EGU General Assembly 2020
© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Meter-scale Measurements of VHF structure of natural leader streamers

Brian Hare1, Olaf Scholten1, Joseph Dwyer2, Ute Ebert3, Sander Nijdam3, and the LOFAR CR KSP*
Brian Hare et al.
  • 1University of Groningen, KVI-CART, Groningen, Netherlands
  • 2Department of Physics and Space Science Center (EOS), University of New Hampshire, Durham NH 03824 USA
  • 3U/e, Eindhoven University of Technology, Eindhoven, The Netherlands
  • *A full list of authors appears at the end of the abstract

We will present maps of negative leaders imaged in the 30-80 MHz band by the LOFAR radio telescope, which is a distributed radio telescope in the Northern Netherlands that can map lightning with meter length and nanosecond timing accuracy. These VHF images show that negative leaders emit bursts of VHF that are about 1-3 µs in duration, most likely in relation to leader stepping. The median time between bursts is around 40 μs, and the median distance is about 7.5 m. Each of these bursts contains around 3-10 discrete VHF pulses. 2/3 of these pulses are consistent with coming from the same location (with 1 meter location accuracy), and the other 1/3 come from up to 3 m away. These data are consistent with the hypothesis that these VHF bursts are due to corona flashes during leader stepping, that the discrete pulses we locate are due to the few very strongest streamers in the corona flash, and the majority of streamers in a corona flash are too weak to be observed as discrete VHF pulses. From these data, we estimate that the strongest streamers in a natural corona flash emit about 4x10-6 J in our 30-80 MHz band.


A. Bonardi, S. Buitink, A. Corstanje, H. Falcke, T. Huege, J.R. Horandel, G.K. Krampah, P. Mitra, K. Mulrey, B. Neijzen, A. Nelles, H. Pandya, J.P. Rachen, L. Rosetto, T.N.G. Trinh, S. ter Veen, T. Winchen

How to cite: Hare, B., Scholten, O., Dwyer, J., Ebert, U., and Nijdam, S. and the LOFAR CR KSP: Meter-scale Measurements of VHF structure of natural leader streamers, EGU General Assembly 2020, Online, 4–8 May 2020, EGU2020-7957,, 2020

Comments on the presentation

AC: Author Comment | CC: Community Comment | Report abuse

Presentation version 1 – uploaded on 29 Apr 2020
  • CC1: Comment on EGU2020-7957, Alexander Kostinskiy, 04 May 2020

    Dear Prof. Hare,

    On slide 3 (Conclusions) you write that the charge of one streamer can be 8 μC, and on slide 18 you ask the question: how many streamers are in the flash? There is a definite answer to this question: never one streamer at pressures of 0.1-1 atmosphere can carry such a large charge. A similar phenomenon at the steps of a negative leader is always an outbreak of many streamers. The charge of one streamer head was measured due to sectioned electrodes (1x1 mm) and it is 0.2-1.4 nanocoulomb (p.174, Bazelyan, E. M. and Raizer, Y. P. (1998), Spark discharge. Boca Raton, FL: CRC Press). Thus, your flash has about 5-10 thousand streamers. Most likely, you are dealing not with one streamer flash, but with several streamer flashes going one after another during one microsecond (for example, Figures 1, 2 a, b, Kostinskiy, A. Yu., Marshall, T.C., Stolzenburg, M . (2019), arXiv: 1906.01033).

    On slides 10-12, you analyze the parameters of the VHF sources. I’m not sure that such an analysis can be carried out for all sources at once, since streamers are emitted by the main negative channel, space leaders and space stems, which are always several in the streamer crown of a negative leader. A more detailed analysis could be proposed, which is based on the known properties of these components of the streamer corona of the negative leaders of a long spark. You got a very interesting result. The average length of the negative lightning leader is only 7.5 meters. This coincides with our ideas that there can be no steps of a negative leader 100 meters long.


    • AC1: Reply to CC1, Ute Ebert, 05 May 2020

      You are doubting that a single streamer can carry 8 uC of charge. But be aware that streamer currents, charges, velocities and radii can vary by orders of magnitude, much beyond the values in the old literature that you are citing. Please check papers by Briels et al in 2006 and 2008, that are quoted as references 30 and 35 in our recent Phys. Rev. Lett. DOI: 10.1103/PhysRevLett.124.105101 .

      Best regards,

      Ute Ebert

      • CC3: Reply to AC1, Alexander Kostinskiy, 05 May 2020

        Dear Ute,

        I have no doubt that in your theoretical articles one streamer can gain such a large charge. I know well your wonderful theoretical and experimental work on short streamers that have not lost their galvanic connection with the electrode. But in my reasoning, I also rely on a huge experimental material (including our group) of the last 50 years on long streamers (0.5-3 meters) and long sparks (3-150 m). I am talking only about positive streamers that move in electric fields of 450-1000 kV / (m atm) and negative streamers (> 1000-1200 kV / (m atm)), as well as pulse fronts no shorter than 0.3-1 μs. Under these conditions, all the experimental material says that even 1-10 nanocoulomb cannot be realized in one streamer. I'm not talking about subnanosecond discharges and streamers in external electric fields of 10-30 MV / (m atm), which are not applicable to lightning conditions. You refer to excellent theoretical work. But is there at least one experiment where a streamer with a charge of 1 microcoulomb was measured in such fields? I refer not to old, but to classical works, which in their experimental part will never become obsolete, since they are based on direct experiments. The interpretations of these experiments may change, but the experimental data cannot. All streamer flashes that have been observed in long sparks contain many (tens and hundreds) of streamers at the same time, which you yourself have observed many times in your very interesting experimental work with Pavel Kochkin and his leader. With an electrode with a diameter of 10 mm, the total charge of the first streamer flash, which is injected into the volume, is 6–8 microcoulomb. A charge of 10-50 microcoulomb is injected during the steps of the positive and negative leader 0.5-2 meters long (in old works and right now Kostinskiy et al. 2018, doi: 10.1029 / 2017JD027997). Do you think that the entire discharge takes 1 streamer? And what did you see in the pictures that Pavel shot? Colleagues and ourselves directly shot with streak cameras, infrared cameras and cameras that you used (4Picos), and we always saw a huge number of streamers in the flash. At the same time, we streamed the streamer corona with microwave radiation, and this is a direct measurement of the average plasma concentration and we have never seen anything like it with an accuracy of several orders of magnitude (Bogatov, Kostinskiy, Mareev, Rakov, 2020, doi: 10.1029 / 2019JD031826). How such a large charge does not branch and is not divided into small charges? If the charge branches, then why does it remain so large? If there are a lot of streamers in the flash and they are of the same sign, then they reduce each other's electric field very much.

        Best regards,

        • CC4: Reply to CC3, Brian Hare, 05 May 2020

          Dear Alexander,

          I should be clear, we are open to other explanations. The biggest difficult is explaining slide 13. How do you explain the pulse structure with corona flashes? IE- a series of desrcete pulses, each of which has nano-second order width, with a random order and exponential amplitude distribution? Keep in mind that large-scale (ie 10 m and larger) currents are below our frequency range. We are not sensitive to the main stepping current. You do raise some very good questions, I certainly wouldn't mind explaining our data with a less-contraversial hypothesis, but I cannot explain slide 13 without streamers. Pavlo's negative laboratory leader with Ute, for example, only shows three corona flashes, nothing like what we see.  

        • AC2: Reply to CC3, Ute Ebert, 05 May 2020

          Dear Alexander,

          your reply is based on the assumption that I refer to theory and simulations, but the references I gave are to experimental work! We have done quite some experimental work in the course of years in Eindhoven, with Tanja Briels and Sander Nijdam et al with detailed plasma diagnostics with up to 60 kV at applied physics, and with Pavlo Kochkin, Lex van Deursen et al at electrical engineering with a 2.4 MV Marx generator. Experimental activities are continuing, and keep producing puzzling new results.

          Best regards,


          • CC6: Reply to AC2, Alexander Kostinskiy, 05 May 2020

            Dear Ute,

            Thanks so much for your clarification. I read most of your experimental articles carefully and tried to understand. Most of your articles, together with Pavlo Kochkin and Lex van Deursen, where the discharge gap was about 1 meter and the voltage reached 1 MV, I read even more carefully. Maybe I missed some of your articles, but in these works there were very interesting results on energetic particles (Pavlo called them photons), but there were nowhere measurements of the charge of streamers or streamer flashes that would contradict earlier classical results (charge of one streamer about 1 nanocoulomb). If you sent links to such new experimental work that I do not know, then I would be very grateful to you.

            Best regards,



            • AC3: Reply to CC6, Ute Ebert, 06 May 2020

              Dear Alexander,

              you are refering to my papers with Pavlo and Lex, but most of my experimental papers are with Sander Nijdam and the early ones with Tanja Briels. 

              In my first reply I was already refering to Briels et al, 2006 and 2008. Here are the references explicitly: 

              Circuit dependence of the diameter of pulsed positive streamers in air, 
              T.M.P. Briels, J. Kos, E.M. van Veldhuizen, U. Ebert, J. Phys. D: Appl. Phys. 39, 5201-5210 (2006)

              Positive and negative streamers in ambient air: measuring diameter, velocity and dissipated energy, 
              T.M.P. Briels, J. Kos, G.J.J. Winands, E.M. van Veldhuizen, U. Ebert, J. Phys. D: Appl. Phys. 41, 234004 (2008)

              Positive and negative streamers in ambient air: modeling evolution and velocities, 
              A. Luque, V. Ratushnaya, U. Ebert, J. Phys. D: Appl. Phys. 41, 234005 (2008)

              A major insight is that streamer diameters in air depend on the voltage rise time. In experiments you need nanosecond rise times to form thick streamers. In the lab they grow from inception clouds as we have analyzed in more detail in later papers, both experimentally and theoretically. A slow voltage rise time creates a small cloud that destabilizes into thin and slow streamers.

              If you have trouble approaching the journal you can pick the articles up from .

              By the way, I have reviewed these results in JGR in 2010, and there is a new topical review on streamers together with Sander Nijdam and Jannis Teunissen presently under review.

              Best regards,


              • CC8: Reply to AC3, Alexander Kostinskiy, 06 May 2020

                Dear Ute,

                Thank you so much for your articles and clarifications. I have carefully read these articles, I have them, and I will gladly read them again in connection with our discussion. In this discussion, too, I never questioned your breakthrough results in the field of short streamers, thin electrodes with low capacitance, and extremely short voltage rise fronts. But my first comment was on streamers (more precisely, streamer flashes) that appear during the development of leaders who are several meters or more in size. As you well know, the capacitance of these plasma formations is much larger than in your experiments with nanosecond fronts, since the capacitance of the leaders is determined by the ionic cover (sheath) of the leaders, which is at least 2-3 meters in diameter. This large capacity of the leaders, as well as the structure and capacity of the streamer zone of the leader, will never allow the formation of a front of increase in voltage on the leader’s head less than 0.3-1 μs. Therefore, as you correctly described the mechanism, the streamers of even the most powerful steps of the negative and positive leaders will be initiated in flashes (that is, hundreds and thousands of streamers at the same time) and have a charge of about one nanocoulomb per streamer (speculatively it can be assumed that the charge of one streamer may reach even 10 nanocoulomb, but so far we don’t know such data for a long spark). The charge of the entire flash, as you know, can reach up to 100 microcoulomb in a long spark, and in lightning it can be up to 0.8-1 millicoulomb. This is just what I wanted to explain in my initial remark on the slides in the presentation of your esteemed colleagues. 



    • CC2: Reply to CC1, Brian Hare, 05 May 2020

      the time structure of our pulses is inconsistant with being due to corona flashes. For example, we have an exponential spectrum of many pulses per leader step, and the pulses come in a random (IE independant) rate, as opposed to a regular rate. Corona flashes cannot do this.

      • CC5: streamer flashes, Alexander Kostinskiy, 05 May 2020

        Dear Brian

        Since the plasma channels of lightning are in free space, absolutely any movement, association, and propagation of leader channels is accompanied by streamer flashes, since without streamers, current will not flow through the plasma channels. And if the current stops flowing, then the electric field is displaced from the plasma and the plasma decays. Flashes can be very small, both when moving a positive leader without large jumps, and large, reaching tens of meters in a negative leader. It’s just that your distribution combines several different plasma processes of different scales. For each case, of course, a thorough individual analysis is required, but from what we now know about a long spark (up to 150 meters) and lightning (up to tens of kilometers), we can assume. For a negative leader, there are the largest steps up to 10-20 meters in size, when a powerful potential is brought to the border of the streamer zone and the largest streamer flashes of the negative leader occurs. After this, space-stem (the second object) arise, which are connected by positive streamers to the main channel and small jumps, due to flashes of negative leaders from the other end, move the negative streamer zone forward. There are always several. And everywhere there are streamer flashes 1-15 A. Inside the space-stems, space-leaders (the third object) are born. The act of their nucleation is very likely ionization-heating instability, which also occurs with a streamer flash, which in turn gives rise to a new space-stem in front of the space-leader. Even in the negative leader’s crown, there are several space leaders, as well as space-stems, in the long spark of a leader. All of them move at different speeds, giving rise to flashes of streamers of different spectra. All these strange elements are connected exclusively by streamers and nothing more. But at the other end of the bi-directional leader, the positive leader moves with small jumps due to the same ionization-heating instability, which generates very small pulsations in the positive streamer corona. But there are theoretically completely incomprehensible situations now when the positive leader of a long spark makes a giant jump, for example, Kostinskiy et al. 2018, doi: 10.1029 / 2017JD027997. The positive leader of lightning also has such large jump in some modes (pp. 270-272, Rakov, V. A., and M. A. Uman (2003), Lightning: Physics and Effects, Cambridge Univ. Press, Cambridge). But that is not all. There are very large flashes of light and changes in the electric field that occur during IBPs. In addition, there are powerful VHF flashes called CID or NBE (their light is small). Of course, you know all these plasma objects, but I have listed them to emphasize that all these processes are necessarily accompanied by flashes of the streamer corona. Apparently they have developed into your spectrum. Of course, there are also fluctuations in the electric potential inside the hot plasma channels, which are reflected from their ends and add to this system a whole lot of all kinds of radiation of a different range. You probably mean them. Maybe I'm wrong, but an analysis looks more promising, which relies on the various structural elements of the plasma formations of lightning and their interaction, which will lead to several graphs.

        Best regards,

        • CC7: Reply to CC5, Brian Hare, 06 May 2020

          Thank you for your extensive reply. I purposefully worded my slides to try to try and encourage this kind of discussion. Writing this work was difficult, as we had to work through each of these different mechanisms you mentioned, and have become convinced that they do not explain our data. Please keep in mind, our interest is the origin of a single pulse, as opposed to a train of pulses (we call a burst). One burst can last a microsecond, and we believe is on the whole due to corona flashes, but the individual pulses have, at most, a time width of sigma = 10 ns. This severely limits the potential origin of these pulses. I will go through many of the potentialities you mention, and try to explain why we believe they are excluded.

          1) Corona Flash. Corona flashes last longer than 10 ns, thus they are too low-frequency to explain a single pulse. Each corona flash may produce multiple pulses, but individual pulses must be due to structures INSIDE the corona flash (which is only streamers).

          2) Space Stems/Space Leaders. Besides the time-width problem, if you assume 1 space stem/leader leads to one pulse, this naturally implies that there must be a physical spread between the locations of radio pulses. We see that each radio pulses are almost all from the same location (this is a hypothesis-independent observation). If a space stem/leader makes multiple radio pulses, then we are back to the question of what is the origin of a single radio pulse.

          3) IBP, CIDS, etc all occur at the beginning of the flash. In this work we have stayed away from these times and only focused on normal negative leaders, to avoid this confusion.

          4) long scale currents. Eg, positive leader jumps and main negative leader stepping current pulses. These are all well below our frequency regime, IE, they have widths longer than 10 ns. We do not see anything with 1 sigma widths longer than 10 ns (this would require the signal to either be incoherent or narrow band, which we do not observe). Note that the tip of the positive leader is entirely invisible to us. (See: ). It is possible that REALLY strong positive leaders (say, positive leaders approaching ground) may become visible, but we have not observed any of those.


          We are not claiming we see normal streamers. The fact that positive leader propagation is invisible implies that normal streamers are also invisible. Here we focus on the idea that the streamers must be very large. The other possibility is that we have normal sized streamers that accelerate very quickly. Based on our spatial distribution my guess is that this may be the case when the ionization wave breaks up into streamers, at the very beginning of a corona flash.


          The most sure way to convince me I am could be wrong (which I wouldn’t mind), would be to take a current pulse from a corona flash, calculate the radiated E-field, apply a band-pass filter from 30-80 MHz (exact shape not critical at this stage), and show me that the resulting width is consistent with the pulse I show on slide 15.


          • CC9: Reply to CC7, Alexander Kostinskiy, 06 May 2020

            Dear Brian,

            Thank you very much for the great and detailed answer,

            I have 2-3 candidates for the position of a pulse emitter with a duration of 10 nanoseconds (on slide 15) and they are not streamers. So that my answer is reasoned and you understand exactly what I'm talking about, I will need to send you several articles where I could refer to specific images. If you are interested in such a specific answer, please send me your email address. My address is .