麻豆传媒

JOSEPH DWYER had never been a big fan of flying. Things took a distinct turn for the worse, though, when his plane accidentally flew into a thunderstorm. 鈥淚 was certain that we were all about to die,鈥� he says.

But it comes with the territory. When Dwyer, a lightning researcher at the University of New Hampshire, began a study that would take him near the edges of thunderclouds in a Gulfstream V jet, he knew it would mean getting up close and personal.

The first two flights were smooth 鈥� fun, even. Then he and his colleagues headed up again. 鈥淲e went inside a cloud and I couldn鈥檛 see anything any more. Then all hell broke loose,鈥� says Dwyer. The plane started to violently rock back and forth, before plummeting several thousand feet. 鈥淚 completely lost my sense of what was up and down, and so to me it felt like we were barrel-rolling toward the ground.鈥�

Luckily, they landed safely. But they had encountered one of the central problems of lightning research 鈥� getting a peek into the stormy nurseries where it arises can be not only difficult, but also dangerous. As a result, we still know comparatively little about how lightning forms. This is surprising given that the phenomenon is so widespread 鈥� roughly a hundred lightning flashes happen somewhere around the globe every second. Slowly, though, researchers like Dwyer are shedding light on the electrical processes responsible for this spectacle. What they are finding is making us realise we may have had lightning back to front all along.

Dwyer follows a long line of physicists who have attempted to understand these mysterious bursts of charge. One early pioneer was Benjamin Franklin. In 1752, so the story goes, he attached a metal key to a kite and flew it into a thunderstorm. As he subsequently went to touch it, sparks flew 鈥� much like the ones that pop when you shuffle across a shaggy carpet and reach for a metal doorknob. The observation was enough to make him think of lightning as simply a scaled-up version of the same phenomenon.

When you walk across a carpet, the friction between your feet and the floor scrapes negatively charged electrons off the carpet that run up through your body and give you an overall negative charge. This build-up of charge may seem trivial, but over short distances the electric field it generates can grow surprisingly powerful.

As your finger edges towards the neutral doorknob, the electric field increases in strength until free electrons in the air can no longer resist it. If in some region the field reaches the critical value of 3 million volts per metre, known as the breakdown field, the electrons start to accelerate along those powerful field lines. Their motion knocks other electrons loose from nearby atoms, increasing the local field and making more of the air conductive. This creates a bridge that can sustain a current running from your hand to the doorknob, resulting in that noisy, somewhat painful electrical discharge.

Friction is the root of the scaled-up electric field in thunderstorms, too, although its origin is different: hail falling through clusters of ice crystals (see 鈥淗ow lightning forms鈥�). The friction rubs electrons from the crystals, and a positive charge builds towards the top of the cloud where the crystals collect, buffeted by the storm鈥檚 strong updrafts. The now negatively charged hail, meanwhile, continues falling to the bottom of the cloud. This charge separation creates an electric field between the top and bottom of the cloud, just like the one between you and the doorknob.

鈥淭he most promising idea at present is that lightning comes from outer space鈥�

But when the field somewhere inside a thundercloud grows strong enough to cause breakdown, the results are rather more spectacular than a spark off a doorknob. Electrons carve ionised channels through the air 鈥� each about as wide as a finger 鈥� hunting for the nearest positive charge. During a familiar cloud-to-ground lightning flash, the negative charge finds this at Earth鈥檚 surface. But the most common type of lightning discharge is an intra-cloud flash, where it runs up to the positive region at the top of the cloud. Either way, once one of these so-called lightning leaders reaches a region of opposite charge, electric current explodes between the two points, creating lightning flashes five times hotter than the surface of the sun.

There鈥檚 just one problem with this picture: although we have been sending balloons and aircraft into lightning-charged thunderstorms since the 1950s, we haven鈥檛 observed that 3 million volts per metre electric field needed to cause breakdown. Instead, the field is typically 10 times weaker than the ones we generate on deep-pile carpets.

That suggested that lightning operates in a different way to a conventional electric spark. Unlikely as it sounds, perhaps the most promising alternative is that lightning gets a boost from outer space.

Every second, billions of high-energy particles crash into our atmosphere. For the most part, these cosmic rays pass unnoticed, but if they collide with a free electron in a thunderstorm, they give it a speed boost. This newly energised electron ionises large numbers of air molecules, triggering an avalanche of high-energy electrons. The resulting sudden accumulation of charge briefly intensifies the electric field, albeit on a tiny scale. Although the details are still murky, the added effect of this localised field may be enough to spark lightning in an effect known as a 鈥渞unaway breakdown鈥�, without the background electric field being anywhere near 3 million volts per metre.

Support for this theory came in 1991, shortly after NASA launched the Compton Gamma Ray Observatory into orbit. Its mission was to search for gamma rays, the most powerful form of radiation in the universe, usually created when stars explode. So it was a surprise when the observatory detected these high-energy photons coming from thunderstorms in Earth鈥檚 atmosphere as well as from objects beyond the Milky Way.

Physicists were quick to connect the dots. As cosmic-ray-accelerated electrons in our atmosphere gather speed, they not only produce further high-energy electrons, but also a side of high-energy photons. The detection of such gamma rays was therefore a sign that runaway breakdown was at work.

Yet as tantalising as this link appears to be, it could be a coincidence: we have no direct evidence that this process is what鈥檚 going on. The trouble is that when it comes to what鈥檚 happening within thunderstorms, we鈥檙e more or less flying blind. For a start, lightning is a very localised process, says , North Carolina. Even in Florida, the lightning capital of the US, lightning strikes a given square kilometre maybe 10 times a year. To boot, lightning鈥檚 low predictability and high speed makes the birth of bolts hard to spot. 鈥淚n terms of space, you鈥檝e got a box that might be a few hundred metres on a side,鈥� Cummer says. 鈥淚n terms of time, you鈥檝e got maybe one millisecond.鈥� Then there鈥檚 the fact lightning comes in many guises (see 鈥�Spark of recognition鈥�).

Another challenge with identifying the electric fields within thunderstorms is that instruments such as balloons or aircraft used to penetrate storm clouds fundamentally alter the cloud鈥檚 environment. Balloons, for example, are often struck by lightning, drastically lowering any nearby electric field. This raises questions about much of the electric field data collected in clouds over the past few decades.

So if we want to see what鈥檚 going on in the roiling black heart of a thunderstorm, we need to find ways of peering in from a distance. Luckily, there is a way. The fluctuating electric fields produced by lightning results in large amounts of radio noise being emitted, causing a crackle similar to that heard on old-fashioned analogue radios. In the mid-1990s, physicist and his colleagues realised they could use GPS receivers to precisely map that radio noise, and therefore the lightning flashes. Today, Rison鈥檚 Lightning Mapping Array stretches across 16 stations in the Magdalena mountains of nearby central New Mexico. It can take 3D images of lightning within a thundercloud, but there鈥檚 one caveat: its time resolution is not so good.

鈥淟ightning emits a lot of radio noise, crackling like an old-fashioned receiver鈥�

Rison and his team are working on that. Last year, they developed an interferometer to detect the radio waves, fitted with a high-speed camera capable of capturing lightning flashes at more than 180 million frames per second. With this device working alongside the Lightning Mapping Array, the video is accompanied by a full, three-dimensional map providing the most accurate images yet of lightning bolts in action.

When the researchers went into the field to test it out, they hoped to get detailed pictures of runaway breakdown. The results, however, weren鈥檛 what they expected. Instead, an alternative mechanism seemed to be at play, a mighty spark hidden deep within the clouds that could raise the electric field without the need for extraterrestrial assistance. But rather than beginning in a negative region within the cloud and running to a nearby positive region, as expected, it did the opposite.

鈥淲e spent weeks arguing among ourselves,鈥� Rison says. 鈥淚s this real? Do we understand how our instrument is working? Did we flip a sign somewhere in our measurement?鈥�

Not exactly. What Rison and his team realised was that the perceived flow of positive charge amounted to a rush of electrons going in the opposite direction 鈥� much like the apparent accumulation of sand left behind by a receding tide.

The culprit, in their view, might well be a tiny ice crystal somewhere in the cloud with a negative charge on one side and a positive charge on the other. If the positive charge grows strong enough to tear off electrons in the air in front of it, it can ionise the air and create another positive charge just beyond the crystal. They called this weird observation fast positive breakdown. This ionised patch of air then behaves like the crystal鈥檚 new tip, a process that repeats itself for as long as the electric field is strong enough to sustain it.

鈥淲hat you get is this little wiggly ribbon of ionised air that starts growing past the end of the ice crystal,鈥� says , Santa Cruz. 鈥淎s that positive 鈥榮treamer鈥� grows, it鈥檚 like a vacuum cleaner sucking up negative charge.鈥� And it sends that charge back towards the ice crystal, like a swimmer pushing water behind her. Once the ice crystal accumulates enough added charge, it can spark a lightning leader, the forerunner to a strike.

These streamers have been observed in nature before 鈥� but never as a precursor to lightning initiation. And although Rison鈥檚 team thinks the fast positive breakdown they observed may have been caused by one of these, they can鈥檛 be sure. Getting a clearer picture of the microscopic physics involved will require tools much more precise than radio wave detectors.

of Dwyer鈥檚 at the University of New Hampshire, has been using computer simulations to try to resolve the matter. Already, he has successfully replicated positive streamers emanating from ice crystals, but the 5 centimetres he has so far been able to model don鈥檛 capture the full picture: in nature, such channels have been observed stretching for 100 metres.

For some researchers in the field, the evidence for fast positive breakdown is a crucial step. 鈥淚t鈥檚 kind of like when you鈥檙e doing a big jigsaw puzzle,鈥� says Dwyer. 鈥淭he interferometer data is that first edge or corner that lets you start the rest.鈥�

But that doesn鈥檛 mean runaway breakdown doesn鈥檛 happen. The background field strengths measured within thunderclouds perfectly match the lower fields required for runaway breakdown, and many doubt that match is a mere coincidence. Smith thinks these runaway electrons act as a thermostat, generating small discharges that bring the electric field鈥檚 strength back down again whenever it gets too high. . 鈥淚t鈥檚 kind of like a slow leak in your tyre,鈥� he says. 鈥淵ou can never exceed a certain point because if you do then the leak increases.鈥�

But Dwyer thinks it could still play a role in lightning鈥檚 formation, along with a number of other processes. 鈥淭he real answer is probably complicated,鈥� he says. And as with any recipe, it might require multiple ingredients.

So he will once again focus on those high-energy photons. Although the gamma glows emanating from thunderstorms aren鈥檛 proof that cosmic rays cause lightning, Dwyer finds it promising that they tend to occur right before lightning begins to form.

So, each summer for the next three years, Dwyer and Liu will visit Florida, armed with a grant from the National Science Foundation, weather balloons equipped with gamma ray detectors, and powerful cameras.

This upcoming work, together with follow-up observations from Rison鈥檚 instruments, should bring scientists closer than ever to cracking the mystery. 鈥淚 think in 10 years, people will be pretty confident they understand how lightning is initiated,鈥� says , a physicist at the University of Florida. And if Dwyer is lucky, without ever needing to fly into a thunderstorm again.

This article appeared in print under the headline 鈥淔lash of inspiration鈥�