
Lightning Strikes Twice, and Scientists Still Don’t Know What’s Going On
A pair of discoveries only reveals how clueless we are about how lightning starts and the “weird stuff” it does next.
You may have heard lightning never strikes twice. Totally wrong. The Empire State Building is said to be struck 23 times a year, on average. Here’s a video of it being walloped three times in less than a minute:
You might also have the impression that scientists understand how lightning works. Wrong again.
“Despite over 250 years of research, how lightning begins is still a mystery,” says Ningyu Liu, a professor of physics at the University of New Hampshire Space Science Center, who makes his living by studying lightning. Seems he’s got pretty good job security.
Liu was involved in one of two recent studies that revealed previously unknown things about lightning, which just tends to get more mysterious the more we know.
Brian Hare, a postdoc researcher in the KVI-CART institute at Groningen University in The Netherlands, studies atmospheric physics and electrodynamics. His team just discovered previously unseen “lightning needles” embedded in a thunderstorm. I asked him to characterize the state of lightning research.
“We don’t know how lightning gets started, grows through the cloud, connects to the ground, or why it does all the weird stuff that it does afterwards,” Hare said.
I’ll explain both studies. But first, let’s back up.
What We Know
Serious lightning research goes back a bit more than 250 years—267, to be precise, when Ben Franklin’s kite wasn’t struck by lightning. It was on a June afternoon in 1752, with a storm descending on Philadelphia, and what actually happened was the kite “picked up the ambient electrical charge from the storm,” according to the Franklin Institute. Had it been actually struck by a bolt, Franklin would likely have been toast.
Another thing Franklin didn’t do was discover electricity. But he did show a connection between electricity and lightning, setting the stage for a century or so of roughly zero progress on the understanding of electrical storms.

In the late 1800s, scientists got a better look at lightning, taking advantage of photography to slow things down. Slow-motion video more recently revealed even more about the complex phenomenon.
Here’s what we’re pretty sure about today: Lightning is not hurled by Zeus, and it’s not created by Thor striking his hammer on an anvil.
Instead, the genesis of lightning is in roiling thunderstorm clouds, where violent updrafts and downdrafts face off. Here’s the process, step-by-step, based on my reading of several expert descriptions and corrective input from both Hare and Liu (who, by the way, were not involved in each others’ work):
- Ice particles collide amid the turbulence, and electrons are stripped away, creating charges of different polarity that then get separated.
- Smaller particles tend to gain a positive charge, the experts think, and larger particles tend to take on negative charges. Updrafts carry more of the lighter particles higher into the clouds, creating a difference in charge. As you know, opposites attract.
- When the charge becomes large enough — millions of volts — the resistance of the air between breaks down (air is a pretty good insulator).
- The breakdown of air, called dielectric breakdown, creates a superheated gas called plasma, which is a good conductor of electricity.
- Two plasma channels, called leaders, head off in opposite directions based on their charge. Positive leaders tend to grow smoothly, while negative leaders (called stepped leaders) jump in steps, lurching up to a few dozen feet in microseconds, resting for a few microseconds, then jumping forward again, searching like a crazed mouse in a maze for the path of least resistance.
- Leaders seek areas or objects of opposite charge to connect with. And then a bunch of things happen that we call lightning.
- Some bolts reach down to Earth. Often (but not always) this involves a negative leader heading down, then a positive channel shooting up from the surface to greet it. When they connect, a return stroke shoots back into the clouds, creating the visible flash. It all happens in a little as one-millionth of a second.

However…
There are several weak points in the scientific understanding of these steps. Take Step 3, for example.
That the air “breaks down” is “technically true,” Hare says, but “we still don’t actually know how this happens.” The basics of dielectric breakdown here on Earth are understood, but not up in the clouds. “When lightning scientists say ‘undergo dielectric breakdown’ this is often code for ‘we don’t actually know how this works,’” Hare told me.
In an email, Liu delved into some detail on the known and the unknown of Step 3, based on lab experiments and electrical-discharge theory:
“The electric field has to exceed a threshold value in order to trigger electrical breakdown of air to form an electrical discharge,” he said. “Decades of efforts to measure thunderstorm electric field, such as sending sensors into thunderclouds by launching balloons, have found the maximum thunderstorm electric field is well below that threshold value. It’s possible that the balloon missed the high electric field region inside thunderstorms prior to lightning occurrence. A theory on the table is that cloud ice particles can significantly enhance the electric field in their vicinity to trigger electrical breakdown, but it’s not clear at present how the breakdown leads to the formation of the first lightning channel.”
A Positive Discovery
As we dig further into the details, things get even more confusing, for us and for the experts.
For one thing, scientists had been thinking lightning started by a process called “fast positive breakdown” of air, by which, in one example, a plasma channel opened as a positive charge moves from the top of a cloud to a negative charge in the middle.
But in their new study, Liu and his colleagues found just the opposite: fast negative breakdown, with the charge moving upward.
“These findings indicate that lightning creation within a cloud might be more bidirectional than we originally thought,” said Julia Tilles, a Ph.D. candidate at the Space Science Center and lead author of the study, which was detailed April 9, 2019 in the journal Nature Communications.
For you and me, that might not seem like a bolt out of the blue. But for lightning researchers, it’s pretty enlightening. Yet it still leaves much unanswered.
“We don’t actually know the physical mechanism behind [fast breakdown],” Hare points out. “We have a working hypothesis, but this hypothesis is poorly tested. Furthermore, if fast breakdown initiates lightning, how does the fast breakdown get started?”
Good question!
Liu adds that fast breakdown is but one way for lightning to get started. “The latest research indicates that not all lightning starts as fast breakdown, and lightning can also begin without identifiable fast-breakdown signatures.”
And what are those other ways? Good question!
Lightning Strikes Again
Eight days after Liu and Tilles published their study, Hare’s team added more understanding (and mystery) to the mix.
Hare and his colleagues used the Dutch radio telescope LOFAR (Low Frequency Array) to peer into thunderstorms. They spotted needle-shaped structures in lightning that nobody had ever seen before. These lightning needles, they wrote, “might help to explain” why lightning does not always discharge just once in a strike, “as was thought for a long time, but can strike several times within seconds.”
The lightning needles “appear to drain charge from the leader, and are probably the reason why positive leaders disconnect from negative ones, and why cloud-to-ground lightning connects to the ground multiple times,” the researchers explained in the journal Nature on April 17, 2019.
“This finding is in sharp contrast to the present picture, in which the charge flows along plasma channels directly from one part of the cloud to another, or to the ground,” said Hare’s colleague, Olaf Scholten, a physics professor at the University of Groningen.
When you see phrases like “appear to” and “probably the reason” in scientific communications, you know the mysteries aren’t solved. More research is needed, as they say.
Meanwhile, the slo-mo video these scientists created is pretty cool. During a strike that spans one-tenth of a second, yellow indicates the plasma channels, and one of the newfound needle structures is shown in red:
Lightning Strikes Twice, Thrice and Quadrice
The fact that multiple lightning bolts sometimes hit the surface during a brief, single “flash event” has been known for a long time. And in fact the same spot can be hit twice during a single event. You might see it as a flickering, since it all happens in the blink of an eye.
Back in 1997, University of Arizona researchers videotaped 558 cloud-to-ground lightning strikes during 386 flash events, meaning each flash event produced, on average, 1.4 ground strikes. A handful of those events produced three or four ground strikes.

The researchers, William Valine and Philip Krider, found that when there was a second stroke in an event, it usually struck in a different yet nearby spot (as pictured above). But third and fourth strokes typically struck in the same spot as the second stroke.
Whether it’s a four-stroke event or multiple strikes during the year, there are at least two known reasons why a certain location might be struck multiple times.
- The location is tall — a mountain, tree or building — and therefore closer to the clouds. Air is a pretty good insulator, so the closer a charged object is to an oppositely charged part of a cloud, the easier it is for nature to make the connection.
- The location has a lot of salt, moisture, metal or some other good conductor of electricity.

Then again, lightning can strike a spot twice in any given time period just as a statistical fluke. There are, after all, more than 30 million cloud-to-ground lightning strikes each year in the United States alone.

Struck by This
All this got me wondering about a very basic question, which surely must have a good answer: Why does some lightning stay in the clouds, spidering its way across the sky, and other times the gods hurl a flaming bolt to the ground. Here’s what we know about that:
“Details of why a discharge stays within a cloud or comes to ground are not understood.”
That’s a statement on a lightning explainer page written by the folks at NASA and the University of Alabama in Huntsville. If they don’t know, I sure as heck can’t explain it.
But I do know this: Thanks to a variety of factors, from education and awareness to more lightning-proof construction and fewer people working outside, lightning deaths in the US have declined in recent decades, from an average of 93 per year between 1959 and 1994 to 27 per year from 2009 to 2018.

Assuming you don’t take undue risks, such as running out into a thunderstorm as I do to delight in nature’s fury, you are more likely to be killed by sunstroke, a bee sting or a dog attack than by lightning.