•Research Topics:
1. X-Ray & Gamma-Ray Emission From Lightning
Rocket Triggered Lightning
Thunderstorm Energetic Radiation Array (TERA)
2. X-Ray Emission From Laboratory Sparks
3. Terrestrial Gamma-Ray Flashes
4. Runaway Breakdown Theory
5. Jovian Lightning
The Mysteries of Lightning
  by Dr. Joseph Dwyer
•A garden variety spark?
Consider the conventional spark, the kind you get when you touch a doorknob after walking across a carpet. When you traverse the carpet, your shoes rub off electrons and you accumulate an electrical charge, which produces an electric field between you and other objects in the room. For small electric fields, air is a good insulator and electrical current cannot flow in any appreciable amount. However, as your finger approaches the knob, the electric field becomes locally enhanced. If it reaches a critical value of about 3 million volts per meter, called the breakdown field, then a discharge occurs; the air becomes a conductor and current bridges the gap, i.e. you get a spark. Because in some ways lightning resembles a big spark, it seems natural to assume that something similar must be going on inside thunderstorms. Unfortunately, there is a problem with this simple picture: Decades of balloon, aircraft and rocket measurements made directly inside the clouds rarely find fields above about 150,000 volts per meter, much too low to cause air to breakdown like it does when we touch a doorknob.

Scientists have been trying to solve this conundrum of what causes lightning for many years, and, until recently, mainly had two explanations: First, perhaps large electric fields really do exist inside thunderstorms but only in relatively small volumes that are unlikely to be measured. Although such a scenario cannot be ruled out observationally, it is not all-together satisfying, since we are replacing one problem with another: how do you get big electric fields in small volumes? The second explanation comes from laboratory experiments that have shown that the electric field needed to produce a discharge is reduced substantially when raindrops or ice particles are present in the air--as they are inside thunderstorms. Unfortunately, while the addition of rain or ice does alleviate some of the discrepancy, thunderstorm electric fields still appear to be too low by perhaps 50% to be explained entirely by this mechanism. In addition to the process that initiates lightning, scientists are also uncertain about how lightning actually propagates many kilometers through the air. It is known that the mechanism involves the formation of a leader, a hot channel that can breakdown the air and transport charge over long distances. However, exactly how this all occurs is somewhat mysterious and previous efforts to model these processes have not been entirely successful.

•Stranger than we thought?
In 1992, a new idea emerged that has shown promise for explaining what is happening inside thunderstorms and how lightning is occurring. A. V. Gurevich of the P. N. Lebedev Institute of Physics in Moscow, G. M. Milikh of the University of Maryland and R. Roussel-Dupre of Los Alamos National Laboratory proposed a new and unusual kind of discharge that involves very fast electrons moving under the influence of electric fields. It turns out that in sufficiently strong fields electrons can gain very large amounts of energy, and the electrons are said to run away. Actually, the idea of runaway electrons is not new and dates back to Nobel Laureate C. T. R. Wilson, who in 1925 suggested that under some circumstances electrons can run away in the atmosphere. However, an early attempt by A. V. Gurevich in 1961 to use runaway electrons to model a discharge was widely considered inapplicable to the atmosphere, since the electric field required for such a discharge is at least a factor of 10 larger than the conventional breakdown field. In contrast, the new theory of runaway breakdown uses an avalanche of relativistic electrons and only requires an electric field 1/10 as large as that needed for a conventional breakdown in dry air. At thunderstorm altitudes, the runaway breakdown threshold is about 150,000 volts per meter-precisely in the range of values measured inside the clouds. Indeed, it is probably not a coincidence that the maximum observed electric field inside thunderclouds and the threshold for runaway breakdown are about the same value, since runaway breakdown would efficiently discharge the field if it were to rise much higher.

In normal discharges, all the electrons have low energies and travel slowly, so at most the emitted electromagnetic spectrum extends into the ultra-violet range. In runaway breakdown, on the other hand, the electrons travel very close to the speed of light, and, consequently, produce large quantities of x-rays as they collide with air molecules. Therefore, one way to test for runaway breakdown is to search for x-rays.

Because of Wilson's work, scientists have attempted to observe x-rays from thunderstorms and lightning since the 1930's. Unfortunately, such measurements are very challenging to make and so until recently have produced mostly ambiguous results. One difficulty is that x-rays don't travel very far through the atmosphere and are usually absorbed within a few hundred meters of the source. Another difficulty is that thunderstorms are electromagnetically very noisy environments, and lightning, in particular, emits large amounts of radio frequency noise. This noise produces the familiar crackle on an AM radio-and that can be from kilometers away. Now imagine trying to make sensitive electrical measurements with lightning less than 100 meters away. It is, therefore, important to be able to distinguish real electrical signals produced by x-rays from spurious ones produced by the radio frequency emission. For this reason, many of the early measurements were not readily accepted.

The situation got more interesting in the 1980's when G. K. Parks, M. McCarthy and collaborators at the University of Washington made aircraft observations inside thunderstorms, and later K. B. Eack (now at New Mexico Tech) and collaborators made a series of balloon soundings through thunderclouds. These observations provided tantalizing hints that occasionally large bursts of x-rays are produced by thunderstorms. The source of these x-rays could not be deduced, but it seemed to be associated with the enhanced electric fields inside the cloud. Interestingly the x-ray emission sometimes began right before lightning and stopped once the lightning occurred, perhaps because lightning shorted out the electric fields needed to produce runaway breakdown. Besides runaway breakdown, there are no other mechanisms known to produce such large quantities of x-rays in our atmosphere. Even during the bright return stroke of lightning, when the channel can reach 30,000 °C (54,000 °F), virtually no x-rays are produced by this high temperature.

•Superman, put on your sunglasses!
Our view of lightning took a radical turn in 2001, when C. B. Moore and collaborators at New Mexico Tech reported observing strong x-ray emission from several natural lightning strikes on top of a tall mountain. Unlike the earlier aircraft and balloon observations, these x-rays seemed to be produced by the lightning itself and not by the large-scale electric fields inside the thundercloud. Furthermore, the emission seemed to be associated with the lightning stepped-leader process, which was something entirely new.

All right, here is where I enter the picture: As a physicist I have always been interested in how x-rays and gamma-rays are produced. While such energetic radiation is commonly created both at distant astrophysical sites and in our solar system, usually this only occurs in the vacuum of space where energetic particles can travel unimpeded. Consequently, I became fascinated by the idea that the same kind of x-rays produced in places like solar flares could also be made by thunderstorms and lightning. Because I live in Florida, which has lots of thunderstorms, I decided to look for myself to see if these supposed x-rays really existed. In 2002, with funding from the National Science Foundation, my group (H. K. Rassoul, M. Al-Dayeh, L. Caraway, B. Wright and A. Chrest) at the Florida Institute of Technology, in collaboration with M. A. Uman and V. A. Rakov at the University of Florida, began a systematic campaign to search for x-ray emission from lightning, thereby clarifying the role of runaway breakdown in our atmosphere.

To reduce the problems of spurious electromagnetic signals we placed sensitive x-ray detectors inside heavy aluminum boxes, designed to keep out moisture, light and radio frequency noise. In order to get close enough to lightning to measure x-rays, we set up our instruments at the International Center for Lightning Research and Testing (ICLRT) at Camp Blanding, FL. The ICLRT is a facility run by the University of Florida, equipped to measure, among other things, the electric and magnetic fields, and optical emission associated with lightning. Moreover, the facility is capable of artificially triggering lightning from natural thunderstorms using small rockets.

At the ICLRT, when a thunderstorm is overhead and the electric field on the ground reaches several thousand volts per meter, a small rocket is launched that uncoils a spool of thin Kevlar coated copper wire, one end of which remains attached to the ground. As the rocket rises, the vertical grounded wire enhances the electric field at the rocket's tip, resulting in an upward propagating leader that eventually snakes its way up into the thundercloud. Electrical current into the leader quickly vaporizes the wire, and the resultant lightning usually strikes the rocket launcher. The advantage of using triggered lightning over natural lightning is that with triggered lightning the exact time and place of the lightning strike is experimentally controlled. Furthermore, the experiment can be repeated over and over with dozens of triggered lightning flashes produced at the ICLRT each summer.

To be honest, when we first set up our instruments at the ICLRT, I wasn't really expecting to measure any x-rays from lightning. The New Mexico Tech observations were fascinating but at that point had not been independently verified, and so in light of the long history of negative or ambiguous x-ray results, they needed to be viewed with caution. I figured we would do a series of careful experiments and most likely rule out lightning as a source of x-rays. For this reason, after we made our first triggered lightning measurements, I didn't get around to looking at our data for over a week. When I finally sat down with my graduate student, Maher Al-Dayeh, and plotted the data from the x-ray detectors, I nearly fell off my chair. To my surprise--and to the surprise of just about everyone else--we discovered that not only does triggered lightning produce x-rays, it produces lots and lots of x-rays just about every time. Indeed, the flashes were so intense in x-rays that our instruments were temporarily blinded by the radiation.

Subsequent experiments over the next year showed that the x-ray emission is produced by the lightning dart leader with possibly some contribution from the beginning of the return stroke, and it is composed of x-rays with energies extending to at least twice that found in a chest x-ray. Furthermore, the x-ray emission was not produced continuously, but instead occurred in rapid bursts a millionth of a second apart. Indeed, if we somehow had x-ray vision, like Superman, lightning would look quite different than what we are used to: As the lightning leader propagates downward, we would see a rapid series of bright flashes descending from the clouds. The flashes would intensify as they approached the ground, ending with a very intense burst at the instant the return stroke begins. The optically bright return stroke that follows, on the other hand, would look black in x-rays.

The mechanism for producing these x-rays is presently not known, but almost certainly involves runaway breakdown. The measurements also suggest that the electric fields produced by lightning leaders are much, much larger than what was previously believed possible. The observed x-rays don't seem to have anything to do with the vaporized wire used to trigger the lightning, since they all occur well after the wire is gone. Indeed, since the initial discovery of x-rays from triggered lightning, we have observed natural lightning strikes at the ICLRT as well. These data also show beautiful x-ray emission from the stepped-leader phase, confirming the earlier New Mexico Tech measurements.

The observations of x-rays from lightning illustrate that our current understanding of how lightning works is far from complete, since none of the conventional models of lightning can explain the x-ray emission. The observations also show that runaway breakdown may be a common phenomenon is our atmosphere. Because air molecules hinder the acceleration of fast electrons, runaway breakdown has a much easier time at thunderstorm altitudes or above where the air is thinner, yet we see evidence for runaway breakdown near the ground at sea level (most of the x-rays we observe come from the bottom 50 meters or so of the lightning channel). In other words, if it can happen there it can happen anywhere.

•Meanwhile, back inside the thunderstorm
So, what about lightning initiation inside thunderstorms? In the last few years,there have been substantial theoretical advances in understanding runaway breakdown and promising models have emerged that involve the combination of high-energy cosmic-ray air showers and runaway breakdown. If these models are valid, lightning on Earth has a direct physical connection to exploding stars in other parts of our galaxy. A fascinating piece of evidence further supporting runaway breakdown inside thunderstorms came from our experiment at the ICLRT last summer. During the last rocket launch of the season we fortuitously caught a huge burst of very high-energy x-rays, more appropriately called gamma-rays, using three detectors placed 650 m from the lightning channel. The energies of the individual gamma-ray photons extended up to over 10 MeV, much larger than emitted by the dart leaders and about 100 times larger than found in dental x-rays! Anyone who pictures scientists as calm and reserved should have seen how we reacted when we saw that gamma-ray flash. Instead of looking at data on a computer, one might have thought we had just seen our favorite team score the winning touchdown at the super bowl.

Interestingly, using lightning channel current and electric field data, along with properties of the gamma-rays we have inferred that the source of the emission was likely many kilometers up in the thundercloud. We did not expect to see gamma-rays from such an altitude due to the absorption in the atmosphere, but apparently the gamma-ray intensity at the source was so large that some gamma-rays were able to make it down to the ground. The observation suggests that massive runaway breakdown must have occurred within the thundercloud in a process related to the initiation of triggered lightning. Furthermore, these observations demonstrate that it is possible to study this phenomenon on the ground, which is experimentally much simpler.

We are now in the process of greatly expanding the number of x-ray instruments at the ICLRT from 5 to over 36, covering the entire 1 square kilometer of the Camp Blanding site. This will allow the practical study of natural lightning as well as triggered lightning and should increase the odds of measuring more gamma-ray bursts from the thunderclouds. The x-ray and gamma-ray emission should serve as a probe to help determine the electric fields in regions that are otherwise very difficult to measure and it should allow us to better understand the breakdown processes that initiate lightning and facilitate its propagation over many kilometers through the air. Using x-rays to study lightning is still new and consequently just about every time we record x-rays from lightning we discover something we didn't know before-a rare and exciting opportunity for a scientist like me. We have already learned that lightning is not just an ordinary spark like we get when we touch a doorknob. It involves a more exotic kind of discharge that produces runaway electrons and x-rays. Since x-rays allow us to look at lightning in a novel way, I believe that this research may help us finally solve the puzzle started by Benjamin Franklin two and a half centuries ago.

•Conventional Electrical Breakdown
When positive and negative electrical charges are separated, they create a potential difference, measured in volts, abbreviated as V. For example, the potential difference between the positive and negative terminals of an AA battery is 1.5 V. The electric field is the rate at which the potential changes with distance. The average electric field across a 5 cm long AA battery is about 0.3 V/cm or, equivalently, 0.003 V/m.

In a conventional discharge, free electrons move under the influence of an electric field, with frequent collisions with air molecules preventing the individual electrons from gaining too much energy. Some of the electrons are absorbed by attaching to oxygen atoms. Free electrons can also be created, for instance, when an electron hits an air molecule hard enough to knock out an additional electron. The original electrons plus the newly freed ones are then accelerated by the electric field. For low electric fields, the rate of free electron loss through attachment is greater than the rate of free electron gain through collisions. As the electric field increases, however, the balance shifts. Above the so-called breakdown field, free electrons are created faster then they are lost, and an avalanche of free electrons is produced, with the number increasing exponentially with time. Because moving electrons create an electrical current, above the breakdown field air becomes a good conductor of electricity. For example, dry air breaks down and becomes a conductor, e.g. it sparks, at about 3,000,000 V/m, so if instead of 1.5 V, an AA battery had a voltage of roughly 150,000 V, it would spark across its terminals.

•Runaway Breakdown
The runaway breakdown of air occurs when the rate of energy gain experienced by an electron in an electric field exceeds the rate of energy loss through collisions with air molecules. The numerous collisions result in an effective drag force on the electron, which works to slow it down. For subatomic particles like electrons this drag force does not behaves according to our ordinary experience. When you stick your hand outside your car window while driving down the road, you feel a drag force due to the motion of your hand through the air. As you go faster, the drag force increases, and as you slow down the drag force decreases. Electrons, on the other hand, behave differently: when electrons are moving sufficiently fast, the drag force actually decreases the faster the electron goes. As a result, if an electron is moving very fast its drag force can become smaller then the force from the electric field and the electron will run away, continuously gaining energy until it is moving very close to the speed of light. The electric field necessary for runaway breakdown to occur is about 300,000 V/m at sea level and lower at thunderstorm altitudes, a factor of ten lower than the field needed for a conventional breakdown.

As electrons run away, they will occasionally bump hard into air molecules knocking off other fast electrons. These "knock-off" electrons can subsequently run away, producing more energetic "knock-off" electrons and so on. The result is a large avalanche of high-energy electrons that increase exponentially with time and distance. The whole process can be initiated by as little as one high-energy "seed" electron, which are plentiful in our atmosphere due to the steady background from cosmic-ray air showers and radioactive decays.

As the runaway electrons move through the air they create large amounts of ionization, e.g. low-energy electrons and ions, plus they produce x-rays and gamma-rays as a byproduct of collisions with the air molecules. This process is known as bremsstrahlung (German for braking radiation) and is exactly how an x-ray tube in a dentist office works.

More information on Runaway Breakdown

•Rocket-Triggered Lightning
Both natural and triggered lightning flashes are usually composed of several strokes. For triggered lightning, each stroke starts as a downward propagating column of charge called a dart leader that, near ground, more or less follows the path left by the rocket and triggering wire. The dart leader brings down charge from the cloud and ionizes the channel as it moves. Once the dart leader connects to the ground, a short circuit is created and a large pulse of current, called the return stroke, flows through the channel. The current in the return stroke quickly heats the channel causing the visible light that we see, and the subsequent rapid expansion of the hot air causes the thunder that we hear. After the return stroke, another dart leader can follow and the whole process repeats. The quick succession of strokes is what causes the lightning channel to flicker. In natural lightning, the role of the rocket is played by a stepped leader, which forges the ionized path, extending from the cloud to the ground. However, the strokes of triggered lightning are very similar to the subsequent strokes of natural lightning.
•X-ray Instrumentation
The x-rays from lightning are detected using crystals of sodium-iodide (NaI). When sodium-iodide absorbs energetic radiation, such as x-rays, it emits a large optical light pulse. This light is captured by a photomultiplier tube, which converts the light pulse into a current pulse that is recorded. The size of the current pulse is proportional to the size of the light pulse, which in turn is proportional to the energy of the detected x-rays. As a result, measuring the size of the output electrical signal from the photomultiplier tube gives a direct measurement of the x-ray energy.

To prevent spurious signals from being detected, the sodium-iodide/photomultiplier tube detectors are placed inside heavy aluminum boxes, which are completely sealed to keep out light, moisture and radio noise, and only allow x-rays to enter. The instruments are battery powered and transmit their data via fiber optics to a shielded trailer that contains the data acquisition system.


The Thunderstorm Energetic Radiation Array (TERA) is an experiment designed to measure energetic electromagnetic radiation (x-rays and gamma rays) from lightning. We are currently looking at the radiation from stepped leaders, dart leaders, and return strokes. We've even caught a gamma ray burst associated with a triggered CG flash! Ultimately, the goal is to determine what role runaway breakdown plays in lightning initiation and propagation.

More information on TERA

Copyright © GPL, All Rights Reserved.

Title Photograph © Marc-André Besel