Electromagnetic pulses related to seismic waves can trigger

lightning and TLEs

Leonid V. SorokinEconomic & Mathematical modeling Department,

Peoples’ Friendship University of Russia, Moscow, Russia

AK135• To discover the effect a big amount of data is

needed. The catalogue of earthquakes for 30-year period has been computed to find out the cases of earthquake triggering by seismic waves from other earthquakes at the big angular distances.

• All evaluations for definition of possible phases of seismic waves and evaluations of their travel times were conducted with the use of model AK135.

• Seismic waves: P, Pdiff, PKP, PKiKP, pP, pPdiff, pPKP, pPKiKP, sP, sPdiff, sPKP, sPKiKP, PP, P'P', S, Sdiff, SKS, pS, pSdiff, pSKS, sS, sSdiff, sSKS, SS, S'S', PS, PKS, SP, SKP, SKiKP, PcP, PcS, ScP, ScS, PKKP, PKKS, SKKP and SKKS.

Computation of seismic waves travel times

• Buland. R and C.H. Chapman, The computation of seismic travel times, Bull. Seism. Sot. Am, v. 73, pp. 1271-1302, 1983.

• Kennett B.L.N. and E.R. Engdahl, Traveltimes for global earthquake location and phase identification, Geophys. J. Int., 105, 429-465, (1991).

• Kennett B.L.N., IASPEI 1991 Seismological Tables, Research School of Earth Sciences, Canberra, (1991).

Method

• At the next stage of this analysis the space-time coupling of earthquakes and positive lightning at big angular distances for the big number of events was found.

• After that the cases of electromagnetic pulses generation at the big angular distances by seismic waves from the earthquakes have been described.

• The detected signals have sub-millisecond duration and high intensity.

Electromagnetic field observation

• The previous version of the 50 kHz broadband 3D receiver system was based on 16-bit ADC, 100000 samplers per second in each channel.

• It has the restriction of 0.352 A/m at 10 kHz due to the sensitivity of data-server and 100-BaseTX networking components for the sub-millisecond electromagnetic pulses.

The new 100 kHz broadband 3D receiver consists of:

• magnetic loop antennas in north-south (NS), east-west (EW) and vertical (Z) directions;

• radio time signal receiver; • personal computer; • 24-bit multitrack (4 channels in use) hard

disk recorder dynamic range is 120 dB and its data sampling rate is 192000 records per second for each channel (one hour record in each channel consumes the 2.4 GB of HDD disk space).

The new 100 kHz broadband 3D receiver system covers:

• super low frequency (SLF, 30 Hz to 300 Hz);

• ultra low frequency (ULF, 300 Hz to 3 kHz);

• very low frequency (VLF, 3 kHz to 30 kHz);

• and a part of low frequency band (LF, 30 kHz to 100 kHz).

The 100 kHz bandwidth receiver system measures:

• three components of magnetic field (BNS, BEW, BZ);• provides U.T. time synchronization (with the <1 msec

timing accuracy); • calculates 3D elements: • arrival Azimuth • output wave angle (Zenit=0°);• absolute magnetic field waveform

(1)

This facility is very important for determining the exit point of the electromagnetic waves and selection of the atmospheric or underground nature of the source.

)( 222ZEWNS BBBB

January, 14 - February, 26, 2003 series

• In the period from January, 14 till February, 26, 2003 a series of 3D electromagnetic field observations (total duration of 516 hours) has been carried out in the Moscow region.

• The previous 3D recording system (16-bit ADC, 50 kHz bandwidth) registered 1088 sub-millisecond electromagnetic pulses with absolute magnetic field intensity more than 0.01 A/m.

• The top threshold of registration was 0.352 A/m (10 kHz) due to the electromagnetic stability of the measuring complex. A number of events with higher magnetic field intensity were missed due to the system failure.

• At the same 516 hours period 932 earthquakes have been registered by USGS NEIC (catalogue PDE). Comparison of the earthquake data with time and 3D elements from electromagnetic field observations showed the existence of 867 sub-millisecond electromagnetic pulses at the moments when seismic waves from earthquakes were passing through a place of registration.

• All 867 sub-millisecond electromagnetic pulses associated with earthquakes at the period of observation (516 hours) are presented at figures 1 and 2.

Fig. 1. Intensity distribution of sub-millisecond electromagnetic pulses: vertical axis – absolute magnetic field intensity (A/m) and

horizontal axis – number of the sorted event corresponding to this intensity

• Sub-millisecond electromagnetic pulses absolute magnetic field intensity distribution (fig. 1) is very similar to the earthquake magnitude distribution. Earthquakes with high levels of magnitude are very rare, but the lower magnitude value the higher number of the earthquakes.

• The same 867 events on the distribution of total number sub-millisecond electromagnetic pulses associated with the exact angular distance to the epicentres of the earthquakes (fig. 2) visualize the distribution of seismic activity around the globe from the place of electromagnetic field observation. Each peak at the figure 2 represents a group of earthquakes around one earthquake centre at the period of observation (516 hours). Seismic waves from far away earthquakes passing from the place of 3D electromagnetic field observation due to the seismo-electromagnetic emission were registered.

Fig. 2. Distribution of total number of sub-millisecond electromagnetic pulses associated with the exact angular distance to the earthquake epicentres: N –

total number of the events; AD – Angular Distance to the epicentres of the earthquakes from the point of observation

Triggering of positive polarity lightning• We can move in the coordinate system concerning the seismic

wave. Traveling around the globe “sitting” on seismic wave one can see the destruction caused by wave front propagation through the medium. The seismic wave is the wave package with its entry, body and tail.

• All evaluations for definition of possible phases of seismic waves and evaluations of their travel times were conducted with the use of model AK135 based on IASPEI-91.

• The AK135 model calculates the travel times (entry) for possible phases of seismic waves but the maximum amplitude of the wave package can be seen only on the seismogram which alone can give the information about the exact position of the maximum. In reality the maximum can be reached at any moment beginning from the entry and furthermore during the interval of several minutes.

• So the travel times from AK135 can point the entry with accuracy ±10 seconds, but not the maximum of the wave package amplitude. In this paper due to this fact the confidence interval for possible phases of seismic waves selection is set to ±100 seconds.

Malaysian LDN• Malaysia is situated in the very active region of thunderstorm

activity and very close to the most seismic active regions of Australia, Indonesia, China and Japan. So at comparatively small time period a big amount of data were obtained.

• At the period from May, 1 till August, 31, 1998 (total duration of 4 months) data from Malaysian lightning detection network (LDN, Malaysia, Universiti Tenaga Nasional) were computed together with earthquake data from USGS NEIC (catalogue PDE).

• In this period Malaysian LDN had registered 465866 lightning discharges, among them 454671 negative polarity lightning (–CG) and 11195 positive polarity lightning (+CG).

• The –CG lightning were 40.61 times more often then +CG lightning discharges. At the same 4 months period 7601 earthquakes had been registered by USGS NEIC (catalogue PDE).

• Comparison of the earthquake data with LDN data showed the existence of 9233 coupling events at the moments when seismic waves from earthquakes were passing through a place of lightning.

Fig. 1. Dependence of total number +CG lightning discharges on the earthquake magnitude and peak current: vertical axis – total number of +CG events; Peak current

axis – peak current of +CG lightning (kA); Earthquake Magnitude axis – magnitude (NEIC)

• In the center of the figure one can see the compact area of triggered +CG lightning. Starting from magnitude close to 1.5 and lasting to the maximum registered level it has a number of peaks in the interval from 3 to 5. The absolute maximum was reached for earthquake magnitude close to 5. The maximum number of +CG lightnings corresponds to the peak currents from 40 to 60 kA.

• Not all +CG lightnings have the space-time coupling with exact seismic waves from the earthquakes.

• Stochastic component of peak current corresponds to “zero level” Magnitude. These events have nothing to do with earthquake triggers and have enough energy to maintain self lightning activity.

• Due to this fact we can select the most confidential events with earthquake magnitude greater then 2.

Fig. 2. Distribution of +CG lightning total number associated with the exact angular distance to the earthquake epicentres: N – total number of +CG events; AD – Angular

Distance to the earthquake epicentres from the point of +CG lightning

• The same 9233 events on the distribution of +CG lightning total number associated with the exact angular distance to the earthquake epicentres visualize the distribution of seismic activity around the globe from the place of +CG lightning.

• One can see that faraway earthquakes situated around the globe can trigger the +CG lightning discharges. The dark area corresponds to the 8034 +CG lightning associated with earthquake magnitude greater then 2. The grey area corresponds to the 1199 events associated with earthquake magnitude lesser then 2.

• The figure 1 contains all 9233 +CG lightning. This number is equal to 82.47% of the total 11195 +CG lightnings registered. And only 71.76% (8034) of them had been associated with exact seismic waves from the earthquakes (M>2).

• The selected 8034 +CG lightning discharges (Malaysia LDN) associated with earthquakes (M>2) in the 4 months period are presented at the next figures.

Fig. 3. The total number of +CG lightning presented in the coordinates: the exact angular distance to the earthquake epicentres and seismic wave travel time: vertical axis – total number of +CG events; time axis – seismic wave travel time; horizontal axis – Angular

Distance to the +CG lightning from the point of earthquake epicentres

• The selected data (8034 +CG lightning, M>2) are presented on figure 3 with additional time axis. This projection is typical for hodograph. Now the information on +CG lightning relation to the exact seismic waves from the earthquakes corresponding to its angular distance are quiet clear. One can see that seismic waves scattering around the globe can trigger the +CG lightning even 57 minutes after the earthquakes. The seismic waves propagating through the earth’s mantle, core, and reflected from the backside of crust can trigger the +CG lightning with high efficiency.

Peak current isolines• It is very interesting to have a look on peak current

isolines. The upper isoline corresponds to all selected events (8034 +CG lightning, M>2). In the middle of the distribution one can see the moment (0 sec) of seismic waves passage through the place of +CG lightning.

• Triggered events have higher peak currents due to simultaneous involving in the lightning discharges of bigger cloud volume. The events with peak currents 63.1 and 100.0 kA are highly localized in the center of distribution (fig. 4). So the +CG lightning discharges with higher peak currents are more probable during the moments when seismic waves from earthquakes pass through a place of +CG lightning.

Peak current isolines corresponding to the +CG lightning: vertical axis – total number of +CG events; horizontal axis – time difference between seismic wave travel time and

passage of seismic waves through the place of +CG lightning

Triggering of transient luminous events

• Cloud-to-ground (+CG) lightning discharges, with charge moment changes in excess of ~600 C·km can generate sprites (W.A. Lyons, 2002).

• The triggered +CG lightning discharges can have the higher peak currents. So the higher probability for TLEs exists during the moments when seismic waves from earthquakes passing through a place +CG of lightning discharges.

• TLEs video observations (Taylor et al. 1996; Mitchell et al. 1997; Vaughn et al. 1997) have been studied together with the NLDN (USA) data.

• On the base of coordinated observations of TLEs space-time coupling between exact earthquake seismic waves and eight high-altitude electric discharges in atmosphere (6 Red Sprites, Vertical Pulse, Elve) were found on the big angular distances. One of these events is described in this paper as an example.

Fig. 5. Multiple Persistent Sprites. Image from ISOCON camera with 665 nm filter (Courtesy Mike Taylor, USU)

Taylor M.J., Clark S., High resolution CCD and video imaging of Sprites and

Elves in the N2 first-positive band emission, EOS Trans., AGU 77

(Supplement), F60, 1996.

• The earthquake with magnitude mb=4.4 (NEIC) occurred on July 19, 1996 at 04:23:13 U.T. (geographic coordinates: Lat=–0.80, Lon=136.71), and its depth was 33°km. The multiple persistent sprites (Taylor et al. 1996; Mitchell et al. 1997; Vaughn et al. 1997) were associated with +CG lightning (NLDN, USA) with peak current +38 kA occurred on July 19, 1996 at 04:49:26 U.T. (geographic coordinates: Lat=40.080, Lon=–101.799). It is important to note, that the sprites video record (July 19, 1996 at 04:49:27 U.T. fig. 5) was done 1574 sec after the earthquake on the angular distance of 114.121 degrees from the earthquake epicentre.

• These multiple persistent sprites have been associated with passage of seismic waves from earthquake (mb=4.4 (NEIC), July 19, 1996, 04:23:13 U.T., Lat=–0.80, Lon=136.71) through the place of +CG lightning (NLDN). The pSKSdf seismic wave (travel time 1560.00 sec) passed the place of +CG lightning 13.00 seconds before and sSKSdf seismic wave (travel time 1563.72 sec) passed this place 9.28 seconds before. The multiple persistent sprites took place within 1 second after +CG lightning. At the moment of this event there was an interference of pSKSdf and sSKSdf seismic waves in the place of multiple persistent sprites. The accuracy of 9.28 seconds before the +CG lightning corresponds with the AK135 model. So the space-time coupling between +CG lightning and multiple persistent sprites with exact seismic waves from the earthquake is established.

Acknowledgments • Walter A. Lyons, CCM (FMA Research, Inc.

USA) and Michael Taylor (Utah State University, USA) for the sprites video records, R. Buland (National Earthquake Information Centre, U.S. Geological Survey Golden, Colorado) and Brian Kennett (Research School of Earth Sciences, Australian National University Canberra, Australia) for possibility to use the software IASPEI91 and AK135. Special thanks to the experts on atmospheric electricity Halil Hussin and Mohd Pauzi Bin Yahaya (Malaysia, Universiti Tenaga Nasional).

• The records from new 100 kHz bandwidth 3D receiver system can serve as an example of waveform and spectrum characteristics of sub-millisecond electromagnetic pulses.

• The Greece earthquake with magnitude Mw=5.8 (NEIC) occurred on March 25, 2007 at 13:57:59.89 U.T. (geographic coordinates: Lat=38.420, Lon=20.494), and its depth was 10°km. At the figure 3 we can see one-hour record from the beginning of strong earthquake in Greece (vertical Z-channel) registered by OBN seismological station (Lat=55.114, Lon=36.569). During this event a number of sub-millisecond electromagnetic pulses “geospherics” were registered by the new 100 kHz bandwidth 3D receiver system situated in Moscow (Lat=55.640, Lon=37.517).

On the seismogram one can see the arrival of very intense seismic P-wave on the 271.3 second from the earthquake.

In Moscow at the place of electromagnetic field

observation P-wave arrived 11 seconds later

BHE00

BHN00

BHZ00

One-hour record of OBN station in Z-channel (Greece, Mw=5.8 (NEIC), March 25, 2007 at 13:57:59.89 U.T.)

• The broadband 3D electromagnetic field observations during the winter period at the complete absence of lightning activity at the distances up to 1000 km showed the existence of natural abnormal strong signals.

• These natural sub-millisecond electromagnetic pulses have been associated with passage of seismic waves through the place of electromagnetic field observations.

At the moment of Greece earthquake the new 100 kHz bandwidth 3D receiver system situated in Moscow has made the direct registration of this seismic event. At the

next figure we can see the 10 ms record of BZ-component of electromagnetic field with direct

registration of Greece earthquake. This record has been done at the angular distance of 20.642 degrees from

the earthquake epicentre. Azimuth 28° to the earthquake epicentre corresponds to the direction of

electromagnetic wave arrival.

Direct registration of sub-millisecond electromagnetic pulse from the earthquake epicentre (Greece, Mw=5.8 (NEIC), March 25, 2007 at 13:57:59.89 U.T.) on the angular

distance 20,642 degrees

Registration of sub-millisecond electromagnetic pulse associated with P-wave from the earthquake

(Greece)

This figure shows the 10 ms record of BZ-component of electromagnetic field with arrival of P-wave from Greece

earthquake. The duration of P-wave electromagnetic signal is 0.3 msec. It is important to note, that the record

of seismo-electromagnetic emission was done on the angular distance of 20.642 degrees from the earthquake epicentre and the signal had sub-millisecond duration.

The spectrum shows that all energy of the event is localized within the 38 kHz band. The electromagnetic

wave propagated from underground source situated under the place of observation (Moscow). The deeper the source the lower frequencies can be observed in the spectrum on

the surface of the ground.

Infrared Cloud Image from Sat24.com (23:00 UTC 3 September or 01:00 CET

4 September 2008)

Radar Data from Meteox.com (23:00 UTC 3 September or 01:00 CET 4 September 2008)

Infrared Cloud Image from Sat24.com

(23:00 UTC 3 September or 01:00 CET 4 September 2008)

1st TEA – IS Summer School, June 17th – June 22nd 2012, Málaga, Spain

EuroSprite-2008 Campaign

• We can see the synchronization of lightning activity within one second (21:15:49 UTC 3 September 2008). The lightning packet situated in the compact area and consists of 7 positive polarity lightning and one negative polarity lightning.

• It is important that in the time interval of 5 sec. before and after it there was no lightning at all. At the same time the Corsican System registered the big multiple Sprite (EuroSprite-2008 Campaign).

• It is a good example of the space-time coupling between exact seismic waves from the earthquake and lightning.

The Positive polarity Lightning and Transient Luminous Event (Big Multiple Sprite) Associated with Seismic Waves

from Earthquake (21:15:49 3 September 2008).

Angul. Azimuth from Time (sec) Seismic Earthquake Positive Lightning TLE

Dist. EQ L EQ-L TTime Diff. Wave Time M Time Current  

140.38 333.41 38.31 2135 2147.64 -12.64 SKKSdf 20:40:14.07 4.4 21:15:49.04 151700 Sprite

140.51 333.24 38.44 2135 2147.39 -12.39 SKKSdf 20:40:14.07 4.4 21:15:49.12 27000  

140.17 333.42 38.43 2135 2148.04 -13.04 SKKSdf 20:40:14.07 4.4 21:15:49.12 68500  

139.93 333.51 38.45 2135 2148.49 -13.49 SKKSdf 20:40:14.07 4.4 21:15:49.37 59800  

140.52 333.48 38.13 2135 2147.36 -12.36 SKKSdf 20:40:14.07 4.4 21:15:49.42 17500  

139.33 333.74 38.54 2135 2149.63 -14.63 SKKSdf 20:40:14.07 4.4 21:15:49.50 93000  

140.51 329.40 45.98 2135 2123.31 11.69 SKKSac 20:40:14.07 4.4 21:15:49.54 16100  

140.09 333.08 38.91 2135 2148.20 -13.20 SKKSdf 20:40:14.07 4.4 21:15:49.67 -5900  

EuroSprite-2008 Campaign, Corsican System, multiple Sprite

(21:15:49.072 UTC 3 September 2008)

The Positive polarity Lightning and Transient Luminous Events Associated with Seismic Waves from Earthquakes (from

22:40:00 3 September up to 00:10:00 4 September 2008).

Ang. Azimuth from Time (sec) Seismic Earthquake Positive Lightning TLE

Dist. EQ L EQ-L Ttime Diff. Wave Time M Time Current  

11.67 293.06 102.82 176 173.76 2.24 PnPn 23:04:03.80 3.5 23:06:59.12 19600 Sprite

11.48 294.98 104.97 176 171.14 4.86 PnPn 23:04:03.80 3.5 23:06:59.27 39800 Sprite

11.29 294.27 104.42 719 727.55 -8.55 ScP 23:04:03.80 3.5 23:16:02.26 34600 Sprite

34.30 302.67 96.09 2403 2406.82 -3.82 P'P'df 22:43:13.00 5.1 23:23:16.68 36400 Sprite

34.33 302.53 95.83 722 732.82 -10.82 S 23:14:45.50 3.4 23:26:47.85 40600 Sprite

90.23 320.51 54.70 1432 1431.58 0.42 ScS 23:11:45.20 3.7 23:35:37.30 167300 Sprite

50.90 300.43 71.24 582 569.96 12.04 sP 23:35:12.54 3.4 23:44:54.46 73900 Sprite

51.32 300.02 70.70 609 605.23 3.77 PcP 23:35:12.54 3.4 23:45:21.30 -23200 Elve

11.46 293.05 103.01 229 219.66 9.34 PgPg 23:47:01.10 3.6 23:50:50.12 15100 Sprite

50.52 300.58 71.66 1208 1207.69 0.31 SKiKP 23:35:12.54 3.4 23:55:20.76 38600 Sprite

11.46 294.13 104.12 956 945.3 10.70 ScS 23:47:01.10 3.6 00:02:57.94 27200 Sprite

71.98 24.49 343.01 2355 2354.77 0.23 P'P'df 23:28:56.24 3.0 00:08:11.94 144900 Sprite

Conclusion

• We can see that seismic waves scattering around the globe, propagating through the earth’s mantle, core and reflected from the backside of crust can trigger with high efficiency Positive polarity Lightning and Transient Luminous Events associated with lightning.

• To demonstrate this effect we can have a look on the fragment of data obtained during the EuroSprite-2008 Campaign and the earthquake catalog data (NEIC PDE-W). Figure contains the data from 22:40:00 3 September up to 00:10:00 4 September 2008 and represents 11 Sprites and one Elve.

Space-time coupling between Transient Luminous Events and exact seismic waves from the earthquakes

(22:40:00 3 September - 00:10:00 4 September 2008)

EuroSprite-2008 Campaign, Corsican System, (2-3 September 2008)

The group of two Elves and 84 Sprites associated with different lightnings took place in the seismological active region (Lion).

The lightning activity together with TLE’s registration can be one of the new earthquake predictors

The group of two Elves and 84 Sprites associated with different lightnings took place in the seismological active region (Lion). The central point of TLE’s group was in the coordinates: Lat(mean)=45.17, Lon(mean)=5.97. TLE’s started one day after the closest earthquake (Lat=45.982, Lon=6.165, Depth=2 km, Mw=2.5 (NEIC), 30 August 2008, 2:33:06.6 U.T.).

The seismic activity in this place lasted for four months after that up to the end of 2008. The next earthquake in this place happened with higher Magnitude 4.2 on February 26, 2012 (Lat=44.538, Lon=6.658, Depth=2 km, Mw=4.2 (NEIC), at 22:37:55.79 U.T.).

Acknowledgments

• First of all I would like to thank Torsten Neubert (DTU Space and National Space Institute) for the opportunity to operate the Pic du Midi and Corsican Systems during the EuroSprite-2008 Campaign.

• My deep gratitude to Olivier Chanrion for supporting the Instruments during his vacations and to Oscar Van Der Veld for the Weather information and Storm forecasting.

• My special thanks to Earle Williams (Massachusetts Institute of Technology, Cambrigde) and Walter A. Lyons, CCM (FMA Research, Inc. USA) for very fruitful discussion on the subject.

• I also thank R. Buland (National Earthquake Information Centre, U.S. Geological Survey Golden, Colorado) and Brian Kennett (Research School of Earth Sciences, Australian National University Canberra, Australia) for the software IASPEI91 and AK135.

References• L.V. Sorokin. Triggering of Positive Lightning and High-Altitude Atmospheric

Discharges. // Proceedings of the NATO Advanced Study Institute on “Sprites, Elves and Intense Lightning Discharges” Edited by Martin Fullenkrug, Eugene A. Mareev and Michael J. Rycroft. Corte, Corsica, France 24-31 July 2004, NATO Science Series II: Mathematics, Physics and Chemistry – Spinger 2006, Vol. 225, pp. 384-385.

• L.V. Sorokin Anomalous seismo-electromagnetic emission related with seismic waves // The proceedings of “7-th International Symposium on Electromagnetic Compatibility and Electromagnetic Ecology”. Saint-Petersburg, 26 – 29 June 2007, pages 293-296, DOI: 10.1109/EMCECO.2007.4371715.

• L.V. Sorokin Lightning triggering related with seismic waves // The proceedings of “7-th International Symposium on Electromagnetic Compatibility and Electromagnetic Ecology”. Saint-Petersburg, 26 – 29 June 2007, pages 297-300, DOI: 10.1109/EMCECO.2007.4371716,

• L. V. Sorokin Seismo-electromagnetic emissions related to seismic waves can trigger TLEs // Coupling of thunderstorms and lightning discharges to Near-Earth Space. Editors Norma B. Crosby, Tai-Yin Huang, Michael J. Rycroft. Proceedings of the Workshop (Corte, France, 23-27 June 2008), Melville, New York, 2009, AIP Conference proceedings. – Vol. 1118. – Pp. 58-67.

Thank you for your attention!