IX International Symposium on
Lightning Protection
26th-30th November 2007 – Foz do Iguaçu, Brazil
LIGHTNING PHENOMENOLOGY AND PARAMETERS IMPORTANT
FOR LIGHTNING PROTECTION
V.A. Rakov
University of Florida
[email protected]
P.O. Box 116130, Gainesville FL 32611-6130
Abstract - Phenomenology and various parameters of lightning discharges are reviewed. Both common negative and less
frequent but potentially more destructive positive and bipolar flashes are considered. Although positive lightning discharges
account for 10% or less of global cloud-to-ground lightning activity, there are several situations that appear to be conducive to
the more frequent occurrence of positive lightning. Bipolar lightning flashes occur not less often than positive flashes, at least
when tall strike objects are involved.
1 INTRODUCTION
Lightning, or the lightning discharge, in its entirety, whether it strikes ground or not, is usually termed a “lightning flash”
or just a “flash”. A lightning discharge that involves an object on ground or in the atmosphere is sometimes referred to
as “lightning strike”. A commonly used non-technical term for lightning discharge is “lightning bolt”. The terms
“stroke” or “component stroke” apply only to components of cloud-to-ground discharges. Each stroke involves a
downward leader and an upward return stroke and may involve a relatively low level “continuing current” that
immediately follows the return stroke. Transient processes occurring in a lightning channel while it carries continuing
current are termed M-components. First strokes are initiated by “stepped” leaders while subsequent strokes following
previously formed channels are initiated by “dart” or “dart-stepped” leaders.
The global lightning flash rate is some tens to a hundred per second or so. The majority of lightning discharges,
probably three quarters, do not involve ground. These are termed cloud discharges and sometimes are referred as ICs.
Cloud discharges include intracloud, intercloud, and cloud-to-air discharges (see Fig.1).
From the observed polarity of the charge effectively lowered to ground and the direction of propagation of the initial
leader, four different types of lightning discharges between cloud and Earth have been identified. The term “effectively”
is used to indicate that individual charges are not transported all the way from the cloud to ground during the lightning
processes. Rather, the flow of electrons (the primary charge carriers) in one part of the lightning channel results in the
flow of other electrons in other parts of the channel, as discussed by Uman (1987) [1]. For example, individual electrons
in the lightning channel move only a few meters during a return stroke that transfers a coulomb or more of charge to
ground.
The four types of cloud-to-ground lightning, illustrated in the Fig. 2, are (a) downward negative lightning, (b) upward
negative lightning, (c) downward positive lightning, and (d) upward positive lightning. Discharges of all four types can
be viewed as effectively transporting cloud charge to the ground and therefore are usually termed cloud-to-ground
discharges (sometimes referred to as CGs). It is believed that downward negative lightning flashes, type (a), account for
about 90% or more of global cloud-to-ground lightning, and that 10% or less of cloud-to-ground discharges are
downward positive lightning flashes (type (c)). Upward lightning discharges, types (b) and (d), are thought to occur only
from tall objects (higher than 100 m or so) or from objects of moderate height located on mountain tops. The
classification of lightning discharges shown in Fig. 2 covers only “unipolar” flashes that transport to ground charge of
one polarity. This classification should be extended to additionally include bipolar lightning flashes transporting to
ground both negative and positive charges. The latter type of cloud-to-ground discharges is discussed in Section 4.
Fig. 1 - General classification of lightning discharges.
Fig 2 - Four types of lightning effectively lowering cloud charge to ground. Only the initial leader is shown for each type. In each
lightning-type name given below the sketch, the direction of propagation of the initial leader and the polarity of the cloud charge
effectively lowered to ground are indicated.
2 NEGATIVE LIGHTNING
2.1 General characterization
A typical negative cloud-to-ground flash is composed of 3 to 5 strokes (leader/return stroke sequences), with the
geometric mean interstroke interval being about 60 ms. Occasionally, two leader/return stroke sequences occur in the
same lightning channel with a time interval between them as short as 1 ms or less. The percentages of single-stroke
flashes observed, using accurate stroke count methods, in different countries are summarized in Table 1. It follows from
Table 1 that the percentage of single-stroke flashes presently recommended by CIGRE, 45% (Anderson and Eriksson,
1980 [2]), is likely to be an overestimate by a factor of two or so. Roughly half lf all lightning discharges to earth, both
single- and multiple-stroke flashes, strike ground at more than one point with the spatial separation between the
channel terminations being up to many kilometres. The correction factor of about 1.5 to 1.7 is needed for measured
values of ground flash density to account for multiple channel terminations on ground (Kitagawa et al., 1962 [3]; Rakov
and Uman, 1990 [4]; Valine and Krider, 2002 [5]; Saba et al., 2006 [6]). First-stroke current peaks are typically a factor
of 2 to 3 larger than subsequent-stroke current peaks. However, about one third of cloud-to-ground flashes contain at
least one subsequent stroke with electric field peak, and, by theory, current peak, greater than the first-stroke peak.
Table 1: Number of strokes per flash and percentage of single-stroke flashes.
2.2 Parameters derived from direct current measurements
The lightning current parameters for 101 downward negative flashes, the types that normally strike flat terrain and
structures of moderate height, are summarized by Berger et al. (1975 [9]). This summary, based on direct current
measurements in Switzerland, is used to a large extent as a primary reference in the literature on both lightning
protection and lightning research and is reproduced in Table 2 . The table gives the percentages (95%, 50%, and 5%) of
cases exceeding the tabulated values, based on the log-normal approximations to the respective statistical distributions.
The action integral represents the energy that would be dissipated in a 1- resistor if the lightning current were to flow
through it. All the parameters presented in Table 2 are estimated from current oscillograms with the shortest measurable
time being 0.5 µs (Berger and Garbagnati, 1984 [10]). It is thought that in Table 2 the distribution of front durations
might be biased toward larger values and the distribution of di/dt toward smaller values. Anderson and Eriksson (1980)
[2] digitized the return-stroke current oscillograms of Berger et al. (1975) [9] and determined additional wavefront
parameters. A similar analysis for 31 negative downward flashes in Brazil is presented by Visacro et al. (2004) [11].
Peak currents in Brazil appear to be somewhat larger than in Switzerland.
The median 10-90% risetime estimated for subsequent strokes by Anderson and Eriksson (1980) [2] from Berger et al.’s
(1975) [9] oscillograms is 0.6 µs, comparable to the median values ranging from 0.3 to 0.6 µs for triggered lightning
strokes (Leteinturier et al. 1991 [12]; Fisher at al. 1993 [13]). The median 10-90% current rate of rise reported for
natural subsequent strokes by Anderson and Eriksson (1980) [2] is 15 kA/µs, almost three times lower than the
corresponding value of 44 kA/µs in data of Leteinturier at al. (1991) [12] and more than twice lower than the value of
34 kA/µs found by Fisher et al. (1993) [13]. The largest value of maximum rate of rise of 411 kA/µs (see Fig. 3) was
measured by Leteinturier et al. (1991 [12], Fig. 4) for a triggered lightning stroke terminating on a launcher grounded to
salt water. The corresponding directly measured current was greater than 60 kA, the largest reported for summer
triggered lightning. The mean value of current derivative peak reported by Leteinturier et al. (1991) [12] is 110 kA/µs.
The higher observed values of current rate of rise for triggered-lightning return strokes than for natural return strokes are
likely due to the use of better instrumentation (digital oscilloscopes with better upper frequency response), although the
influence of lightning triggering conditions has not yet been ruled out.
Fig. 3 - Relation between peak value of current rate of rise and current from triggered-lightning experiments conducted at the NASA
Kennedy Space Center, Florida, in 1985, 1987, and 1988 and in France in 1986. The regression line for each year is shown, and the
sample size and the regression equation are given. Adapted from Leteinturier et al. (1991) [12].
The minimum peak current value included in Berger’s distributions is 2 kA. Clearly, the parameters of statistical
distributions can be affected by the lower and upper measurement limits. Rakov (1985) [14] showed that, for a lognormal distribution, the parameters of a measured, “truncated” distribution and a knowledge of the lower measurement
limit can be used to recover the parameters of the actual, “untruncated” distribution. He applied the recovery procedure
to the various lightning peak current distributions found in the literature and concluded that the peak current
distributions published by Berger at al. (1975) [9] can be viewed as practically unaffected by the lower measurement
limit of 2 kA (actually no first strokes with peak currents less than 5 kA were observed). Note from Table 2 that the
median return-stroke current peak for first strokes is 2 to 3 times higher than that for subsequent strokes. Also, negative
first strokes transfer about a factor of four larger total charge than do negative subsequent strokes. On the other hand,
subsequent return strokes are characterized by 3 to 4 times higher maximum steepness (the current maximum rate of
rise). Only a few percent of negative first strokes are expected to exceed 100 kA, while about 20% of positive strokes
have been observed to do so, although the 50-% (median) values, which for a log normal distribution are equal to the
geometric mean values, are rather similar. It is worth noting that directly measured current waveforms of either polarity
found in the literature do not exhibit peaks exceeding 300 kA or so, although less reliable peak current estimates from
the residual magnetization of ferromagnetic detectors (magnetic links) and inferences from remotely measured electric
and magnetic fields suggest the existence of currents up to 500 kA and even higher (e.g., Le Boulch and Plantier 1990
[15]). Lyons et al. (1998b) [16], using U.S. National Lightning Detection Network (NLDN) data for 14 selected
summer months from 1991-1995, reported that the largest current peaks were 957 and 580 kA for negative and positive
flashes, respectively. It is important to note that peak current estimates reported by the NLDN and by other similar
systems are based on an empirical formula the validity of which has been tested, using triggered lightning in Florida,
only for negative subsequent strokes with peak currents not exceeding 60 kA (Orville, 1999 [17]; Jerauld et al. 2005
[18]).
Table 2: Lightning current parameters for negative flashes (Berger et al., 1975 [9])
Parameters
Peak current (minimum 2 kA)
First strokes
Subsequent strokes
Charge (total charge)
First strokes
Subsequent strokes
Complete flash
Impulse charge
(excluding continuing current)
First strokes
Subsequent strokes
Front duration (2 kA to peak)
First strokes
Subsequent strokes
Maximum di/dt
First strokes
Subsequent strokes
Stroke duration
(2 kA to half peak value on the
tail)
First strokes
Subsequent strokes
Action integral ( i2dt)
First strokes
Subsequent strokes
Time interval between strokes
Flash duration
All flashes
Excluding single-stroke flashes
Units
Sample Size
Percent Exceeding Tabulated Value
95%
50%
5%
kA
101
135
14
4.6
30
12
80
30
C
93
122
94
1.1
0.2
1.3
5.2
1.4
7.5
24
11
40
90
117
1.1
.22
4.5
0.95
20
4
s
89
118
1.8
.22
5.5
1.1
18
4.5
kA/ s
92
122
5.5
12
12
40
32
120
90
115
30
6.5
75
32
200
140
91
88
133
6.0x103
5.5x102
7
5.5x104
6.0x103
33
5.5x105
5.2x104
150
94
39
0.15
31
13
180
1100
900
C
s
A2s
ms
ms
2.3 Return-stroke speed
Return-stroke speeds are usually measured using optical techniques. The optically measured return-stroke speed
probably represents the speed of the region of the upward-moving return-stroke tip where power losses are greatest, the
power per unit length being the product of the current and the longitudinal electric field in the channel.
Lightning return stroke speed is an important parameter in lightning protection studies. Some researchers (e.g.,
Lundholm, 1957 [19]; Wagner, 1963 [20]) have suggested that the return-stroke speed should increase with increasing
peak current. If such relationship indeed existed, which is generally not supported by experimental data, it could be used
in relating return-stroke peak current to the electric potential of the preceding leader and estimating the lightning striking
distance (e.g., Hileman, 1999 [21], pp. 221-222). Further, return-stroke speed is a parameter in the models used in
evaluating lightning-induced effects in power and communication lines (e.g., Rachidi et al., 1996 [22]). Finally, an
explicit or implicit assumption of the return-stroke speed is involved in inferring lightning currents from remotely
measured electric and magnetic fields (e.g., Norinder and Dahle, 1945 [23]; Uman and McLain, 1970b [24]; Uman et
al., 1973a, b [25,26]; Dulzon and Rakov, 1980 [27]; Krider et al., 1996 [28]; Cummins et al., 1998 [29]; Rakov, 2005
[30]).
Schonland et al. (1935) [31] found that the first return stroke speed at the channel base was typically near 1 x 108 m/s,
and at the top of the main channel it was typically near 5 x 107 m/s. A summary of more recently measured return-stroke
speeds averaged over the lowest some hundreds of meters of the channel for both natural and rocket-triggered lightning
is given in Table 3. The more recently measured return-stroke speeds presented in Table 3 (typically between one-third
and one-half of the speed of light) are generally higher than the earlier results of Schonland et al. (1935) [31], probably
due in part to the fact that the recent measurements were made closer to the ground where the return-stroke speed tends
to be higher.
Table 3: Summary of measured return stroke speeds in natural and triggered negative lightning.
Reference
Min.
Speed,
m/s
Max. Speed
m/s
Mean Speed,
m/s
St. Dev.
m/s
Sample
Size
2.0 x 107
1.2 x 108
0.71 x 108
2.6 x 107
12
2.9 x 107
2.4 x 108
1.1 x 108
4.7 x 107
63
2.0 x 107
8.0 x 107
2.6 x 108
>2.8 x 108
5 x 107
7 x 107
54
43
4.5 x 107
1.7 x 108
9.9 x 107
4.1 x 107
13
6.7 x 107
1.7 x 108
1.2 x 108
2.7 x 107
56
1.0 x 108
1.5 x 108
1.2 x 108
1.6 x 107
9
1.2 x 108
1.9 x 108
1.5 x 108
1.7 x 107
18
6.0 x 107
6.0 x 107
1.6 x 108
2.0 x 108
2 x 107
4 x 107
40
39
Comments
Natural Lightning
Boyle and Orville
(1976) [32]
Idone and Orville
(1982) [33]
Mach and Rust
(1989a, Fig. 7) [34]
1.3±0.3 x 108
1.9±0.7 x 108
Streak camera, 2-D
speed
Streak camera, 2-D
speed
Long channel
Short channel
(Photoelectric, 2-D)
Triggered Lightning
Hubert and Mouget
(1981) [35]
Idone et al. (1984)
[36]
Willett et al. (1988)
[37]
Willett et al. (1989a)
[38]
Mach and Rust
(1989a, Fig. 8) [34]
1.2±0.3 x108
1.4±0.4 x108
Photoelectric, 3-D
speed
Streak camera, 3-D
speed
Streak camera, 2-D
speed
Streak camera, 2-D
speed
Long channel
Short channel
(Photoelectric, 2-D)
We now review the optical measurements of lightning return-stroke speed within the bottom 100 m or so of the channel.
This channel segment corresponds to the time when the initial peaks of the channel-base current (the typical 10-90%
risetime of subsequent return-stroke currents is 0.3 to 0.6 µs (see Fisher et al., 1993 [13], Fig. 6) are formed. It is this
value of speed that is needed for estimating the current peak from measured radiation field peak and distance, using
simple return-stroke models (e.g., Rakov and Uman, 1998 [39]).
Wang et al. (1999c) [40] reported on two-dimensional speed profiles within 400 m of the ground for two return strokes
in triggered lightning. These speed profiles were obtained in 1997 at Camp Blanding, Florida, using the digital optical
imaging system ALPS having a time resolution of 100 ns and a spatial resolution of 30 m. The return-stroke speeds
within the bottom 60 m of the channel were found to be 1.3 x 108 and 1.5 x 108 m/s.
Olsen et al. (2004) [41], using a vertical array of four photodiodes, estimated return-stroke speeds in the bottom 170 m
of the channel for five strokes in one flash triggered at Camp Blanding, Florida, in 2003. Return-stroke speed values
estimated tracking the 20% of the peak point on the front of return-stroke light pulse for three different segments of the
lightning channel, 7 to 63 m, 63 to 117 m, and 117 to 170 m, are summarized in Table 4. For the lowest channel
segment, 7 to 63 m, the speed values are 1.2 to 1.3 x 108 m/s. For higher channel segments, speed values are generally
higher, varying from 1.6 to 1.8 x 108 m/s over the 63 to 117 m segment and from 1.2 to 1.7 x 108 m/s over the 117 to
170 m segment. The speed tends to be higher if a point lower than 20% of the peak is tracked. On the other hand, some
speed values obtained tracking the 10% point were significantly affected by noise. Speeds over the channel segment
from 7 to 117 m estimated using the so-called slope-intercept method, which probably yielded the upper speed bound,
ranged from 1.8 to 2.3 x 108 m/s. For comparison, the speed range found for the same channel segment but tracking the
20% point was 1.4 to 1.5 x 108 m/s.
Thus, the return-stroke speed in the bottom some tens of meters to 100 m of the lightning channel, that is, at the time
when the initial peak of the channel-base current is formed is typically one-third to two- thirds of the speed of light.
Table 4: Return-stroke speeds (x108 m/s) estimated tracking the 20% point on the light-pulse front for triggered lightning
flash F0336 (Olsen et al., 2004 [41]) .
Height
range, m
7 – 63
63 – 117
117 – 170
1
1.3
1.6
1.7
2
1.2
1.8
1.2
Stroke order
4
5
1.2
1.2
1.8
1.8
1.5
1.6
6
1.2
1.6
1.5
Estimated
error, %
10
15
21
No data are available for stroke 3.
Idone and Orville (1982) [33] found that the negative return-stroke speed usually decreases with height, for both first
and subsequent strokes, by 25% or more over the visible part of channel relative to the speed near ground. In computing
lightning electromagnetic fields, the return-stroke speed is often assumed to be constant (particularly for subsequent
strokes) over the radiating channel section (e.g., Rakov and Uman, 1998 [39]). Gorin (1985) [42] suggested a nonmonotonic return-stroke speed profile. According to his nonlinear distributed-circuit model for a first stroke, the speed
initially increases to its maximum over a channel length of the order of some hundreds of meters and decreases
thereafter. The initial speed increase in Gorin’s (1985) [42] model is associated with the so-called break-through phase
(also called the final jump or switch-closing phase) thought to be responsible for the formation of the initial rising
portion of the return-stroke current pulse (see also, Rakov and Dulzon 1991 [43]; Rakov et al. 1992b [44]). Srivastava
(1966) [45] proposed, based on the experimental data published by Schonland (1956) [46], a bi-exponential expression
for the first return-stroke speed as a function of time, according to which the speed rises from zero to its peak and falls
off afterwards. Variation of the return-stroke speed with height for triggered-lightning strokes 2, 4, 5, and 6 in Table 4
suggests that the speed indeed initially increases and then decreases with increasing height.
2.4 Summary
The downward negative lightning discharge to ground is the most common and most studied type of cloud-to-ground
lightning. Each ground flash typically contains three to five strokes. The maximum number of strokes per flash observed
to date is 26. The overwhelming majority, about 80% or more, of flashes contain more than one stroke. The time interval
between successive return strokes in a flash is usually several tens of milliseconds, although it can be as large as many
hundreds of milliseconds if a long continuing current is involved or as small as one millisecond or less. The total
duration of a flash is typically some hundreds of milliseconds. The total charge lowered to ground is some tens of
coulombs. About half of all negative ground flashes create more than one termination on ground, the spatial separation
between the channel terminations being up to many kilometres. Each stroke is composed of a downward-moving leader
and an upward-moving return stroke. The leader creates a conductive path between the cloud charge source and ground
and deposits negative charge along this path, while the return stroke traverses that path, moving from ground toward the
cloud charge source, and neutralizes the negative leader charge. The stepped leader initiating the first return stroke
moves intermittently while subsequent-stroke leaders usually appear to move continuously. The return-stroke speed in
the bottom 100 m of the lightning channel is typically one-third to two-thirds of the speed of light.
In the remainder of this review, the focus is placed on considerably less studied positive and bipolar lightning
discharges, although the salient properties of downward negative lightning discharges are summarized in Section 5.
3 POSITIVE LIGHTNING
Positive lightning discharges (flashes) are defined as those effectively transporting positive charge from cloud to Earth.
It is thought that less than 10% of global cloud-to-ground lightning is positive (e.g., Uman 1987 [1]).
Positive lightning discharges have recently attracted considerable attention for the following reasons:
1) The highest recorded lightning currents (near 300 kA) and the largest charge transfers to ground (hundreds of
coulombs or even more) are thought to be associated with positive lightning (see Fig. 4).
2) Positive lightning can be the dominant type of cloud-to-ground lightning during the cold season, during the
dissipating stage of a thunderstorm, and in some other situations, discussed in Section 3.1.
3) Positive lightning has been recently found to be preferentially related to luminous phenomena in the middle and
upper atmosphere known as sprites (e.g., Lyons et al. 1998a, b [47,16]).
4) Reliable identification of positive discharges by lightning locating systems (LLS), such as the U.S. National
Lightning Detection Network (NLDN), has important implications for various meteorological and other studies that
depend on LLS data (e.g., Petersen and Rutledge 1992 [48]; Seimon 1993 [49]; MacGorman and Burgess 1994 [50];
MacGorman and Morgenstern 1998 [51]; Lyons et al. 1998a [47]).
On the other hand, positive lightning largely remains a mystery. Experimental data on this type of lightning discharge
are limited and often controversial.
Fig. 4 - Directly measured currents in three positive lightning discharges in Japan. Note the very large peaks, 340, 320, and 280 kA,
of the initial pulses followed by low-level continuing currents whose durations are of the order of 10 ms (top and middle panels) or
100 ms (bottom panel). The middle and bottom panels have inserts (labeled “Expansion”) that show the same current waveform, but
on an expanded (1 ms) time scale. Transferred charges (currents integrated over time) are 330, 180, and 400 C, respectively.
Adapted from Goto and Narita (1995) [52].
3.1 Conditions conducive to the occurrence of positive lightning
Although the overall percentage of positive lightning discharges is relatively low, there are five situations, listed below,
that appear to be conducive to the more frequent occurrence of such discharges. The genesis of positive lightning in
these situations is not fully understood yet.
(1) The dissipating stage of an individual thunderstorm. The tendency for positive lightning to occur toward the end of a
thunderstorm has been reported, for example, by Fuquay (1982) [53] and by Orville et al. (1983) [54]. Pierce (1955b)
[55] suggested that positive flashes are initiated from the upper (main) positive charge region of thunderclouds after
much of the main negative charge, located below the main positive charge, has been removed by negative ground
flashes. On the other hand, Krehbiel (1981) [56] reported on three positive flashes in Florida that apparently involved,
or were a byproduct of, long (longer than 40 km) horizontal lightning discharges that effectively removed positive
charge from a layer near the 0º C isotherm where frozen precipitation was melting (that is, from a region that was
considerably lower than the main positive charge region). It has been recently suggested (Rust and MacGorman 2002
[57]) that some clouds may have “inverted” charge structure (the main positive charge below the main negative charge),
which can facilitate the production of positive lightning by such clouds.
(2) Winter thunderstorms. There is a clear tendency for positive lightning to occur during the cold season. Orville et
al. (1987) [58], using one year'
s data from the U.S. East Coast lightning locating network found that positive lightning
accounted for about 80% of all ground discharges in the northeastern United States in February and for less than 5%
during the summer. From the NLDN data for 1992-1995, Orville and Silver (1997) [59] reported that the monthly
percentage of positive flashes over the contiguous United States ranged from 3% (August 1992) to about 25%
(December 1993). In 1995-1997, the smallest percentage (6.5%) was observed in July 1995, and the largest (about
25%) in January 1996 (Orville and Huffines, 1999 [60]). In the coastal area of the Sea of Japan, the percentage of
positive flashes reached a maximum of 60% in December (Hojo et al. 1989 [61]). In France, a maximum percentage of
positive flashes of 44% was observed in February (Le Boulch and Plantier 1990 [15]).
The production of large positive lightning discharges by winter storms in Japan has been reported by Takeuti et al.
(1978) [62] and by Brook et al. (1982) [63] from multiple-station electric field measurements in conjunction with optical
observations. Brook et al. (1982) [63] observed that positive flashes constituted about 40% of the total number of
cloud-to-ground flashes. About one-third of winter lightning currents recorded by Miyake et al. (1992) [64] on two
towers of 88-m and 200-m height in Japan were of positive polarity. Brook et al. (1982) [63] suggested that positive
flashes originated from the upper positive charge that was displaced horizontally by vertical wind shear from the lower
negative charge and thereby exposed to the ground, which implies that winter clouds have a charge structure similar to
that of summer clouds. Later studies (Kitagawa and Michimoto 1994 [65]), however, indicate that some winter
thunderclouds in Japan may contain predominantly positive charge, this is, may have the essentially monopolar charge
structure, during most of their life cycle. According to Kitagawa and Michimoto (1994) [65], graupel particles, which
are thought to be both the main carriers of negative charge and also the carriers of the lower positive charge, are present
in the cloud for only a relatively short period of time. As a result, the lifetime of the dipolar or tripolar charge structure
is very short, usually less than 10 minutes in early or late winter and less than several minutes in midwinter.
(3) Trailing stratiform regions of mesoscale convective systems (MCSs). The production of predominantly positive
flashes by relatively shallow clouds, including the trailing stratiform regions of MCSs (Engholm et al. 1990 [66]), has
been observed in both winter and summer seasons. This observation might be due largely to the tendency for the
occurrence of negative flashes to decrease dramatically with decreasing cloud depth. (It is worth noting that the
difference between the main convective region and the trailing stratiform region of an MCS in terms of cloud depth is
not very large.) Some thunderstorm systems produce positive and negative flashes whose ground strike locations tend to
be separated in space by polarity, thereby forming the "bipolar" pattern (Orville et al. 1988 [67]). It has been found that
most positive flashes are associated with the relatively shallow region of the system, while negative flashes tend to occur
in the deepest-convection region. Perhaps winter thunderstorms, discussed above, can be included in this category since
their depth is usually small.
(4) Severe storms. The occurrence of infrequent, widely scattered positive flashes in the mature and later stages of
severe springtime storms over the Great Plains in the United States was first observed by Rust et al. (1981) [68], who
used electric field measurements in conjunction with optical observations. More recently, some severe storms have been
observed (e.g., Seimon 1993 [49]; MacGorman and Burgess 1994 [50]; Stolzenburg 1994 [69]; Perez et al. 1997 [70];
Carey and Rutledge 1998 [71]; Williams 2001 [72]) in which the positive flashes detected by lightning locating systems
such as the NLDN occurred relatively frequently and outnumbered negative ground flashes for more than 30 minutes.
The period of a storm'
s lifetime in which positive flashes dominated varied, but was often during the earlier severe
stages of the storm. The cause of this behavior and its relationship to severe weather production is unclear.
(5) Thunderclouds formed over forest fires or contaminated by smoke. Vonnegut and Orville (1988) [73] found that
25% of about 50 cloud-to-ground flashes apparently associated with the forest fires in Yellowstone National Park
lowered positive charge to Earth. Latham (1991) [74], studying lightning discharges from a cloud generated by a
prescribed forest fire, reported that in a sample of 28 lightning flashes, 19, or two-thirds, were identified as positive
flashes. A relatively high percentage of positive flashes in the central United States detected by the NLDN in Spring
1998 has been associated with cloud contamination by smoke from massive forest fires in Mexico that were up to
thousands of kilometers away (Lyons et al. 1998a [47]; Murray et al. 2000 [75]).
3.2 General characterization
The following is a list of observed lightning properties that are thought to be characteristic of positive lightning
discharges.
(1) Positive flashes are usually composed of a single stroke, whereas about 80% of negative flashes contain two or more
strokes. Multiple-stroke positive flashes do occur but they are relatively rare. Heidler et al. (1998) [76], from electric
field measurements in 1995-1997 in Germany, found that out of a total of 36 positive flashes, 32 contained one stroke
and 4 contained two strokes. On the other hand, Lyons et al. (1998b) [48], using NLDN data for 14 selected summer
months from 1991-1995, reported on 1002 positive flashes (about 0.04% of a total of 2.7 million positive flashes)
composed of more than 10 strokes. However, it is likely that some of these multiple-stroke events are actually
misidentified cloud discharges.
(2) Positive return strokes tend to be followed by continuing currents that typically last for tens to hundreds of
milliseconds (e.g., Fuquay 1982 [53]; Rust et al. 1981, 1985 [77,78]). Brook et al. (1982) [63], from multiple-station
electric field measurements, inferred continuing currents in positive flashes in excess of 10 kA, at least an order of
magnitude larger than for negative flashes, for periods up to 10 ms. Directly measured positive continuing currents in the
kiloamperes to tens of kiloamperes range in winter lightning in Japan are seen following the initial current pulses in Fig.
2. (For comparison, continuing currents in negative flashes are typically in the tens to hundreds of amperes range.)
Such large continuing currents are probably responsible for the unusually large charge transfers by positive flashes.
Brook et al. (1982) [63], for one positive lightning in winter storm in Japan, inferred a charge transfer in excess of 300 C
during the first 4 ms. (For comparison, a typical negative flash transfers to ground a charge of 20 C.) Charge transfers
during the first 2 ms estimated by Berger (1967) [79] for summer positive lightning in Switzerland are of the order of
tens of coulombs. Charge transfers of the order of 1000 C were reported, from direct current measurements, by Miyake
et al. (1992) [64] for both positive and negative winter lightning in Japan. However, these latter events may well be
unusual forms of lightning discharges since the grounded strike object tip was very close to or inside the cloud.
(3) From electric field records, positive return strokes often appear to be preceded by significant in-cloud discharge
activity lasting, on average, in excess of 100 ms (Fuquay 1982) [53] or 200 ms (Rust et al. 1981 [77]). This observation
suggests that a positive discharge to ground can be initiated by a branch of an extensive cloud discharge. Negative
cloud-to-ground discharges are less often preceded by such long-lasting in-cloud discharge activity.
(4) Several researchers (e.g., Fuquay 1982 [53]; Rust 1986 [80]) reported that positive lightning discharges often
involve long horizontal channels, up to tens of kilometers in extent. It is presently not clear why.
(5) It appears that positive leaders can move either continuously or intermittently (in a stepped fashion), as determined
from time-resolved optical images. This is in contrast with negative leaders which are always optically stepped when
they propagate in virgin air. Further, distant (radiation) electric and magnetic field waveforms due to positive
discharges are less likely to exhibit step pulses immediately prior to the return-stroke waveform than are first strokes in
negative lightning. Finally, positive leaders usually do not radiate at VHF and UHF as strongly as negative leaders and
therefore are usually not detected by VHF-UHF lightning imaging systems. In the case of a positive leader, electrons
present or produced ahead of the leader tip move toward the tip since they are attracted to the positive charge on it, and
the resultant ionization occurs in the strong field near the tip. In the case of a negative leader, electrons tend to “run
ahead” of the moving leader tip to where the field is relatively low since they are repelled by the negative charge on the
leader tip. Thus ionization occurs under less favorable conditions for the negative leader than for the positive leader.
As a result, streamer zone formation in the negative leader requires a higher tip potential than for the positive leader,
which may be related to the apparently different intensity of VHF-UHF radiation produced by these two types of
leaders.
The lightning cuirrent parameters for 26 Berger’s positive flashes in Switzerland are summarized in Table 5.
Table 5. Lightning current parameters for positive flashes (Berger et al., 1975 [9])
Sample
Percent Exceeding Tabulated Value
Parameters
Units
Size
95%
50%
5%
Peak current (minimum 2 kA)
kA
26
4.6
35
250
Charge (total charge)
C
26
20
80
350
Impulse charge (excluding continuing
C
25
2.0
16
150
current)
Front duration (2 kA to peak)
s
19
3.5
22
200
Maximum di/dt
kA/ s
21
0.20
2.4
32
Stroke duration (2 kA to half peak value
s
16
25
230
2000
on the tail)
Action integral ( i2dt)
A2s
26
2.5x104
6.5x105
1.5x107
Flash duration
ms
24
14
85
500
3.3 Peak current
A reliable distribution of positive-lightning peak currents applicable to objects of moderate height on the flat ground is
presently unavailable. The sample of 26 directly measured positive-lightning currents analyzed by Berger et al. (1975)
[9] is commonly used as a primary reference both in lightning research and in lightning protection studies. However, this
sample is apparently based on a mix of (1) discharges initiated as a result of junction between a descending positive
leader and an upward connecting negative leader within some tens of meters of the tower top and (2) discharges initiated
as a result of a very long (1 to 2 km) upward negative leader from the tower making contact with an oppositely charged
channel inside the cloud. These two types of positive discharges, which differ by the height above the tower top of the
junction between the upward connecting leader and the oppositely charged overhead channel (descending positive
leader or positively-charged in-cloud channel), are expected to produce very different current waveforms at the tower, as
illustrated in Figs. 5a and b. The “microsecond-scale” current waveform shown in Fig. 5a is probably a result of
processes similar to those in downward negative lightning, whereas the “millisecond-scale” current waveform shown in
Fig. 5b is likely to be a result of the M-component mode of charge transfer to ground (Rakov et al. 2001 [81]). (M
component is a transient process, an increase in current and associated luminosity, that occurs in a lightning channel
carrying continuing current.) It is possible that such millisecond-scale waveforms are characteristic of tall objects,
capable of generating very long upward connecting leaders. On the other hand, the distribution of positive lightning
peak currents inferred from electric or magnetic fields recorded by multiple-station lightning locating systems, such as
the NLDN, are influenced by the uncertainties of the conversion of the measured field to current. The NLDN formula
that is used for this conversion is based on the linear regression equation relating the NLDN-measured field peak to the
directly measured current peak for negative triggered-lightning strokes and is extrapolated to natural positive strokes.
Additionally, the lower end of the positive-lightning peak current distribution is contaminated by misidentified cloudflash pulses (e.g., Cummins et al. 1998 [29]).
Fig. 5. Examples of two types of positive lightning current waveforms observed by K. Berger: (a) “microsecond-scale” waveform
(right panel), similar to those produced by downward negative return strokes, and a sketch (left panel) illustrating the lightning
processes that might have led to the production of this waveform; (b) “millisecond-scale” waveform (right panel) and a sketch (left
panel) illustrating the lightning processes that might have led to the production of this current waveform. Arrows indicate directions
of the extension of lightning channels.
3.4 Return-stroke speed
Mach and Rust (1993) [82], from photoelectric measurements, reported on two-dimensional propagation speeds for 7
positive and 26 negative natural-lightning return strokes. They presented their speed measurements in two groups: one
included values averaged over channel segments less than 500 m and the other included values averaged over channel
segments greater than 500 m. For the "short-segment" group, Mach and Rust (1993) [82] found an average speed of 0.8
x 108 m/s for positive return strokes and 1.7 x 108 m/s for negative return strokes (the typical negative return stroke
speed is 1 x 108 to 1.5 x 108 m/s). Two-dimensional measurements of positive return-stroke speed were also reported by
Idone et al. (1987) [83] for one positive return stroke that was part of an eight-stroke rocket-triggered lightning flash in
Florida (KSC), the other seven strokes being negative, and by Nakano et al. (1987, 1988 [84,85]) for one natural
positive lightning stroke in winter in Japan. Idone et al.'
s (1987) [83] measurements yielded a value about 108 m/s for
8
the positive stroke and values ranging from 0.9 x 10 to 1.6 x 108 m/s for the seven negative strokes, all averaged over a
channel segment of 850 m in length near ground. Nakano et al. (1987, 1988) [84,85] reported a significant decrease in
two-dimensional speed with increasing height over a 180-m section of the channel, from 2 x 108 m/s at 310 m to 0.3 x
108 m/s at 490 m. On the other hand, Mach and Rust (1993) [82] found no significant speed change with height for
positive return strokes. Clearly, more data on positive return-stroke speed are needed.
3.5 Summary
Our knowledge of the physics of positive lightning remains much poorer than that of negative lightning.
Many
questions regarding the genesis of positive lightning and its properties cannot be answered without further research. It is
worth noting that attempts to initiate positive lightning using the rocket-and-wire technique generally result in discharges
that are composed of the initial stage (relatively low-level, long-lasting current component) not followed by positive
leader/return stroke sequences.
Although positive lightning discharges account for 10% or less of global cloud-to-ground lightning activity, there are
five situations that appear to be conducive to the more frequent occurrence of positive lightning. These situations
include (1) the dissipating stage of an individual thunderstorm, (2) winter thunderstorms, (3) trailing stratiform regions
of mesoscale convective systems, (4) some severe storms, and (5) thunderclouds formed over forest fires or
contaminated by smoke. The highest directly measured lightning currents (near 300 kA) and the largest charge transfers
(hundreds of coulombs or more) are thought to be associated with positive lightning. Two types of impulsive positive
current waveforms have been observed. One type is characterized by risetimes of the order of 10 s, comparable to
those for first strokes in negative lightning, and the other type by considerably longer risetimes, up to hundreds of
microseconds. The latter waveforms are apparently associated with very long, 1 to 2 km, upward negative connecting
leaders. The positive return-stroke speed is of the order of 108 m/s. Positive flashes are usually composed of a single
stroke. Positive return strokes often appear to be preceded by significant in-cloud discharge activity, tend to be followed
by continuing currents, and involve long horizontal channels. In contrast to negative leaders which are always optically
stepped when they propagate in virgin air, positive leaders seem to be able to move either continuously or in a stepped
fashion.
4 BIPOLAR LIGHTNING
Lightning flashes that transfer to ground both negative and positive charges are referred to as bipolar lightning flashes.
Most of the bipolar lightning events were identified in direct current measurements on tall grounded objects.
Lightning current waveforms exhibiting polarity reversals were first reported by McEachron (1939, 1941) [86,87] from
his studies at the Empire State Building (New York). According to Hagenguth and Anderson (1952) [88], the number of
bipolar flashes observed at this object over a 10-year period was 11 (14%) from a total of 80 flashes for which polarity
could be determined. The charge transfer was reported to be greater for negative polarity, probably due to the fact that
the initial stage current (e.g., Rakov, 2003 [89]) was mostly negative. Interestingly, no flashes transferring only positive
charge to ground were observed in the Empire State Building studies. Berger (1978) [90] reported that 72 (6%) of 1196
discharges observed in 1963-1973 at Monte San Salvatore (Switzerland) were bipolar, with 68 of them being of the
upward type. For 30 bipolar events, he found median current peaks for the negative and positive parts of the waveform
to be 350 A and 1.5 kA, respectively. The corresponding median charge transfers were 12 and 25 C. Gorin and Shkilev
(1984) [91] reported that 6 (6.7%) of 90 upward discharges initiated from the Ostankino tower (Russia) were bipolar, all
of which initially transported negative charge to ground. The total number of bipolar lightning discharges observed on
the Peissenberg tower (Germany) was two (Heidler et al., 2000 [92]), both of which initially transported negative charge
to ground. Overall, it appears that bipolar flashes constitute 6 to 14% of summer lightning in Europe and USA. Many
bipolar current waveforms have been observed in winter lightning studies in Japan, with the reported frequency of
occurrence ranging from 5 to 33% (see Table 4). At least one bipolar lightning discharge was reported from each of the
triggered-lightning experiments in France, Japan, New Mexico, Florida, and China.
4.1 Classification of bipolar flashes
There are basically three types of bipolar lightning discharges that are illustrated in Fig. 6, although some events may
belong to more than one category.
(1) The first type, illustrated in Fig. 6a, is associated with a polarity reversal during a slowly-varying (millisecond-scale)
current component, such as the initial-stage current in object-initiated lightning or in rocket-triggered lightning. The
initial-stage current in Fig. 6a is followed by two return-stroke current pulses, but bipolar flashes of type 1 may be
composed of the initial stage only. The polarity reversal may occur one or more times and may involve an appreciably
long no-current (or unmeasurable-current) interval between opposite-polarity portions of the waveform. Channel
geometry and current record for a lightning flash that belongs to category 1 are shown in Fig. 7. This flash exhibited a
polarity reversal from negative to positive and contained no leader/return stroke sequences.
Table 6. Observed polarity of currents in winter lightning in Japan.
Number of
Positive
(percent)
Number of
Negative
(percent)
Number of
Bipolar
(percent)
Total
(percent)
1978-1986
41
(33)
78
(62)
6a
(5)
125
(100)
150-m meteorological tower at Maki
1982-1993?
25
(17)
91
(63)
29
(20)
145
(100)
Nagai et al.
(1996) [93]
500-kV GendenTsuruga
transmission line
tower
1986-1992
1
(4)
15
(63)
8
(33)
24
(100)
Miki et al. (2004)
[94]
200-m stack at
Fukui
1989-2002
18
(8)
148
(70)
47 b
(22)
213
(100)
Strike
Object
Observation
Period
Miyake et al.
(1992) [64]
88-m weather
observation tower at
Kashiwazaki and
200-m stack at
Fukui
Goto and Narita
(1995) [52]
Source
a
b
Five changes from negative to positive and one change from positive to negative.
Types 1, 2, and 3a, illustrated in Fig. 4a, b, and c, respectively, were observed.
Fig. 6 - Sketches of overall-flash current records to illustrate different types of bipolar lightning discharges. RS = Return Stroke.
Artwork by J. Jerauld.
Fig. 7 - Bipolar lightning (# 81-11) triggered in winter at the Kahokugata site, Japan. (a) The discharge channels, photographically
observed (solid line) or reconstructed from three- station acoustic measurements (broken line); (b) the current to ground. The current
waveform exhibits a polarity reversal (from negative to positive) during the initial-stage current. This flash contained no downward
leader/upward return stroke sequences. Currents of both polarities followed the same channel to ground, but from different,
oppositely charged regions in the cloud. The negative current was supplied by branch A and the positive current by branch B.
Adapted from Horii (1982, 1986) [95,96] and Akiyama et al. (1985) [97] .
(2) The second type of bipolar discharge, illustrated in Fig. 6b, is characterized by different polarities of the initial-stage
current and of the following return stroke or strokes. Fernandez (1997) [98] reported a positive initial-stage current in
triggered lightning at Camp Blanding, Florida, that was followed by leader/return stroke sequences transferring negative
charge to ground. An example of lightning flash that belongs to category 2 is shown in Fig. 8. This flash, labeled “T1”
(see the right optical channel image in Fig. 8a and top current record in Fig. 8b), exhibited negative initial-stage current
followed by a positive return-stroke current waveform.
Fig. 8 - (a) Concurrent upward bipolar flash (on the right) and upward negative flash (on the left) from towers T1 and T2,
respectively, on Monte San Salvatore (event 6439). (b) The currents measured at tower T1 (top trace) and at tower T2 (bottom trace).
The time scale (which is in seconds) on the bottom trace also applies to the top trace. Current waveform measured at tower T1
exhibits initial-stage current of negative polarity followed, after a no-current interval, by a return stroke of positive polarity. The
bottom current record in Fig. 8b is typical for an upward negative lightning flash composed of the initial stage followed by
leader/return stroke sequences. Adapted from Berger and Vogelsanger (1966) [99] .
(3) The third type of bipolar discharge, illustrated in Figs. 6c and d for natural upward (or rocket-triggered) and natural
downward flashes, respectively, involves return strokes of opposite polarity.
Janischewskyi et al. (1999b) [100]
observed three return strokes in an upward discharge initiated from the CN tower in Toronto, Canada, with currents of 10.6, +6.5, and -8.9 kA. The time interval between the first and second strokes was 300 ms, and between the second and
the third strokes was 335 ms. All three strokes followed the same channel, as observed within about 535 m of the tower
top. The waveshape characteristics of all three strokes were similar. Idone et al. (1987) [83] studied a Florida rockettriggered flash that contained eight return strokes, one of which was positive and the other seven negative. The positive
stroke was the third in the flash, with the preceding and following interstroke intervals being 374 and 369 ms,
respectively. The time intervals between the negative strokes in this flash ranged from 39 to 101 ms. The positive
stroke attached to ground in a separate location from the first two negative strokes, each of which terminated on the
rocket launcher, and hence its current could not be directly measured but was identified by the electric-field polarity.
Fig. 9 - Bipolar lightning flash 03-31 of type 3a triggered at the ICLRT at Camp Blanding, Florida. ICV = Initial Current Variation,
associated with explosion of the triggering wire and its replacement by a plasma channel. Adapted from Jerauld et al. (2004) [101].
The positive stroke peak current was estimated to be about +21 kA, based on a single-station electric field record
obtained 2.2 km away from the channel. The positive stroke was initiated by a positive leader that propagated along the
previously formed channel and then (about 150 m above ground) departed from that channel. Examples of bipolar
flashes of type 3a for the cases of rocket-triggered and object-initiated lightning are shown in Figs. 9 and 10,
respectively.
Shown in Fig. 9 are overall flash current records, displayed on the same time scale but on two different amplitude scales,
for a two-stroke rocket-triggered flash reported by Jerauld et al. (2004) [101]. The initial stage and first stroke of this
flash lowered negative charge to ground, and the second stroke lowered positive charge via the same channel. The
negative return stroke peak current was -11 kA and the positive +5 kA, with the total charge transfers being -1 C and
+24 C, respectively. The two return strokes are separated by about 58 ms. This was the only bipolar flash of type 3a
triggered at the International Center for Lightning Research and Testing (ICLRT) at Camp Blanding, Florida, in 15
years (1993-2007) of its operation, with the total number of triggered flashes being 317.
Fig. 10 - Portion of four-stroke bipolar flash #312 of type 3a initiated from the 100-m Gaisberg tower, Austria. In this flash, the
negative initial-stage current (not shown in Fig. 10) was followed by a sequence of negative and positive return strokes separated by
17.2 ms. Adapted from Schulz and Diendorfer (2003) [102].
As opposed to the positive stroke in the bipolar flash reported by Idone et al. (1987) [83] and discussed above, the
positive stroke shown in Fig. 9 was initiated by a leader that propagated entirely along the previously formed channel, at
least within the bottom several hundred meters imaged above its termination point.
Shown in Fig. 10 is a portion of the overall current record of a four-stroke bipolar flash initiated from the 100-m
Gaisberg tower in Austria and reported by Schulz and Diendorfer (2003) [102]. Only two strokes separated by 17.2 ms
are presented. The first of them is negative and has peak current of -15.7 kA and the second one positive having +8.3
kA peak current. Note that the latter sequence of two strokes is very similar to that shown in Fig. 9, including a
considerably larger charge transferred to ground by the positive stroke.
The two strokes shown in Fig. 10 were detected by the Austrian lightning locating system ALDIS that reported peak
currents of -19.8 kA and +8.4 kA for the negative and positive strokes, respectively. Overall, Schulz and Diendorfer
(2003) [102] reported that they identified about 55,000 (about 3% of) bipolar flashes out of 1.6 million flashes in the
ALDIS data set they analyzed. The total number of bipolar flashes identified by them in currents measured at the
Gaisberg tower in 2000 to 2002 was six, five of which were of type 1 or 2 and one (shown in Fig. 10) of type 3a.
Bipolar flashes of type 3b are illustrated in Fig. 11. Shown in this figure is the overall electric field waveform for a
natural three-stroke bipolar flash that occurred at Camp Blanding, Florida, in 2002 and was reported by Jerauld et al.
(2003) [103]. The first two return strokes, separated by approximately 53 ms, transported positive charge to ground.
These two strokes occurred in separate channels and were followed, 525 ms after the second stroke, by a negative stroke
along the second-stroke channel. The U.S. National Lightning Detection Network (NLDN) located these strokes and
reported peak currents of about +50 kA, +50 kA, and -14 kA. The video, magnetic field, and NLDN records indicate
that the two channel termination points were separated by some hundreds of meters. This event is the first documented
bipolar flash of type 3b, downward flash containing return strokes of opposite polarity.
Fig. 11 - Overall electric field waveform, measured at Station 6 of the multiple-station field measuring system at Camp Blanding,
Florida, for natural downward bipolar flash MSE-0202, on an 800 ms time scale. Two positive strokes are followed by a negative
stroke at about 600 ms. Strokes 2 and 3 apparently followed the same path to ground, different from the first-stroke channel.
Adapted from Jerauld et al. (2003) [103].
4.2 Discussion
For winter lightning in Japan, Narita et al. (1989) [104] suggested that in a bipolar discharge, currents of both polarities
follow the same channel to ground, but from different, oppositely charged regions in the cloud, as illustrated in Fig. 12.
It is likely that the explanation of bipolar current waveshapes suggested by Narita et al. (1989) [104] for winter lightning
also applies to summer bipolar lightning. The hypothetical scenario shown in Fig. 12 implies that the cloud charge
structure cannot always be described by a simple, vertically stacked charge model (e.g., Rakov and Uman, 2003 [105],
Ch. 3). For this scenario to occur, positively and negatively charged regions should be displaced horizontally (should
exist at about the same height in the cloud). The existence of such charge distributions is confirmed by experimental
data of Imyanitov et al. (1971) [106] who reported, from aircraft measurements, pockets of relatively high charge
density that appeared to be chaotically located in the cloud. The numerous pockets of positive and negative cloud charge
observed by them are illustrated in Fig. 13. Note that average net charges at different heights usually formed the
classical positive dipole (the positive charge above the negative charge; see larger plus and minus signs in Fig. 13).
Although Fig. 13 was given by Imyanitov (1971) [106] for cumulus clouds having heights up to 4 km, they stated that it
also applies to cumulonimbus clouds (thunderclouds) if one adds a third, net positive charge layer near the bottom of the
cloud. According to Imyanitov et al. (1971) [106] , the average dimensions of cloud regions (pockets) containing the
largest charges in active thunderstorms are of the order of a few hundred meters. Further, Davis and Standring (1947)
[107] inferred the existence of oppositely charged regions separated by a horizontal distance of 300 m by measuring the
currents in the cables of kite balloons flying at a height of 600 m under thunderstorm conditions. Also, Bateman et al.
(1999) [108] found, from in situ measurements, precipitation particles carrying charge of either polarity at nearly all
altitudes. Thus, it appears that charges of both polarities can coexist in any cloud charge layer, regardless of the polarity
of the net charge of that layer. The “lumpy” cloud charge structure illustrated in Fig. 13 should be conducive to the
occurrence of bipolar flashes.
Fig. 12 - A possible explanation of observed bipolar lightning currents, such as that shown in Fig. 5b. Currents of both polarities
follow, in turn, the same channel to the ground, with negative charge transfer from the cloud charge region labeled “1” being
followed by positive charge transfer from the charge region labeled “2”. Adapted from Narita et al. (1989) [104] .
Fig. 13 - Electrical structure of a convective cloud. Note pockets of positive and negative charge (numerous smaller encircled plus
and minus signs) at different altitudes, while the net charge at the top of the cloud is positive and that at the bottom of the cloud is
negative (larger encircled and minus signs). Adapted from Imyanitov et al. (1971) [106].
4.3 Summary
Bipolar lightning is a poorly understood and often unrecognized phenomenon. However, bipolar events constitute 6 to
14% of summer lightning flashes observed on tall objects in USA, Switzerland, and Russia and 5 to 33% of winter
lightning flashes in Japan. Bipolar flashes were also reported from triggered-lightning experiments in France, Japan,
New Mexico, Florida, and China. Bipolar lightning flashes can be grouped in three categories: (1) type 1 is associated
with a polarity reversal during a millisecond-scale current component (such as the initial-stage current), type 2 is
characterized by different polarities of the initial-stage current and of the following return stroke or strokes, and type 3
involves return strokes of opposite polarity. Bipolar flashes of types 1 and 2 are initiated from tall objects (or triggered
using the rocket-and-wire technique), whereas type 3 can be either upward (type 3a) or downward (type 3b) flash. Miki
et al. (2004) [94] obtained both optical (high-speed video) and current records for types 1, 2, and 3a in their studies of
winter lightning at the 200-m Fukui stack in Japan. All three types were also observed in the Gaisberg-tower studies in
Austria (Schulz and Diendorfer, 2003 [102]). Jerauld et al. (2003) [103], from observations in Florida, presented the
first documented bipolar flash of type 3b (natural downward). The occurrence of bipolar lightning suggests that the
cloud charged structure cannot always be described by a simple, vertically stacked charge model. It is likely that
positively and negatively charged regions can exist at about the same height in the cloud, as observed by Imyanitov et al.
(1971) [106] .
5 SUMMARY OF SALIENT LIGHTNING PROPERTIES
The salient properties of downward negative lightning discharges, the most common type of cloud-to-ground lightning, are
summarized in Table 7.
Characterization of negative cloud-to-ground lightning
Continuing current (longer than
Overall flash
~40 ms)c
M-component
6 ACKNOWLEDGEMENT
This work was supported in part by NSF grant ATM-0346164.
7 REFERENCES
[1] Uman, M.A. 1987. The Lightning Discharge, Orlando: Academic Press, 377 p.
[2] Anderson, R.B., and Eriksson, A.J. 1980. Lightning parameters for engineering application. Electra 69: 65-102.
[3] Kitagawa, N., Brook, M., and Workman, E.J. 1962. Continuing currents in clout-to-ground lightning discharges. J. Geophys.
Res. 67: 13 183-9.
[4] Rakov, V.A. and Uman, M.A. 1990. Some properties of negative cloud-to-ground lightning flashes versus stroke order,
J. Geophys. Res., 95, 5447-5453.
[5] Valine, W.C. and Krider, E.P. 2002. Statistics and characteristics of cloud-to-ground lightning with multiple ground contacts. J.
of Geophysl Res.107, NO. D20, 4441, 11 p.
[6] Saba, M.M.F., Ballarotti, M.G., and Pinto, O. 2006. Negative cloud-to-ground lightning properties from high-speed video
observations. J. Geophys. Res. 111, D03101.
[7] Coray, V., and Perez, H. 1994. Some features of lightning flashes observed in Sweden. J. Geophys. Res. 99: 10,683-8.
[8] Cooray, V., and Jayaratne, K.P.S.C. 1994. Characteristics of lightning flashes observed in Sri Lanka in the tropics. J. Geophys.
Res. 99: 21,051-6.
[9] Berger, K., Anderson, R.B., and Kroninger, H. 1975. Parameters of lightning flashes. Electra 80: 223-37.
[10] Berger, K., and Garabagnati, E. 1984. Lightning current parameters. Results obtained in Switzerland and in Italy. URSI Conf.,
Florence, Italy.
[11] Visacro, S., Soares Jr, A., 2004. Statistical analysis of lightning current parameters: Measurements at Morro do Cachimbo
Station. J. Geophys. Res. 109, D01105, 11 p.
[12] Leteinturier, C., Hamelin, J.H., and Eybert-Berard, A. 1991. Submicrosecond characteristics of lightning return-stroke
currents. IEEE Trans. Electromagn. Compat. 33: 351-7.
[13] Fisher, R.J., Schnetzer, G.H., Thottappillil, R., Rakov, V.A., Uman, M.A., and Goldberg, J.D. 1993. Parameters of triggeredlightning flashes in Florida and Alabama. J. Geophys. Res. 98: 22,887-902.
[14] Rakov, V.A., 1985. On estimating the lightning peak current distribution parameters taking account of the measurement
threshold level (in Russian), Elektrichestvo, No. 2, 57-59.
[15] Le Boulch, M., and Plantier, T. 1990. The Meteorage thunderstorm monitoring system: A tool for new EMC protection
strategies. In Proc. 20th Int. Conf. on Lightning Protection. Interlaken, Switzerland, Paper 6.13P, 8 p.
[16] Lyons, W.A., Uliasz, M., and Nelson, T.E. 1998b. Large peak current cloud-to-ground lightning flashes during the summer
months in the contiguous United States. Mon. Wea. Rev. 126: 2217-23.
[17] Orville, R.E. 1999. Comments on "Large peak current cloud-to-ground lightning flashes during the summer months in the
contiguous United States". Mon. Wea. Rev. 127: 1937-8.
[18] Jerauld, J., Rakov, V.A., Uman, M.A., Rambo, K.J., Jordan, D.M., Cummins, K.L., and Cramer, J.A. 2005. An evaluation of
the performance characteristics of the U.S. National Lightning Detection Network in Florida using triggered lightning, J.
Geophys. Res., 110, D19106, doi:10.1029/2005JD005924.
[19] Lundholm, R. 1957. Induced overvoltage-surges on transmission lines and their bearing on the lightning performance at
medium voltage networks. Trans. Chalmers Univ. Technol. No. 120, 117 p.
[20] Wagner, C.F. 1963. Relation between stroke current and velocity of the return stroke. AIEE Trans. Power Appar. Syst. 82:
609-17.
[21] Hileman, A.R. 1999. Insulation coordination for power systems, 767 p., New York: Marcel Dekker.
[22] Rachidi, F., Nucci, C.A. Ianoz, M., and Mazzetti, C. 1996. Influence of a lossy ground on lightning-induced voltages on
overhead lines. IEEE Trans. Electromagn. Compat. 38: 250-64.
[23] Norinder, H., and Dahle, O. 1945. Measurements by frame aerials of current variations in lightning discharges. Arkiv. Mat.
Astron. Fysik 32A: 1-70.
[24] Uman, M.A., and McLain, D.K. 1970b. Lightning return stroke current from magnetic and radiation field measurements. J.
Geophys. Res. 75: 5143-7.
[25] Uman, M.A., McLain, D.K., Fisher, R.J., and Krider, E.P. 1973a. Electric field intensity of lightning return stroke. J.
Geophys. Res. 78: 3523-9
[26] Uman, M.A., McLain, D.K., Fisher, R.J., and Krider, E.P. 1973b. Currents in Florida lightning return strokes. J. Geophys.
Res. 78: 3530-7.
[27] Dulzon, A.A., and Rakov, V.A. 1980. Estimation of errors in lightning peak current measurements by frame aerials (in
Russian). Izv. VUZov SSSR, ser. Energetika 11: 101-4.
[28] Krider, E.P., Leteinturier, C., Willett, J.C. 1996. Submicrosecond fields radiated during the onset of first return strokes in
cloud-to-ground lightning. J. Geophys. Res. 101: 1589-97.
[29] Cummins, K.L., Krider, E.P., and Malone, M.D. 1998b. The U.S. National Lightning Detection Network and applications of
cloud-to-ground lightning data by electric power utilities. IEEE Trans. on Electromagn. Compat. 40 (II): 465-80.
[30] Rakov, V.A. 2005. Evaluation of the performance characteristics of lightning locating systems using rocket-triggered lightning,
in Proc. of Int. Symp. on Lightning Protection (VIII SIPDA), Sao Paulo, Brazil, Nov. 21-25, 2005, 697-715.
[31] Schonland, B.F.J., Malan, D.J., and Collens, H. 1935. Progressive Lightning II. Proc. Roy. Soc. (London) A152: 595-625.
[32] Boyle, J.S., and Orville, R.E. 1976. Return stroke velocity measurements in multistroke lightning flashes. J. Geophys. Res. 81:
4461-6.
[33] Idone, V.P., and Orville, R.E. 1982. Lightning return stroke velocities in the Thunderstorm Research International Program
(TRIP). J. Geophys. Res. 87: 4903-15.
[34] Mach, D.M., and Rust, W.D. 1989a. Photoelectric return-stroke velocity and peak current estimates in natural and triggered
lightning. J. Geophys. Res. 94: 13,237-47.
[35] Hubert, P., and Mouget, G. 1981. Return stroke velocity measurements in two triggered lightning flashes. J. Geophys. Res. 86:
5253-61.
[36] Idone, V.P., Orville, R.E., Hubert, P., Barret, L., and Eybert-Berard, A. 1984. Correlated observations of three triggered
lightning flashes. J. Geophys. Res. 89: 1385-94.
[37] Willett, J.C., Idone, V.P., Orville, R.E., Leteinturier, C., Eybert-Berard, A., Barret, L., and Krider, E.P. 1988. An
experimental test of the "transmission-line model" of electromagnetic radiation from triggered lightning return strokes. J.
Geophys. Res. 93: 3867-78.
[38] Willett, J.C., Bailey, J.C., Idone, V.P., Eybert-Berard, A., and Barret, L. 1989a. Submicrosecond intercomparison of radiation
fields and currents in triggered lightning return strokes based on the transmission-line model. J. Geophys. Res. 94: 13,275-86.
[39] Rakov, V.A., and Uman, M.A. 1998. Review and evaluation of lightning return stroke models including some aspects of their
application. IEEE Trans. on Electromagn. Compat. 40: 403-26.
[40] Wang, D., Takagi, N., Watanabe, T., Rakov, V.A., and Uman, M.A. 1999c. Observed leader and return-stroke propagation
characteristics in the bottom 400 m of the rocket triggered lightning channel. J. Geophys. Res. 104: 14,369-76.
[41] Olsen, R.C., Jordan, D.M., Rakov, V.A., Uman, M.A., and Grimes, N. 2004. Observed two-dimensional return stroke
propagation speeds in the bottom 170 m of a rocket-triggered lightning channel. Geophys. Res. Lett., 31, L16107, doi:
10.1029/2004GL020187.
[42] Gorin, B.N. 1985. Mathematical modeling of the lightning return stroke. Elektrichestvo, 4,10-16.
[43] Rakov, V.A., and Dulzon, A.A. 1991. A modified transmission line model for lightning return stroke field calculations. In
Proc. 9th Int. Zurich. Symp. on Electromagnetic Compatibility, Zurich, Switzerland, pp. 229-35.
[44] Rakov, V.A., Thottappillil, R., and Uman, M.A. 1992b. On the empirical formula of Willett et al. relating lightning returnstroke peak current and peak electric field. J. Geophys. Res. 97: 11,527-33.
[45] Srivastava, K.M.L. 1966. Return stroke velocity of a lightning discharge. J. Geophys. Res. 71: 1283-6.
[46] Schonland, B.F.J. 1956. The lightning discharge. In Handbuch der Physik 22: 576-628, Berlin, Springer-Verlag.
[47] Lyons, W.A., T.E. Nelson, E.R. Williams, J.A. Cramer, and T.R. Turner, 1998a: Enhanced positive cloud-to-ground lightning
in thunderstorms ingesting smoke from fires. Science, 282: 77-80.
[48] Petersen, W.A., and S.A. Rutledge, 1992: Some characteristics of cloud-to-ground lightning in tropical northern Australia. J.
Geophys. Res., 97: 11,553-11,560.
[49] Seimon, A., 1993: Anomalous cloud-to-ground lightning in an F5-tornado-producing supercell thunderstorm on 28 August
1990. Bull. Am. Meteor. Soc., 74: 189-203.
[50] MacGorman, D.R., and D.W. Burgess, 1994: Positive cloud-to-ground lightning in tornadic storms and hailstorms. Mon. Wea.
Rev., 122: 1671-1697.
[51] MacGorman, D.R., and C.D. Morgenstern, 1998: Some characteristics of cloud-to-ground lightning in mesoscale convective
systems. J. Geophys. Res., 103: 14,011-14,023.
[52] Goto, Y., and Narita, K. 1995. Electrical characteristics of winter lightning. J. Atmos. Terr. Phys. 57: 449-59.
[53] Fuquay, D.M., 1982: Positive cloud-to-ground lightning in summer thunderstorms. J. Geophys. Res., 87: 7131-7140.
[54] Orville, R.E., R.W. Henderson, L.F. Bosart, 1983: An east coast lightning detection network. Bull. Am. Meteor.
Soc., 64: 1029-1037.
[55] Pierce, E.T., 1955b: The development of lightning discharges. Q. J. R. Meteor. Soc., 81: 229-240.
[56] Krehbiel, P.R., 1981: An analysis of the electric field change produced by lightning, Ph.D. Thesis, University of Manchester
Institute of Science and Technology, Manchester, England. [Available as Report T-11, Geophys. Res. Ctr., New Mexico Inst.
Mining and Tech., Socorro, NM 87801, 1981.]
[57] Rust, W.D., and D.R. MacGorman, 2002: Possibly inverted-polarity electrical structures in thunderstorms during STEPS
[58] Orville, R.E., R.A. Weisman, R.B. Pyle, R.W. Henderson, and R.E. Orville, Jr., 1987: Cloud-to-ground lightning flash
characteristics from June 1984 through May 1985. J. Geophys. Res., 92: 5640-5644.
[59] Orville, R.E., and A.C. Silver, 1997: Lightning ground flash density in the contiguous United States: 1992-95. Mon.
Wea. Rev., 125: 631-638.
[60] Orville, R.E., and G.R. Huffines, 1999: Lightning ground flash measurements over the contiguous United States: 1995-1997.
Mon. Wea. Rev., 127: 2693-2703.
[61] Hojo, J., M. Ishii, T. Kawamura, F. Suzuki, H. Komuro, and M. Shiogama, 1989: Seasonal variation of cloud-to-ground
lightning flash characteristics in the coastal area of the Sea of Japan. J. Geophys. Res., 94: 13,207-13,212.
[62] Takeuti, T., M. Nakano, M. Brook, D.J. Raymond, and P. Krehbiel, 1978: The anomalous winter thunderstorms of the
Hokuriku coast. J. Geophys. Res., 83: 2385-2394.
[63] Brook, M., M. Nakano, P. Krehbiel, and T. Takeuti, 1982: The electrical structure of the Hokuriku winter thunderstorms. J.
Geophys. Res., 87: 1207-1215.
[64] Miyake, K., T. Suzuki, and K. Shinjou, 1992: Characteristics of winter lightning current on Japan Sea coast. IEEE Trans.
Pow. Del., 7: 1450-1456.
[65] Kitagawa, N., and K. Michimoto, 1994. Meteorological and electrical aspects of winter thunderclouds. J. Geophys. Res., 99:
10,713-10,721.
[66] Engholm, C.D., E.R. Williams, and R.M. Dole, 1990: Meteorological and electrical conditions associated with positive cloudto-ground lightning. Mon. Wea. Rev., 118: 470-487.
[67] Orville, R.E., R.W. Henderson, L.F. Bosart, 1988: Bipole patterns revealed by lightning locations in mesoscale
storm systems, Geophys. Res. Lett. 15: 129-32.
[68] Rust, W.D., D.R. MacGorman, and R.T. Arnold, 1981: Positive cloud to ground lightning flashes in severe storms. Geophys.
Res. Lett., 8: 791-794.
[69] Stolzenburg, M., 1994: Observations of high ground flash densities of positive lightning in summertime thunderstorms. Mon.
Wea. Rev., 122: 1740-1750.
[70] Perez, A.H., L.J. Wicker, and R.E. Orville, 1997: Characteristics of cloud-to-ground lightning associated with violent
tornadoes. Wea. Forecast., 12: 425-437.
[71] Carey, L.D., and S.A. Rutledge, 1998: Electrical and multiparameter radar observations of a severe hailstorm. J. Geophys.
Res., 103: 13,979–14,000.
[72] Williams, E.R. 2001: The electrification of severe storms. In Severe Convective Storms, Meteor. Monogr., ed. C.A. Doswell,
vol. 28, no. 50, Am. Meteor. Soc., pp. 527-561.
[73] Vonnegut, B., and R.E. Orville, 1988: Evidence of lightning associated with the Yellowstone Park forest fire (Abstract). Eos
Trans. AGU, 69: 1071.
[74] Latham, D., 1991. Lightning flashes from a prescribed fire-induced cloud. J. Geophys. Res., 96: 17,151-17,157.
[75] Murray, N.D., R.E. Orville, and G.R. Huffines, 2000: Effect of pollution from Central American fires on cloud-to-ground
lightning in May 1998. Geophys. Res. Lett., 27: 2249-2252.
[76] Heidler, F., F. Drumm, and Ch. Hopf, 1998: Electric fields of positive earth flashes in near thunderstorms. Proc. 24th Int.
Conf. on Lightning Protection, Birmingham, United Kingdom, pp. 42-47.
[77] Rust, W.D., D.R. MacGorman, and R.T. Arnold, 1981: Positive cloud to ground lightning flashes in severe storms. Geophys.
Res. Lett., 8: 791-794.
[78] Rust, W.D., D.R. MacGorman, and W.L. Taylor, 1985: Photographic verification of continuing current in positive cloud-toground flashes. J. Geophys. Res., 90: 6144-6146.
[79] Berger, K., 1967: Novel observations on lightning discharges: Results of research on Mount San Salvatore. J. Franklin Ins.,
283: 478-525.
[80] Rust, W.D., 1986: Positive cloud-to-ground lightning. The Earth'
s Electrical Environment, E.P. Krider and R.G. Roble, Eds.,
National Academy Press, 41-45.
[81] Rakov, V.A., D.E. Crawford, K.J. Rambo, G.H. Schnetzer, M.A. Uman, and R. Thottappillil, 2001: M-Component mode of
charge transfer to ground in lightning discharges. J. Geophys. Res., 106, 22,817-22,831
[82] Mach, D.M, and Rust, W.D. 1993. Two-dimensional velocity, optical risetime, and peak current estimates for natural positive
lightning return strokes. J. Geophys. Res. 98: 2635-8.
[83] Idone, V.P., Orville, R.E., Mach, D.M., and Rust, W.D. 1987. The propagation speed of a positive lightning return stroke.
Geophys. Res. Lett. 14: 1150-3.
[84] Nakano, M., Nagatani, M., Nakada, H., Takeuti, T., and Kawasaki, Z. 1987. Measurements of the velocity change of a
lightning return stroke with height. Res. Lett. Atmos. Electr. 7: 25-8.
[85] Nakano, M., Nagatani, M., Nakada, H., Takeuti, T., and Kawasaki, Z. 1988. Measurements of the velocity change of a
lightning return stroke with height. In Proc. 1988 Int. Aerospace and Ground Conf. on Lightning and Static Electricity,
Oklahoma City, Oklahoma, pp. 84-86.
[86] McEachron, K.B. 1939. Lightning to the Empire State Building. J. Franklin Inst. 227: 149-217.
[87] McEachron, K.B. 1941. Lightning to the Empire State Building. AIEE Trans. 60: 885-90.
[88] Hagenguth, J.H., and Anderson, J.G. 1952. Lightning to the Empire State Building - Part III. AIEE Trans. 71: 641-9.
[89] Rakov, V.A. 2003. A review of the interaction of lightning with tall objects. Res. Res. Devel. Geophysics, 5: 57-71.
[90] Berger, K. 1978. Blitzstrom-Parameter von Aufwärtsblitzen. Bull. Schweiz. Elektrotech. Ver. 69: 353-60.
[91] Gorin, B.N., and Shkilev, A.V. 1984. Measurements of lightning currents at the Ostankino tower. Elektrichestvo 8: 64-5.
[92] Heidler, F., Zischank, W., and Wiesinger, J. 2000. Statistics of lightning current parameters and related nearby magnetic
fields measured at the Peissenberg tower. In Proc. 25th Int. Conf. on Lightning Protection, Rhodes, Greece, pp. 78-83.
[93] Nagai, T., Matsui, T., Sonoi, Y., Adachi, M., Sekioka, S., and Sugimoto, O. 1996. Observation on winter lightning at a tall
transmission line tower in Hokuriku district. In Proc. 10th Int. Conf. on Atmospheric Electricity, Osaka, Japan, pp. 476-479.
[94] Miki, M., Wada, A., and Asakawa, A. 2004. Observation of upward lightning in winter at the Coast of Japan Sea with a highspeed video camera. In Proc. 27th Int. Conf. on Lightning Protection, Avignon, France, pp. 63-67.
[95] Horii, K. 1982. Experiment of artificial lightning triggered with rocket. Memoirs of the Faculty of Engineering, Nagoya
University, Nagoya, Japan, 34: 77-112.
[96] Horii, K. 1986. Experiment of triggered lightning discharge by rocket. In Proc. Korea-Japan Joint Symp. on Electrical Material
and Discharge, Cheju Island, Korea, paper 8-3, 6 p.
[97] Akiyama, H., Ichino, K., and Horii, K. 1985. Channel reconstruction of triggered lightning flashes with bipolar currents from
thunder measurements. J. Geophys. Res. 90: 10,674-80.
[98] Fernandez, M.I. 1997. Responses of an unenergized test power distribution system to direct and nearby lightning strikes. M.S.
Thesis, Univ. Florida, Gainesville, 249 p.
[99] Berger, K., and E. Vogelsanger, 1966: Photographische Blitzuntersuchungen der Jahre 1955-1965 auf dem Monte San
Salvatore. Bull. Schweiz, Elektrotech. Ver., 57: 599-620.
[100] Janischewskyi, W., Shostak, V., Hussein, A.M., and Kordi, B. 1999b. A lightning flash containing strokes of opposite
polarities. CIGRE SC 33.99 (COLL), 2 p.
[101] Jerauld, J., M.A. Uman, V.A. Rakov, K.J. Rambo, and D.M. Jordan. 2004. A triggered lightning flash containing both
negative and positive strokes. Geophys. Res. Lett., 31, L08104, doi: 10.1029/2004GL019457.
[102] Schulz, W. and G. Diendorfer. 2003. Bipolar flashes detected with lightning location systems and measured on an
instrumental tower. In Proc. VII Int. Symp. on Lightning Protection, Curitiba, Brazil, pp. 6-9.
[103] Jerauld, J., Rakov, V.A., Uman, M.A., Crawford, D.E., DeCarlo, B.A., Jordan, D.M. Rambo, K.J., and Schnetzer, G.H. 2003.
Multiple-station measurements of electric and magnetic fields due to natural lightning. In Proc. Int. Conf. on Lightning and
Static Electricity, Blackpool, UK.
[104] Narita, K., Goto, Y., Komuro, H., and Sawada, S. 1989. Bipolar lightning in winter at Maki, Japan. J. Geophys. Res. 94:
13,191-5.
[105] Rakov, V.A., and M.A. Uman Lightning: Physics and Effects, 687 p., 2003, Cambridge University Press.
[106] Imyanitov, I.M., Chubarina, Y.V., and Shvarts, Y.M. 1971. Electricity of Clouds, 92 p., Leningrad: Gidrometeoizdat (NASA
Technical Translation from Russian, NASA TT-F-718, 1972).
[107] Davis, R., and Standring, W.G. 1947. Discharge currents associated with kite balloons. Proc. Roy. Soc. (London) A191: 304322.
[108] Bateman, M.G., Marshall, T.C., Stolzenburg, M., and Rust, W.D. 1999. Precipitation charge and size measurements inside a
New Mexico mountain thunderstorm. J. Geophys. Res. 104: 9643-9653.