“I thought I was going to die, I really did,” said Joseph Dwyer, professor of physics and astronomy at the University of New Hampshire. It was August 21, 2009. Cruising at an altitude of about 45,000 feet, the Gulfstream V business jet, modified to enable scientists like Dwyer to study atmospheric physics, was flying around a thunderstorm. “I’m not sure exactly what happened, but we went through the middle of the thunderstorm,” said Dwyer. The ensuing turbulence turned the flight into a violent roller coaster ride, disorienting him. “Your inner ear is all messed up. You can’t tell if you are upside down or right side up,” he said. While he was contemplating his imminent demise, an instrument designed to detect high-energy particles produced in thunderstorms lit up. Dwyer went from thinking “I’m going to die,” to, “Oh, look at that, what’s going on there?”

The instrument, called the Airborne Detector for Energetic Lightning Emissions (ADELE), had captured a gamma-ray glow—long-lasting, high-energy emissions—from a thundercloud. On the same flight, ADELE also captured a much more powerful emission, called a terrestrial gamma-ray flash (TGF), an extremely elusive burst of energetic electromagnetic radiation that only infrequently accompanies lightning in thunderstorms. The TGF came from a thundercloud about 10 kilometers away. Until then, TGFs had only been seen from the ground and by satellites in space. The sighting was the first-ever made from an airplane.

ADELE was birthed in the laboratory of Dwyer’s kindred lightning chaser David Smith, professor of physics at UC Santa Cruz and a member of the Santa Cruz Institute for Particle Physics, which is involved in cutting-edge experimental and theoretical particle physics and particle astrophysics. Dwyer, Smith, and other physicists have doggedly pursued these enigmatic high-energy outbursts from thunderstorms, seeking to gain a better understanding of how lightning develops and behaves. In addition to flying ADELE close to thunderstorms and even through a hurricane, their intrepid quest has also involved deploying instruments near mountaintops in Mexico and on the ground in Japan, all to study the inner physics of thunderstorms.

The research has revealed a smorgasbord of strange phenomena: TGFs that barely last a millisecond but are powerful enough to blind the electronics in satellites overhead; thunderstorms that glow in gamma-ray frequencies; anti-electrons (also known as positrons) generated by the same processes that produce gamma-ray flashes; and even a barrage of neutrons knocked out of air molecules by gamma rays. “It’s a weird wonderland inside thunderstorms,” said Dwyer. “There’s a lot of strange stuff going on.”

Eyes in the sky

The first clear indication of such high-energy strangeness came in 1994, when the Compton Gamma Ray Observatory (CGRO), a NASA satellite launched in 1991 to primarily study astrophysical gamma-ray bursts from space, produced an anomalous result. Composed of photons of the highest possible frequency, these gamma-ray bursts contain the highest possible electromagnetic energies. Unexpectedly, however, the CGRO had detected flashes of gamma rays coming from thunderstorms below, not from space above. Despite the satellite being in space, the intensity of these terrestrial flashes overwhelmed its detectors. Since gamma rays are mostly absorbed by the Earth’s atmosphere, scientists argued that these gamma-ray flashes had to be originating at altitudes of more than 30 kilometers, above the dense regions of the atmosphere. For about a decade following this 1994 detection, TGFs were thought to be associated with “sprites,” jellyfish-shaped lightning that occurs in the uppermost parts of the atmosphere during thunderstorms.

An early hint that TGFs were not connected to sprites and developed from much deeper down in the atmosphere came in 2004. Dwyer, then at the Florida Institute of Technology, was studying lightning triggered by firing meter-long rockets up into thunderstorms. Normally, lightning occurs when negative and positive charges build up in clouds. When these charge centers grow big enough, they can spark, creating intra- or inter-cloud lightning. Positive charges can also accumulate on the ground during thunderstorms, and sometimes, a chunk of charge, called a leader, breaks free and moves toward the negative charge center in the clouds above. This creates a return path for lightning to strike the ground. Dwyer and his colleagues were using rockets to initiate this process. The rockets are tethered to spools of Kevlar-coated copper wire, which unspool as the rockets streak upward. The trailing copper wire creates a leader, providing the return path for lightning to come down and strike the rocket launcher. It was while observing such artificially created lightning that Dwyer’s team detected a downward gamma-ray flash. The finding suggested that if thunderstorms could beam TGFs downward, they could also beam them up to be seen from space. Maybe sprites weren’t the culprits.

Getting ready to fly ADELE (Airborne Detector for Energetic Lightning Emissions) on the NSF/NCAR Gulfstream V in 2009, Bryna Hazelton, then a UCSC graduate student, now a research scientist at the University of Washington, Seattle, and Forest Martinez-McKinney, principal electronics technician at the Santa Cruz Institute for Particle Physics (SCIPP), examine newly arrived parts for the terrestrial gamma-ray detector ADELE. Credit: David Smith.

Smith entered the picture about that time. As an assistant research physicist at UC Berkeley in the 1990s and early 2000s, he worked on a satellite called the Reuven Ramaty High Energy Solar Spectroscopic Imager (RHESSI), which launched in 2002 to study solar flares. It turns out that the exquisite germanium detectors used in RHESSI could also capture TGFs, and Smith, soon to be hired to the UCSC faculty in 2003, and colleagues began to study them. RHESSI allowed the investigators to measure the energy spectrum of many TGFs, revealing that they originated deep within the Earth’s atmosphere and not from sprites. “RHESSI was a real game changer,” said Dwyer.

Runaway electrons

Meanwhile, the theoretical physics that might explain TGFs was also coming into focus. The earliest work dates back to 1925 and Charles Thomson Rees Wilson, who won the 1927 Nobel Prize for inventing the cloud chamber. Wilson showed that electrons encountering the electric fields that exist in storm clouds can “run away” to near relativistic speeds (i.e., near the speed of light) if they gain energy from the field faster than they lose energy through interactions with air molecules. When these runaway electrons hit the nuclei of atoms in the atmosphere, the collisions produce gamma rays. “The process is called ‘bremsstrahlung’,” said Smith. “It’s German for ‘braking radiation.’”

The runaway “seed” electrons that sprint to relativistic speeds inside thunderstorms arise either from the particle showers produced by cosmic rays striking the Earth’s atmosphere, or nearer the ground from the decay of radioactive gases that constantly seep from the Earth’s surface. But these seeds can’t account for the great intensity of TGFs. “If I take every one of those electrons and accelerate it, and let it make gamma rays, it is still millions of times too weak to produce the TGFs we see,” said Smith.

More of the essential physics had been worked out by Russian theorists, led by Aleksandr Gurevich at the Lebedev Physics Institute in Moscow. In 1992, Gurevich and colleagues showed that accounting for a process called Møller scattering—in which electrons scatter off other electrons—results in a snowballing effect, creating more and more relativistic electrons, a phenomenon called the relativistic runaway electron avalanche (RREA).

As physicists began combining the RHESSI data with the theories to explain TGFs, it became clear that something was still amiss. Even if electrons accelerated to relativistic speeds inside thunderstorms and in turn created an avalanche of relativistic electrons, it would still not be enough to account for the brightness of a TGF, said Smith.

Then, in 2003, Dwyer suggested a possible solution: an avalanche of relativistic electrons somehow triggers other avalanches (see sidebar). “If an avalanche can make new avalanches, you immediately have a mechanism to get as bright as you need to get,” said Smith. But testing the theory required more experimentation. “I quickly decided that I didn’t want to wait for satellite data,” said Smith. “I wanted to go for a closer look.”

Stormy beginnings

So, Smith and his students built ADELE at UCSC. At the heart of the instrument are devices called scintillators, which emit light when stuck by particles such as gamma rays or neutrons. This light is amplified and converted into an electric current by a photomultiplier tube, providing information about the incident particles. ADELE was designed to see both faint and bright events by including multiple scintillators of varying sizes. In 2009, the first ADELE began flying aboard the modified Gulfstream V operated by the National Science Foundation (NSF) and the National Center for Atmospheric Research (NCAR). Although the effort captured that first TGF occurring about 10 kilometers away in a thunderstorm (during the flight that had Dwyer fearing for his life), there was another danger besides motion sickness (and crashing) associated with flying into storms looking for TGFs: pilots and passengers could be exposed to potentially lethal doses of gamma rays.

Gregory Bowers, then UCSC graduate student, now postdoctoral researcher at the Los Alamos National Laboratory in New Mexico, stands in front of NOAA's N42 Hurricane Hunter 'Miss Piggy' that flew into Hurricane Patricia with the terrestrial gamma-ray detector ADELE on board. Credit: Gregory Bowers.

This concern led Smith and his then graduate students, Gregory Bowers and Nicole Kelley, to build another iteration of ADELE, a more compact version that flew on NASA’s unmanned Global Hawk. Unfortunately, the work came to naught. NASA didn’t fly its drone anywhere near where you’d expect to see a TGF. “We didn’t know this when we made all the effort to get on board,” said Smith. “NASA was being very careful with a very valuable drone. They were unwilling to try to fly it into, or even over, a storm.”

NASA’s primary concern wasn’t the lightning, but the convective turbulence in thunderstorms. They didn’t want their Global Hawk flying into turbulence, said Bowers. Smith then turned to the only people who dare to tackle such risky conditions: the pilots who fly the Hurricane Hunters for the National Oceanic and Atmospheric Administration (NOAA).

On October 23, 2015, ADELE flew on a NOAA Hurricane Hunter through Hurricane Patricia, at the time the strongest hurricane recorded by the National Hurricane Center, as it approached landfall on the southwest coast of Mexico. Credit: Jeff Schmaltz, LANCE/EODIS Rapid Response.

It was a busy hurricane season in 2015 and ADELE spent a lot of time on a Hurricane Hunter (a Lockheed WP-3D Orion). The pilots even flew the plane through the most intense tropical cyclone ever recorded in the Western Hemisphere, Hurricane Patricia, which slammed into Mexico. “The plane took a straight cut all the way through the outer rain bands right through the eye wall, through the eye, back through the eye wall on the other side, and out,” said Smith. “It’s amazing they fly in those conditions. I would never ask someone to do that for my science, but since they were going to do it anyway, it was nice to be on board. We learned a lot.”

In the eye wall, a column of tall, violent storms that line the eye of the hurricane, ADELE detected positrons beaming down at them, produced by terrestrial gamma rays interacting with atomic nuclei. It was another first. “We were actually underneath a TGF that was pointed upward,” said Bowers.

Despite these successes, flying the instruments on planes poses challenges—not least because the instrument must be built both to handle intense vibrations and be fire-safe. Because of this, Smith and Bowers began to explore the possibility of placing their instruments near mountaintops.

From the ground up

Salvaging old ADELE equipment, Bowers built GODOT (for Gamma-ray Observations During Overhead Thunderstorms)—an instrument the team could simply install on the ground while thunderstorms raged above. The team first placed GODOT alongside the High-Altitude Water Cherenkov (HAWC) Gamma-Ray Observatory, 4100 meters high on a plateau in the shadow of the Sierra Negra volcano near Puebla, Mexico. It waited there patiently, taking in the lightning season that lasts for three long months.

Despite all the lightning, no TGFs were detected by GODOT, further confirming their rarity. GODOT did, however, observe thunderclouds glowing in gamma-rays, a phenomenon the team attributes to some acceleration of electrons and some avalanches, but not enough to produce lightning or TGFs. The gamma-ray glows last much longer than TGFs, but are a million times dimmer, said Smith.

Above left: A top-open view of the GODOT (Gamma-ray Observations During Overhead Thunderstorms) instrument, designed to study terrestrial gamma-ray flashes from the ground. The Lenovo notebook records data from five analog-to-digital converters on the left. The metal box in the bottom left corner is a GPS unit. Two small, shiny cylindrical detectors (each is a scintillator coupled to a photomultiplier tube) sit above two large detectors that are not visible. Above right: GODOT ready to be taken as checked-in baggage from San Francisco to Japan. Credit, both: Gregory Bowers.

Another attractive location for GODOT was suggested by a collaborator in Japan: the town of Uchinada, on the western coast of the island of Honshu. The region gets hit by winter thunderstorms that are very low to the ground. A wind turbine at the mouth of a lagoon there had been repeatedly struck by lightning. The Japanese dismantled the wind turbine, erected a lightning protection tower, and reinstalled the turbine next to it. For researchers studying lightning, this was perfect. “This was a good place to look for lightning strikes from these winter thunderstorms,” said Bowers. “We knew we had a big lightning rod that was getting struck.”

Masashi Kamogawa, a fellow lightning researcher at Tokyo Gakugei University, agreed to host Bowers, so GODOT traveled to Japan. “I took it over as checked luggage,” said Bowers, who now works as a postdoctoral researcher at the Los Alamos National Laboratory. “We deployed it, left it there for the winter, and we got really lucky.”

On December 3, 2015, GODOT detected a unique and unusual signal. “Because of my experience in solar physics, I knew what it was,” said Smith. “It’s what happens when a neutron gets absorbed by hydrogen.” GODOT had seen a theoretically predicted TGF signature. The bremsstrahlung gamma rays can knock out neutrons from air atoms. These neutrons scatter around until they are captured by hydrogen atoms in the scintillator, resulting in additional gamma rays with characteristic energies.

The neutron signal turns out to be a good proxy for a TGF, said Smith, especially because the TGF itself can be blindingly bright for terrestrial instruments. Smith’s team compared the observed TGF event to another lightning strike that happened about four hours earlier, but without an associated TGF. Based on their findings, the team posited that TGFs may not vary in intensity from weak to strong, but rather are “all-or-nothing” events, only occurring when the electric fields in the thunderclouds reach a very high threshold. This might also explain why TGFs are elusive. Not all lightning strikes may have the requisite electric fields to create TGFs. Smith and others have estimated that only about one in a thousand lightning strikes result in a TGF.

The wind turbine and an adjacent lightning tower in Uchinada, Japan, where GODOT saw photoneutrons, which are neutrons knocked out of air molecules by the gamma rays generated during a terrestrial gamma-ray flash. Credit: Gregory Bowers.

Smith now has the funding to build a super-GODOT, an instrument with a wider range of detectors, both smaller and larger, than in the current GODOT. He hopes to deploy this instrument, tentatively called THOR, for Terrestrial High-Energy Observations of Radiation, on mountaintops, in aircraft, next to lightning towers—anywhere he can. “Every time you think of a new way to look, you discover something new,” he said. “Every time you look in a new place, you discover something new.”

One such interesting new place could be detectors flown into thunderstorms in balloons, a project Smith is working on with John Sample, an assistant professor of physics at Montana State University. By going where pilots cannot, the balloon-based detectors should see the weak TGFs if they exist and may permit a closer inspection of the intense ones that could provide further insight into Smith’s “all-or-nothing” hypothesis. “When the balloons go up, we could be flattened or we’ll see nothing,” said Smith. “But maybe nature will surprise us, and that’ll be very exciting too.”