pulses may be due to the negatively-charged leader branch of a bi-directional leader propagating inside the cloud. If we consider the negatively-charged leader branch moving towards an observer, this would give rise to pulses with opposite polarity (to that of the positive return stroke) in the electric field waveforms. However, if the negative leader branch is moving away from the observer, this would give rise to same polarity pulses (to that of the positive return stroke) in the waveforms. Pulse bursts like these have been observed previously by Krider et al. [36] who recorded sequences or bursts of uniform pulses during intracloud lightning discharges, with time intervals between the pulses of typically 5 s. They suggested that based on the pulse shape and time interval, the source of these pulses could be an intracloud dart-stepped leader process. The pulses that we recorded would then be the first observation that such pulses occur in positive ground flash. 5. Conclusion The presence of the same polarity pulses suggests that the leaders can propagate in a stepped manner. It is possible that the pulses were due to the upward connecting negative leader developed under the influence of the downward positive leader. However, since we only have electric field measurements, we cannot rule out the possibility that it might also come from the downward positive leaders since both would produce same polarity pulses (to that of the positive return stroke) in the electric field waveforms.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
If these pulses were attributed to the downward positive leader, the time of initiation obtained from the electric field records can be used to estimate the height the leader begin to appear from the ground. If however, the pulses were attributed to the upward connecting negative leader, the distance the connecting leader travelled would be an estimate of the striking distance of the ground flash. If this is the case, the striking distance would depend on the current on the leader channel and the height of the structure that it originated from. One case of positive ground flash preceded by opposite polarity pulses just before the return stroke was also observed. These opposite polarity pulses could be due to the negatively-charged leader branch of a bi-directional leader inside the cloud that propagates towards an observation point. Acknowledgements The authors would like to express their gratitude to all the people who have directly or indirectly contribute towards the successful completion of this paper. Participation of Dalina Johari is funded by the Ministry of Education of Malaysia and Universiti Teknologi MARA Malaysia. Participation of Prof. Vernon Cooray and Dr. Mahbubur Rahman are funded by the fund from B. John F. and Svea Andersson donation at Uppsala University. Participation of Mohd Muzafar Ismail is funded by the Ministry of Education of Malaysia and Universiti Teknikal Malaysia Melaka. We would also like to thank Thomas Götschl for the assistance in the measurement setup and acquisition of lightning data from the Swedish LLS database. Finally, the authors would like to acknowledge the Division of Electricity, Ångström Laboratory, Uppsala University, for the excellent facility provided to carry out this research. References [1] V. Cooray, Mechanism of the lightning flash, in: V. Cooray (Ed.), The Lightning Flash, 2nd ed., The Institution of Engineering and Technology, London, United Kingdom, 2014, pp. 119–229. [2] V.A. Rakov, M.A. Uman, Positive and bipolar lightning discharges to ground, in: Lightning: Physics and Effects, Cambridge University Press, 2003, pp. 214–240. [3] M.M.F. Saba, C. Schumann, T. a. Warner, J.H. Helsdon, R.E. Orville, High-speed video and electric field observation of a negative upward leader connecting a downward positive leader in a positive cloud-to-ground flash, Electr. Power Syst. Res. 118 (2015) 89–92. [4] D. Wang, N. Takagi, A downward positive leader that radiated optical pulses like a negative stepped leader, J. Geophys. Res. 116 (May (D10)) (2011) D10205. [5] C. Schumann, M.M.F. Saba, L.Z.S. Campos, R.B.G. Silva, W. Schulz, Leader characteristics in positive cloud-to-ground lightning flashes, International Symposium on Winter Lightning (ISWL) (2011) 59–62. [6] X. Kong, X. Qie, Y. Zhao, Characteristics of downward leader in a positive cloud-to-ground lightning flash observed by high-speed video camera and electric field changes, Geophys. Res. Lett. 35 (5) (2008). [7] M.M.F. Saba, K.L. Cummins, T.A. Warner, E.P. Krider, L.Z.S. Campos, M.G. Ballarotti, O. Pinto, S.A. Fleenor, Positive leader characteristics from high-speed video observations, Geophys. Res. Lett. 35 (7) (2008) L07802. [8] J.D. Hill, M.A. Uman, D.M. Jordan, High-speed video observations of a lightning stepped leader, J. Geophys. Res. Atmos. 116 (16) (2011). [9] M. Stolzenburg, T.C. Marshall, S. Karunarathne, N. Karunarathna, R.E. Orville, Transient luminosity along negative stepped leaders in lightning, J. Geophys. Res. D Atmos. 120 (8) (2015) 3408–3435. [10] D.A. Petersen, W.H. Beasley, High-speed video observations of a natural negative stepped leader and subsequent dart-stepped leader, J. Geophys. Res. Atmos. 118 (21) (2013) 12110–12119.
7
[11] W. Lu, L. Chen, Y. Ma, V.A. Rakov, Y. Gao, Y. Zhang, Q. Yin, Y. Zhang, Lightning attachment process involving connection of the downward negative leader to the lateral surface of the upward connecting leader, Geophys. Res. Lett. 40 (20) (2013) 5531–5535. [12] V. Mazur, L.H. Ruhnke, The physics of lightning flash development, in: V. Cooray (Ed.), Lightning Electromagnetics, 1st ed., Institution of Engineering and Technology, 2012, pp. 193–229. [13] V. Cooray, S. Lundquist, On the characteristics of some radiation fields from lightning and their possible origin in positive ground flashes, J. Geophys. Res. 87 (1982) 11203. [14] J. Hojo, M. Ishii, T. Kawamura, F. Suzuki, R. Funayama, The fine structure in the field change produced by positive ground strokes, J. Geophys. Res. 90 (1985) 6139–6143. [15] A. Nag, V.A. Rakov, Positive lightning: An overview, new observations, and inferences, J. Geophys. Res. 117 (2012) D08109. [16] K. Berger, Novel observations on lightning discharges: Results of research on Mount San Salvatore, J. Franklin Inst. 283 (6) (1967) 478–525. [17] Les Renardieres Group, Positive discharges in long air gaps at Les Renardieres —1975. Results and conclusions, Electra 53 (1975) 31–153. [18] V. Cooray, Electromagnetic fields of lightning flashes, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 135–165. [19] A. Galvan, M. Fernando, Operative Characteristics of a Parallel-Plate Antenna to Measure Vertical Electric Fields from Lightning Fields from Lightning Flashes, Report UURIE 285-00, Uppsala University, 2000. [20] Z.A. Baharudin, N.A. Ahmad, M. Fernando, V. Cooray, J.S. Mäkelä, Comparative study on preliminary breakdown pulse trains observed in Johor, Malaysia and Florida, USA, Atmos. Res. 117 (2012) 111–121. [21] Z.A. Baharudin, M. Fernando, N.A. Ahmad, J.S. Mäkelä, M. Rahman, V. Cooray, Electric field changes generated by the preliminary breakdown for the negative cloud-to-ground lightning flashes in Malaysia and Sweden, J. Atmos. Sol. Terr. Phys. 84–85 (2012) 15–24. [22] D. Johari, V. Cooray, M. Rahman, P. Hettiarachchi, M.M. Ismail, Characteristics of preliminary breakdown pulses in positive ground flashes during summer thunderstorms in Sweden, Atmosphere (Basel) 7 (3) (2016) 39. [23] M.M. Ismail, M. Rahman, V. Cooray, S. Sharma, P. Hettiarachchi, D. Johari, Electric field signatures in wideband, 3 MHz and 30 MHz of negative ground flashes pertinent to Swedish thunderstorms, Atmosphere (Basel) 6 (12) (2015) 1904–1925. [24] U. Sonnadara, V. Cooray, T. Götschl, Characteristics of cloud-to-ground lightning flashes over Sweden, Phys. Scr. 74 (5) (2006) 541–548. [25] M. Fernando, A. Galván, T. Götschl, V. Cooray, V. Scuka, Analysis of Swedish lightning using ‘LLP’ data, Light. Prot. (ICLP), 1998 Int. Conf. (1998) 150–155. [26] U. Sonnadara, V. Kathriarachchi, V. Cooray, R. Montano, T. Götschl, Performance of lightning locating systems in extracting lightning flash characteristics, J. Atmos. Sol. Terr. Phys. 112 (2014) 31–37. [27] D.M. Fuquay, Positive cloud-to-ground lightning in summer thunderstorms, J. Geophys. Res. Ocean 87 (August (C9)) (1982) 7131–7140. [28] V. Cooray, Basic electromagnetic theory with special attention to lightning electromagentics, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 29–58. [29] A. Nag, V.A. Rakov, K.L. Cummins, Positive lightning peak currents reported by the U.S. national lightning detection network, IEEE Trans. Electromagn. Compat. 56 (2014) 404–412. [30] E.P. Krider, C.D. Weidman, C.R. Noggle, The electric fields produced by lightning stepped leaders, J. Geophys. Res. 82 (6) (1977) 951–960. [31] M.M.F. Saba, M.G. Ballarotti, L.Z.S. Campos, O.P. Jr, High-speed video observations of positive lightning, International Symposium on Lightning Protection (IX SIPDA) (2007), pp. 1–5. [32] M.M.F. Saba, W. Schulz, T.A. Warner, L.Z.S. Campos, C. Schumann, E.P. Krider, K.L. Cummins, R.E. Orville, High-speed video observations of positive lightning flashes to ground, J. Geophys. Res. 115 (D24) (2010). [33] X. Kong, Y. Zhao, T. Zhang, H. Wang, Optical and electrical characteristics of in-cloud discharge activity and downward leaders in positive cloud-to-ground lightning flashes, Atmos. Res. 160 (2015) 28–38. [34] L.Z.S. Campos, M.M.F. Saba, T.A. Warner, O. Pinto, E.P. Krider, R.E. Orville, High-speed video observations of natural cloud-to-ground lightning leaders—a statistical analysis, Atmos. Res. 135–136 (2014) 285–305. [35] V. Cooray, Mechanism of lightning flash, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 91–116. [36] E.P. Krider, G.J. Radda, R.C. Noggle, Regular radiation field pulses produced by intracloud lightning discharges, J. Geophys. Res. 80 (27) (1975) 3801–3804.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
ARTICLE IN PRESS
EPSR-4971; No. of Pages 7
Electric Power Systems Research xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Electric Power Systems Research journal homepage: www.elsevier.com/locate/epsr
Characteristics of leader pulses in positive ground flashes in Sweden Dalina Johari a,b,∗ , Vernon Cooray a , Mahbubur Rahman a , Pasan Hettiarachchi a , Mohd Muzafar Ismail a,c a
Ångström Laboratory, Division of Electricity, Department of Engineering Sciences, Uppsala University, Box 534, 75121 Uppsala, Sweden Faculty of Electrical Engineering, Centre for Electrical Power Engineering Studies, Universiti Teknologi Mara, 40450 Shah Alam, Selangor, Malaysia Faculty of Electronics and Computer Engineering, Telecommunication Engineering Department, Universiti Teknikal Malaysia Melaka, Hang Tuah Jaya, 76100 Durian Tunggal, Malacca, Malaysia b c
a r t i c l e
i n f o
Article history: Received 19 April 2016 Received in revised form 25 October 2016 Accepted 28 November 2016 Available online xxx Keywords: Positive ground flash Positive leader Negative leader Stepped-leader Bi-directional leader Negatively-charged leader branch
a b s t r a c t This paper presents the characteristics of the electric field pulses observed during leader propagation in positive ground flashes. We analysed in detail the electric field changes occurring just before the first return stroke in 51 positive ground flashes during 2014 summer thunderstorms in Uppsala, Sweden. Pronounced leader pulses (having the same polarity as the return stroke) were observed in 22% of the cases. They were observed to occur within 1.4 ms before the first return stroke. Interpulse duration ranged from 13.3 to 50.3 s with a mean value of 24.7 s. The peak amplitude of the leader pulses relative to the return stroke peak ranged from 2.7 to 17.8%. The presence of these pulses shows that the leaders propagate in a stepped manner. Based on the leader pulses’ time of initiation and average speed of the leader, the distance travelled by the leader was also estimated. One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke is also reported. To the best of our knowledge, this is the first time that such a case in positive ground flashes is reported. We suggest that these opposite polarity leader pulses are due to the negatively-charged leader branch of a bi-directional leader inside the cloud that propagates towards observation point. © 2016 Elsevier B.V. All rights reserved.
1. Introduction 1.1. Positive ground flash Cloud-to-ground lightning can be defined as a transient, highcurrent discharge that transports charges from thunderclouds to the ground. The overall discharge, termed a flash, is composed of a number of processes such as preliminary breakdown, stepped leaders, connecting leaders, return strokes, dart leaders and subsequent return strokes [1]. Depending on the charges being transported to ground, a ground flash can be categorized into positive and negative. Positive ground flash is therefore a ground flash that brings positive charges down to earth. Of the two, positive ground flashes are less dominant and account for about 10% of global cloud-toground lightning [2].
∗ Corresponding author at: Ångström Laboratory, Lägerhyddsvägen 1, Box 534, Uppsala University, 75121 Uppsala, Sweden. Fax: +46 18 471 5810. E-mail addresses: [email protected], [email protected] (D. Johari), [email protected] (V. Cooray), [email protected] (M. Rahman), [email protected] (P. Hettiarachchi), [email protected] (M.M. Ismail).
According to [2], among the five observed properties that are thought to be associated with positive ground flashes is the propagation of the leaders that appears to move either continuously or in a stepped manner. In contrast, negative leaders always propagate in a stepped manner when they travel towards the ground. This means that during negative leader formation, the leader steps appear bright while the leader channel remains dark in between steps formation [1]. Positive leaders, on the other hand, exhibit a continuously luminous channel image either without such steps or with superimposed steps in the forms of luminosity enhancements [2].
1.2. Review on previous studies As part of the process in cloud-to-ground lightning flashes (CG), many studies have been conducted to characterize the leader process in both positive (e.g. [3–7]) and negative ground flashes (e.g. [8–11]). Plenty of information have been discovered for negative leaders [1] but the same cannot be said about positive leaders. Characteristics of leaders in positive ground flashes are still not well understood [7] and very little is known about the downward positive leaders in positive ground flashes compared to leaders in
http://dx.doi.org/10.1016/j.epsr.2016.11.026 0378-7796/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7 2
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
negative ground flashes [12]. Based on the information found in the literature, past observations made on leader process in positive ground flashes, particularly on the stepping behaviour of the pulses, can be briefly summarized as follows:
1.2.1. Electric field measurements Using electric field measurements, several researchers have observed the pulses preceding the first return stroke in their electric field waveforms (essentially radiation). Cooray and Lundquist [13], for example, observed that some positive waveforms were preceded by small amplitudes pulses which they attributed to the leader process. Hojo et al. [14] observed that 26–30% of the waveforms had discernable pulses preceding the positive return stroke. Schumann et al. [5] reported a rather large percentage where they observed leader pulses in 74% of the cases. Nag and Rakov [15] reported that 27% of the first return stroke waveforms were preceded by pronounced step pulses. According to Cooray and Lundquist [13], their electric field observations indicated that in some cases, positive leader behaves in a stepped manner during the last few hundred microseconds of its journey towards the earth. On the other hand, they said that the pulses they observed could also be due to the negative connecting leader that may develop upward from the ground under the influence of the downward positive leader. Based on their results and observations by Berger [16] and Les Renardières Group [17], Hojo et al. [14] inferred that most positive return strokes are preceded by leaders that propagate downwards either continuously or in a stepped manner. Berger [16] had previously obtained a streak-camera image of a positive leader propagating continuously downward. The observations showed that positively charged leaders display a very weak luminosity and less clear, or no stepping as compared to negative leaders that exhibit a very distinct and bright stepping. In contrast, Les Renardières Group [17] observed that a positive leader in a long laboratory spark exhibits stepping depending on the humidity, with higher humidity conducive for stepping. As for Schumann et al. [5], they said that it is difficult to say whether the pulses are due to downward leader or not since both downward positive leader and upward connecting negative leader would produce the same polarity pulses in the waveforms. Nag and Rakov [15] stated that the reason for the occurrence of the field pulses indicative of stepping prior to the return stroke pulse in some positive cloud-to-ground discharges is not known. According to them, the pulses could be associated with a descending positive leader, an upward connecting negative leader, which may be launched in response to the non-stepped positive downward leader, or both.
1.2.2. High-speed video recordings From high-speed video recordings, previous studies showed that downward-moving positive leaders could produce step-like pulses. Wang and Takagi [4] recorded a downward positive leader that radiated optical pulses like a negative stepped leader over the height from 299 m to 21 m above the ground but with a much larger rise time. Kong et al. [6] observed a downward-moving positive leader whose optical images showed a stepped-like development characteristics with a high intensity in the leader tip, and the pulses in the fast electric field 0.5 ms prior to the return stroke also suggested a stepped-like development. Saba et al. [7], however, did not observe any discrete steps that might indicate stepping. Instead, they observed that all downward positive leaders showed continuous progression towards the ground. An interesting finding from a recent study by Saba et al. [3] also showed that the pulses observed just before the return stroke were solely due to the upward connecting leader. Based on this, they suggested that if positive leaders step, their steps do not produce any significant electromagnetic
radiation and are much weaker than those produced by the stepping of the connecting negative leader. 1.3. Objectives This paper presents the characteristics of the leader pulses occurring just before the first return stroke in positive ground flashes during 2014 summer thunderstorms in Uppsala, Sweden (59.837◦ N, 17.646◦ E). The study is motivated by the fact that lightning processes in positive ground flashes are less well investigated than those in negative ground flashes [12]. Since there is still much discussion about the characteristics of positive leaders in positive ground flashes, the findings would add to existing knowledge. We present a comprehensive study of the leader pulses based on electric field measurements. We determined: (1) the percentage of detection of pronounced leader pulses, (2) the time of initiation of pronounced leader pulses before the first return stroke, (3) the peak amplitude of pronounced leader pulses relative to the return stroke peak and (4) the interpulse duration between successive pulses just before the first return stroke. One case of positive ground flashes preceded by opposite polarity leader pulses just before the return stroke is also reported. 1.4. Sign convention The atmospheric sign convention is used throughout the paper according to which a downward directed electric field is considered to be positive. According to this notation, a negative charge in a cloud produces a negative field at ground level and a negative return stroke will produce a positive field change [18]. A positive charge in a cloud will then produce a positive field at ground level and a positive return stroke will produce a negative field change. 2. Methodology 2.1. Measurement setup Electric field measurements were carried out in Uppsala, Sweden (59.837◦ N, 17.646◦ E) from June to August 2014 during summer season. The measurements were conducted using broadband antenna system (up to 100 MHz) consisting of a parallel plate antenna and a vertical whip antenna, a 12-bit Yokogawa transient recorder and a Meinberg M400 GPS system. The parallel plate antenna was used to detect the fast variation of the electric field (i.e. fast field) while the vertical whip antenna was used to detect the slow variation (i.e. slow field). Each antenna has a buffer amplifier circuit with decay time constants of 15 ms for fast field and 1 s for slow field. The sampling rate for the measurement system was 100 Msample/s with 10 ns interval and the measurements were conducted for a record length of 1 s with 200 ms pre-trigger time. The transient recorder was triggered automatically based on the voltage amplitude of the incoming signal from the parallel plate antenna and the trigger level was set above noise level at 50 mV. Detailed description on the antenna system and the buffer electronic circuits can be found in [19]. Further readings can also be found in [20–23] since the measurement setups used were identical. 2.2. Swedish Lightning Location System Information obtained from the Swedish Lightning Location System (LLS) were used to estimate the location and peak currents of the positive return strokes. The overall lightning flash detection efficiency of the Swedish LLS is about 85% and varies from point to point within Sweden [24]. Around the measuring site, the flash detection efficiency of the LLS is approximately 90% [25]. The
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
3
Fig. 1. Observed number of leader pulses (sample size, N = 51).
detection efficiency of the positive ground flashes, however, is not known. Further information about the Swedish LLS and its performance can be found in [24–26]. 2.3. Sample size The measurements were carried out continuously throughout the entire summer during which a total of 3025 lightning data were recorded. The final analysis was restricted to 51 positive ground flashes located 6–130 km from the measuring station. Since the signatures of the electromagnetic fields depend on the physical processes of the lightning flash, they can be used to identify the different processes associated with the ground flash. The return stroke is indicated by an abrupt and considerable electric field change pulse in the fast field record with coincident field change in the slow field record [27]. For leader process, broadband measurements of the fields indicate that small field pulses appear in the electric field preceding the return stroke [1]. We only considered pulses with peak amplitudes equal to or exceeding twice that of the average noise level since this condition would result in readable pulses for the analysis. Pronounced leader pulses have the same polarity as the positive return stroke. The leader pulses would be considered as having opposite polarity if they had polarity opposite to that of the positive return stroke. 3. Results In the study, electric field changes occurring just before the first return stroke in 51 positive ground flashes were analysed. The analysis was done based on the radiation field pulses associated with the leader process in positive ground flashes. 3.1. Percentage of detection Out of 51 waveforms, pronounced leader pulses were observed in 11 (22%) of the cases. In 25 cases, these pulses were not detected while in eight cases, they have very small amplitudes barely visible over the noise level. In six cases, only one or two pulses were visible just before the return stroke. One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke was also observed. Fig. 1 shows the overview of the observation while Fig. 2 shows typical pronounced leader pulses observed just before a positive return stroke. For the 11 positive ground flashes with pronounced leader pulses, the ratio of the positive ground flashes with pronounced leader pulses to the number of positive ground flashes detected is about 8% when the return stroke distance from the measuring station is close (<20 km), 29% for distance in the 20–50 km range,
36% for the 50–100 km range and 11% for distance greater than 100 km. The ratio increases as the return stroke distance increases but then it decreases when the distance is greater than 100 km. In principle, the radiation field is inversely proportional to the distance [28]. Therefore, more leader pulses occurrences should be observed when the return stroke distance is close. However, when distance is close, static field is dominant and the fine features of the radiation field might be hidden inside the large static field (overwhelmed by the static contribution). On the other hand, when the distance is very large (say, greater than 100 km), the noise level might be comparable to the small radiation field pulses (since radiation field becomes smaller when the distance is larger). These could be the reasons why we only observed a small number of leader pulses occurrence when the distance was close and when the distance was greater than 100 km. The sample size is rather small though (N = 11) and the effect of propagation to the electric fields is not known. Therefore, we could not confirm any clear dependence of the leader pulses occurrences with respect to the return stroke distance at this moment.
3.2. Characteristics of pronounced leader pulses In order to determine the characteristics of the pronounced leader pulses, we obtained the time of initiation of the leader pulses before the first return stroke, the peak amplitude of the leader pulses relative to the return stroke peak and the interpulse duration between successive pulses just before the first return stroke. The statistics and comparison with previous studies are given in Table 1. From Table 1, we observed the leader pulses to start 0.26–1.40 ms before the following first return stroke. This is comparable to 0.14–1.32 ms found by Schumann et al. [5]. On the other hand, Nag and Rakov [15] observed the pulses to appear 0.07–0.63 ms before the first return stroke. We also found that the peak amplitude of the leader pulses relative to the return stroke peak ranged between 2.7 and 17.8% of the return stroke peak with an average of 6.4%. In comparison, Nag and Rakov [15] found the largest amplitude of the step pulses ranged from 8 to 19% of the following return stroke pulse with a mean value of 14%. Schumann et al. [5] found the peak amplitude of the leader pulses ranged between 2.1 and 3.6% of the return stroke peak. We also observed that the interpulse duration between successive pulses just before the first return stroke ranged between 13.3 and 50.3 s with a mean value of 24.7 s. Our mean interpulse duration was slightly higher than those found by Hojo et al. [14] (17.4 s) and Kong et al. [6] (17 s, ranging from 3–31 s) but comparable to that found by Nag and Rakov [15] (20 s, ranging
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model
ARTICLE IN PRESS
EPSR-4971; No. of Pages 7
D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
4
Fig. 2. Typical pronounced leader pulses observed just before a positive return stroke (top) and a larger scale (zoom in) of the waveform (bottom) (from fast field records).
Table 1 Time of initiation, relative peak amplitude and interpulse duration between successive pulses of pronounced leader pulses. Researcher
Present study Schumann et al. [5] Nag and Rakov [15] Cooray and Lundquist [13] Hojo et al. [14] Kong et al. [6] a b
Location
Uppsala, Sweden Brazil, US, Austria Florida, US Uppsala, Sweden Niigata, Japan Shandong, China
Sample size, N
11 14 15 96 7 –
Time of initiation (ms)
Ratio of leader peak to first RS peak (%)
Interpulse duration (s)
Max
Min
AMa
Max
Min
AMa
Max
Min
AMa
1.4 1.32 0.63 – – –
0.26 0.14 0.07 – – –
0.95 0.45 – – – –
17.8 – 19 – – –
2.7 – 8 – – –
6.4 2.1–3.6b 14 – – –
50.3 – 37 48 – 31
13.3 – 5.8 12 – 3
24.7 20.2–22.5 20 26 17.4 17
AM = arithmetic mean. Sample size = 7.
Fig. 3. Time of initiation of the pronounced leader pulses versus the LLS-reported peak current of the positive return strokes (sample size, N = 11).
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
from 5.8–37 s), Schumann et al. [5] (20.2–22.5 s) and Cooray and Lundquist [13] (26 s). Next, we plotted the time of initiation of the pronounced leader pulses against the LLS-reported peak current of the positive return strokes, given in Fig. 3. As seen from Fig. 3, there is a tendency for the time of initiation of the leader pulses to increase with respect to the peak current of the return stroke. It should be noted that the LLSestimated peak current values for the positive return strokes may not be accurate since the peak currents estimated by the lightning detection networks is calibrated against direct current measurements for negative triggered-lightning strokes [29].
3.3. Opposite polarity leader pulses One case of positive ground flash preceded by opposite polarity leader pulses just before the return stroke was also observed. The waveform is shown in Fig. 4 while the data for these pulses are given in Table 2. The distance of this positive ground flash, however, could not be estimated since the Swedish LLS did not record the occurrence. From Table 2, the pulses were observed to appear 0.02 ms before the following first return stroke. Compared to that of the pronounced leader pulses (0.26–1.40 ms), the time of initiation of the negative pulses was much closer to the return stroke. Interpulse duration was found to be 1.78 s (also smaller than the minimum of the pronounced leader pulses of 13.3 s) while the peak amplitude was 3.77% of the return stroke peak.
5
Table 2 Data for opposite polarity leader pulses. Sample size, N
Time of initiation (ms)
Interpulse duration (s)
Ratio of leader peak to first RS peak (%)
1
0.02
1.78
3.77
4. Discussion 4.1. Pronounced leader pulses The presence of these pronounced leader pulses (having the same polarity as the return stroke) suggests that the leaders can propagate in a stepped manner. In a recent study, Saba et al. [3] found that the pulses observed just before the return stroke were solely due to the upward connecting negative leader. They observed that these pulses in the electric field waveforms occur simultaneously with the initiation of the upward connecting negative leader in their video recordings. According to the authors, the initiation of the upward connecting negative leader and the pulses occur at approximately 1 ms before the return stroke, and the time interval between the pulses was approximately 23 s. In the literature concerning negative ground flashes, Krider et al. [30] found that the average time interval between the pulses of the radiation field produced by the negative stepped leaders is about 16 s [30]. Similarly, Cooray and Lundquist [13] found that the time interval between successive leader pulses immediately preceding the radiation fields
Fig. 4. A positive ground flash preceded by opposite polarity leader pulses (i.e. opposite to that of the positive return stroke) (top) and a larger scale (zoom in) of the waveform (bottom) (from fast field records).
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7 6
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
from negative first strokes is about 14 s (ranging between 6 and 30 s) while Ismail et al. [23] found the time interval to be larger than 10 s. Considering that the pulses we observed appear within 1.4 ms before the return stroke (mean 0.95 ms), the mean time interval between the pulses was 24.7 s, the time interval ranged between 13.3 and 50.3 s (within the negative stepped leader values), and the fact that negative leaders always appear to advance in steps [1,2], it is possible that the pulses we observed were due to the upward connecting negative leader. However, since we only have observations from the electric field measurements, we cannot rule out the possibility that it might also come from the downward positive leaders (as well as other possibilities).
4.2. Estimation of the height of the leader from the ground Previous researchers have reported the 2-D (two-dimensional) average speed of the downward-moving positive leaders [4,6,7,31–33]. Saba et al. [7] found that the 2-D average speeds for nine positive leaders ranged from 0.33 to 6.0 × 105 ms−1 (mean of 2.7 × 105 ms−1 ), and the speeds have the tendency to increase by a factor of 1.1–6.5 as the leaders approach the ground. They also made four other measurements and obtained average speeds that ranged between 1.6 and 7.8 × 105 ms−1 [31]. Campos et al. [34] determined the 2-D average speeds for 29 positive leaders and found that the values ranged between 0.24 and 11.80 × 105 ms−1 with an average of 2.76 × 105 ms−1 (also reported in [32]). Wang & Takagi [4] observed that one downward positive leader propagated at a speed of about 1.0 × 106 ms−1 at a height of 272–93 m above the ground before accelerating to a speed of 2.5 × 106 ms−1 at a height of about 45 m. Kong et al. [6] found that the 2-D speed of one stepped-like leader increased from 0.1 × 105 ms−1 to 3.8 × 105 ms−1 as it propagated towards the ground. In a recent study, Kong et al. [33] found that the average 2-D speeds for six positive leaders ranged from 0.3 × 105 ms−1 to 2.0 × 105 ms−1 and they increased as the leaders approached the ground. Based on their findings and our results on the pulses’ time of initiation, we can estimate the height above the ground when the leader pulses begin to appear. Assuming that the leader pulses were due to the downward positive leader, and using an average 2-D speed of 2.7 × 105 ms−1 for the downward leader [7] and the average time of initiation of 0.95 ms, we calculated that the distance travelled by the leader was approximately 256.5 m. Now, as the downward positive leaders approach the ground, they will induce upward-moving connecting leaders that will travel towards the positive leaders. The length of the connecting leader can be several tens of metres [35]. Electrical breakdown will occur when the separation between the two leader tips reaches a certain distance, in which case the two will meet resulting in a return stroke. Depending on the length of the connecting leader, several scenarios can be considered. If the length of the connecting leader is very small compared to the downward leader (i.e. negligible), then the distance that we calculated (i.e. the distance travelled by the downward positive leader) is the height above the ground that the leader pulses begin to appear. The height would increase if the connecting leader has a certain length and originated from a structure. In a recent study, Saba et al. [3] captured images of a positive downward leader and a negative upward connecting leader from a 163 m tower preceding a positive ground flash. The estimated peak current of the positive ground flash was 124 kA. The connecting leader travelled a distance of 270 m and the calculated average 2-D speed of the connecting leader was 3.0 × 105 ms−1 . If we consider these (assuming that the downward leader is attached to a connecting leader initiated from a tall structure), then the leader is estimated
to begin to appear at approximately 689.5 m above the ground (i.e. 256.5 + 270 + 163 = 689.5 m). The connecting leader would be longer if the charge on the leader channel and the height of the structure increase as these conditions promote long connecting leaders [35]. So, if the connecting leader were initiated from a shorter structure, its length might be less than 270 m. It then follows that if the connecting leader were initiated from the ground level, the height above the ground that the leader begin to appear should be less than 526.5 m. If however, we considered that the pulses were due to the upward connecting leader itself, then using the average 2-D speed of 3.0 × 105 ms−1 calculated by [3], we estimated that the connecting leader travelled a distance of approximately 285 m from the ground before attachment with the downward positive leader. If the downward positive leader has a higher current, the connecting leader may be initiated when the downward positive leader is still considerably far from the ground. This will result in longer connecting leader length and greater height of the leader pulses when they begin to appear above the ground. If, however, the downward positive leader has a smaller current, the length of the connecting leader may also be smaller because it will be initiated when the downward leader was closer to ground level. This distance that the connecting leader travelled would then represent an estimate of the striking distance i.e. the distance between the tip of the downward moving leader and the point of origin of the connecting leader when they meet [35]. 4.3. Opposite polarity leader pulses For the opposite polarity leader pulses, they were observed to start 0.02 ms before the following positive return stroke. The interpulse duration was 1.78 s and the peak amplitude was 3.77% of the return stroke peak. Both downward positive leader and upward connecting leader in positive ground flashes produce pulses with the same polarity as the positive return stroke in the electric field waveforms. However, the pulses we observed in this case were of the opposite polarity to that of the positive return stroke. Considering that they occurred during the leader process in positive ground flashes, these pulses may be due to the negatively-charged leader branch of a bi-directional leader propagating inside the cloud. If we consider the negatively-charged leader branch moving towards an observer, this would give rise to pulses with opposite polarity (to that of the positive return stroke) in the electric field waveforms. However, if the negative leader branch is moving away from the observer, this would give rise to same polarity pulses (to that of the positive return stroke) in the waveforms. Pulse bursts like these have been observed previously by Krider et al. [36] who recorded sequences or bursts of uniform pulses during intracloud lightning discharges, with time intervals between the pulses of typically 5 s. They suggested that based on the pulse shape and time interval, the source of these pulses could be an intracloud dart-stepped leader process. The pulses that we recorded would then be the first observation that such pulses occur in positive ground flash. 5. Conclusion The presence of the same polarity pulses suggests that the leaders can propagate in a stepped manner. It is possible that the pulses were due to the upward connecting negative leader developed under the influence of the downward positive leader. However, since we only have electric field measurements, we cannot rule out the possibility that it might also come from the downward positive leaders since both would produce same polarity pulses (to that of the positive return stroke) in the electric field waveforms.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026
G Model EPSR-4971; No. of Pages 7
ARTICLE IN PRESS D. Johari et al. / Electric Power Systems Research xxx (2016) xxx–xxx
If these pulses were attributed to the downward positive leader, the time of initiation obtained from the electric field records can be used to estimate the height the leader begin to appear from the ground. If however, the pulses were attributed to the upward connecting negative leader, the distance the connecting leader travelled would be an estimate of the striking distance of the ground flash. If this is the case, the striking distance would depend on the current on the leader channel and the height of the structure that it originated from. One case of positive ground flash preceded by opposite polarity pulses just before the return stroke was also observed. These opposite polarity pulses could be due to the negatively-charged leader branch of a bi-directional leader inside the cloud that propagates towards an observation point. Acknowledgements The authors would like to express their gratitude to all the people who have directly or indirectly contribute towards the successful completion of this paper. Participation of Dalina Johari is funded by the Ministry of Education of Malaysia and Universiti Teknologi MARA Malaysia. Participation of Prof. Vernon Cooray and Dr. Mahbubur Rahman are funded by the fund from B. John F. and Svea Andersson donation at Uppsala University. Participation of Mohd Muzafar Ismail is funded by the Ministry of Education of Malaysia and Universiti Teknikal Malaysia Melaka. We would also like to thank Thomas Götschl for the assistance in the measurement setup and acquisition of lightning data from the Swedish LLS database. Finally, the authors would like to acknowledge the Division of Electricity, Ångström Laboratory, Uppsala University, for the excellent facility provided to carry out this research. References [1] V. Cooray, Mechanism of the lightning flash, in: V. Cooray (Ed.), The Lightning Flash, 2nd ed., The Institution of Engineering and Technology, London, United Kingdom, 2014, pp. 119–229. [2] V.A. Rakov, M.A. Uman, Positive and bipolar lightning discharges to ground, in: Lightning: Physics and Effects, Cambridge University Press, 2003, pp. 214–240. [3] M.M.F. Saba, C. Schumann, T. a. Warner, J.H. Helsdon, R.E. Orville, High-speed video and electric field observation of a negative upward leader connecting a downward positive leader in a positive cloud-to-ground flash, Electr. Power Syst. Res. 118 (2015) 89–92. [4] D. Wang, N. Takagi, A downward positive leader that radiated optical pulses like a negative stepped leader, J. Geophys. Res. 116 (May (D10)) (2011) D10205. [5] C. Schumann, M.M.F. Saba, L.Z.S. Campos, R.B.G. Silva, W. Schulz, Leader characteristics in positive cloud-to-ground lightning flashes, International Symposium on Winter Lightning (ISWL) (2011) 59–62. [6] X. Kong, X. Qie, Y. Zhao, Characteristics of downward leader in a positive cloud-to-ground lightning flash observed by high-speed video camera and electric field changes, Geophys. Res. Lett. 35 (5) (2008). [7] M.M.F. Saba, K.L. Cummins, T.A. Warner, E.P. Krider, L.Z.S. Campos, M.G. Ballarotti, O. Pinto, S.A. Fleenor, Positive leader characteristics from high-speed video observations, Geophys. Res. Lett. 35 (7) (2008) L07802. [8] J.D. Hill, M.A. Uman, D.M. Jordan, High-speed video observations of a lightning stepped leader, J. Geophys. Res. Atmos. 116 (16) (2011). [9] M. Stolzenburg, T.C. Marshall, S. Karunarathne, N. Karunarathna, R.E. Orville, Transient luminosity along negative stepped leaders in lightning, J. Geophys. Res. D Atmos. 120 (8) (2015) 3408–3435. [10] D.A. Petersen, W.H. Beasley, High-speed video observations of a natural negative stepped leader and subsequent dart-stepped leader, J. Geophys. Res. Atmos. 118 (21) (2013) 12110–12119.
7
[11] W. Lu, L. Chen, Y. Ma, V.A. Rakov, Y. Gao, Y. Zhang, Q. Yin, Y. Zhang, Lightning attachment process involving connection of the downward negative leader to the lateral surface of the upward connecting leader, Geophys. Res. Lett. 40 (20) (2013) 5531–5535. [12] V. Mazur, L.H. Ruhnke, The physics of lightning flash development, in: V. Cooray (Ed.), Lightning Electromagnetics, 1st ed., Institution of Engineering and Technology, 2012, pp. 193–229. [13] V. Cooray, S. Lundquist, On the characteristics of some radiation fields from lightning and their possible origin in positive ground flashes, J. Geophys. Res. 87 (1982) 11203. [14] J. Hojo, M. Ishii, T. Kawamura, F. Suzuki, R. Funayama, The fine structure in the field change produced by positive ground strokes, J. Geophys. Res. 90 (1985) 6139–6143. [15] A. Nag, V.A. Rakov, Positive lightning: An overview, new observations, and inferences, J. Geophys. Res. 117 (2012) D08109. [16] K. Berger, Novel observations on lightning discharges: Results of research on Mount San Salvatore, J. Franklin Inst. 283 (6) (1967) 478–525. [17] Les Renardieres Group, Positive discharges in long air gaps at Les Renardieres —1975. Results and conclusions, Electra 53 (1975) 31–153. [18] V. Cooray, Electromagnetic fields of lightning flashes, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 135–165. [19] A. Galvan, M. Fernando, Operative Characteristics of a Parallel-Plate Antenna to Measure Vertical Electric Fields from Lightning Fields from Lightning Flashes, Report UURIE 285-00, Uppsala University, 2000. [20] Z.A. Baharudin, N.A. Ahmad, M. Fernando, V. Cooray, J.S. Mäkelä, Comparative study on preliminary breakdown pulse trains observed in Johor, Malaysia and Florida, USA, Atmos. Res. 117 (2012) 111–121. [21] Z.A. Baharudin, M. Fernando, N.A. Ahmad, J.S. Mäkelä, M. Rahman, V. Cooray, Electric field changes generated by the preliminary breakdown for the negative cloud-to-ground lightning flashes in Malaysia and Sweden, J. Atmos. Sol. Terr. Phys. 84–85 (2012) 15–24. [22] D. Johari, V. Cooray, M. Rahman, P. Hettiarachchi, M.M. Ismail, Characteristics of preliminary breakdown pulses in positive ground flashes during summer thunderstorms in Sweden, Atmosphere (Basel) 7 (3) (2016) 39. [23] M.M. Ismail, M. Rahman, V. Cooray, S. Sharma, P. Hettiarachchi, D. Johari, Electric field signatures in wideband, 3 MHz and 30 MHz of negative ground flashes pertinent to Swedish thunderstorms, Atmosphere (Basel) 6 (12) (2015) 1904–1925. [24] U. Sonnadara, V. Cooray, T. Götschl, Characteristics of cloud-to-ground lightning flashes over Sweden, Phys. Scr. 74 (5) (2006) 541–548. [25] M. Fernando, A. Galván, T. Götschl, V. Cooray, V. Scuka, Analysis of Swedish lightning using ‘LLP’ data, Light. Prot. (ICLP), 1998 Int. Conf. (1998) 150–155. [26] U. Sonnadara, V. Kathriarachchi, V. Cooray, R. Montano, T. Götschl, Performance of lightning locating systems in extracting lightning flash characteristics, J. Atmos. Sol. Terr. Phys. 112 (2014) 31–37. [27] D.M. Fuquay, Positive cloud-to-ground lightning in summer thunderstorms, J. Geophys. Res. Ocean 87 (August (C9)) (1982) 7131–7140. [28] V. Cooray, Basic electromagnetic theory with special attention to lightning electromagentics, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 29–58. [29] A. Nag, V.A. Rakov, K.L. Cummins, Positive lightning peak currents reported by the U.S. national lightning detection network, IEEE Trans. Electromagn. Compat. 56 (2014) 404–412. [30] E.P. Krider, C.D. Weidman, C.R. Noggle, The electric fields produced by lightning stepped leaders, J. Geophys. Res. 82 (6) (1977) 951–960. [31] M.M.F. Saba, M.G. Ballarotti, L.Z.S. Campos, O.P. Jr, High-speed video observations of positive lightning, International Symposium on Lightning Protection (IX SIPDA) (2007), pp. 1–5. [32] M.M.F. Saba, W. Schulz, T.A. Warner, L.Z.S. Campos, C. Schumann, E.P. Krider, K.L. Cummins, R.E. Orville, High-speed video observations of positive lightning flashes to ground, J. Geophys. Res. 115 (D24) (2010). [33] X. Kong, Y. Zhao, T. Zhang, H. Wang, Optical and electrical characteristics of in-cloud discharge activity and downward leaders in positive cloud-to-ground lightning flashes, Atmos. Res. 160 (2015) 28–38. [34] L.Z.S. Campos, M.M.F. Saba, T.A. Warner, O. Pinto, E.P. Krider, R.E. Orville, High-speed video observations of natural cloud-to-ground lightning leaders—a statistical analysis, Atmos. Res. 135–136 (2014) 285–305. [35] V. Cooray, Mechanism of lightning flash, in: An Introduction to Lightning, Springer, Netherlands, 2015, pp. 91–116. [36] E.P. Krider, G.J. Radda, R.C. Noggle, Regular radiation field pulses produced by intracloud lightning discharges, J. Geophys. Res. 80 (27) (1975) 3801–3804.
Please cite this article in press as: D. Johari, et al., Characteristics of leader pulses in positive ground flashes in Sweden, Electr. Power Syst. Res. (2016), http://dx.doi.org/10.1016/j.epsr.2016.11.026