WHAT REALLY GROWS AT THE TIP OF LIGHTNING
A lightning flash is built from hot, electrically conducting channels, above 5,000 kelvin, called leaders. They grow from both ends over many kilometres. Physicists long agreed on one rule: the positively charged end advances smoothly, while the negatively charged end always advances in jumps, or steps. The common explanation is that negative leaders push away the free electrons they need to grow.
Radio telescopes can follow lightning because its growing tips emit millions of short radio pulses. In lightning high in the clouds, the steps of negative leaders appear as a series of fans, every few milliseconds, up to 300 metres apart, each made of several filaments. Those filaments have always been read as a crown of streamers: small, self-propagating discharges that leave no conducting channel behind them. But no one had seen them clearly enough to be sure.
A radio telescope pointed at a storm
M. A. A. Lourens, B. M. Hare and colleagues at ASTRON, the Netherlands Institute for Radio Astronomy, and the University of Groningen, with researchers in the United States and China, went back to a lightning flash recorded on 14 August 2020 by LOFAR. Two hundred antennas, spread over 36 stations across the north of the Netherlands up to 100 kilometres apart, caught its radio pulses between 30 and 80 megahertz.
They re-imaged it with ATRI-D, an imaging method the group recently developed that adds up the signals of hundreds of antennas to locate each faint source in three dimensions. In an earlier test, it pinpointed weak radio sources on an aircraft flying through high clouds to within about 10 centimetres. Here the resolution is coarser, about a metre, because so many sources go off at almost the same time.
Filaments, not sparks
Zooming into a single fan, the team found it does not grow as one block. It is an intricate network of thin filaments, at most about 1.1 metres wide, and possibly much thinner, since this is the limit of the instrument. Each filament grows for 100 to 500 metres at an almost constant speed of about 100 kilometres per second: in one image, filaments cover about 250 metres in 3 milliseconds. They branch often, without changing speed, and emit radio pulses continuously from their tips.
They also behave as if they knew where the others were. New filaments start from older ones but bend around earlier tracks, never merging and never passing twice through the same volume.
Strong radio bursts launch new networks of filaments. These bursts do not happen at the tips of older filaments, but at points along their length, sometimes after milliseconds. Charge must be building up at those points, so a current must be flowing through the filaments: they conduct electricity, but only moderately, or the charge would not gather in separate spots.
Why they cannot be streamers
The authors list four reasons. A single streamer moving at constant speed cannot emit so many radio pulses. Streamers would appear around a leader tip at about a billion per second, far more than the handful of filaments seen. A streamer only keeps a constant speed in a very specific electric field, and speeds up or slows down quickly otherwise. And a streamer loses its conductivity within tens of centimetres behind its head.
A tree of branching streamers is ruled out too: it would spread out at a wide angle. These filaments stay just as thin over about 500 metres. Something must keep them focused, as it does for leaders: heat concentrates at the centre of the tip, which makes it more conductive, which concentrates the heat further.
So the filaments are small negative leaders that advance smoothly, without stepping. Seen with a much blurrier imager, a series of fans gives the impression of discrete steps.
Rethinking the steps
The team writes that this is “a critical change” in how lightning plasma is understood. Negative and positive leaders still differ: negative ones are loud in radio, positive ones quiet. But both can advance without steps, kept narrow by their own heating.
Many questions remain. Why do the filaments gather into repeated fans? Do negative leaders closer to the ground, below about 7 kilometres, work the same way, or are high-altitude fans shaped by local pockets of charge? The faint optical glows seen ahead of lightning leaders, usually attributed to streamers, might themselves be leaders. To find out, the authors call for simultaneous high-speed cameras and new models of how these filaments form.
