The Universe is never short on the bizarre and the unknown. From black holes to exotic planets, scientists have plenty to ponder.

Lately, though, one puzzle has been especially baffling: mysterious flashes in the sky called fast radio bursts.

These bright bursts of radio waves shine for only a few milliseconds. In that instant, they release about a million times more energy than the Sun.

Since their discovery in 2007, astronomers have found fewer than 20 – all coming from outside our galaxy, scattered randomly across the sky. But telescopes typically observe small patches of space at a time. If astronomers extrapolate to the entire sky, they estimate as many as 10,000 bursts could be flashing every day.

And no one knows what they are.

Astronomers have plenty of ideas, of course, some of which are quite exotic: colliding stellar corpses called neutron stars, exploding black holes, snapping cosmic strings, and even aliens.

They're guaranteed to revolutionise our understanding of the Universe

"Right now, there are more theories for the nature of the bursts than the actual bursts themselves," says Duncan Lorimer, an astronomer at the University of West Virginia in the US and leader of the team that discovered the first burst. "It's kind of a theorist's paradise at the moment."

Even if fast radio bursts turn out to be more mundane, they still could be important. These radio signals are like lasers that shoot across the Universe, encountering magnetic fields, plasma, and other cosmic stuff along the way. The bursts thus capture information about intergalactic space, serving as a one-of-a-kind tool for probing the Universe.

"They're guaranteed to revolutionise our understanding of the Universe by making very precise measurements," says Ue-Li Pen, an astrophysicist at the University of Toronto.

Before that happens, though, scientists need a better understanding of what fast radio bursts are. Fortunately, in just the last few months, astronomers have made tantalising progress.

The first thing that struck Lorimer about the signal was its brightness.

I was so excited I couldn't sleep

He and a team of astronomers were sifting through archival data taken with the Parkes radio telescope in Australia. They were searching for pulses of radio waves – from, for instance, rapidly-spinning neutron stars called pulsars. These city-sized stars, as dense as an atomic nucleus, can spin more than a thousand times per second. As they rotate, they swing a beam of radiation around like a lighthouse, producing radio signals that appear as pulsating blips.

But this one signal was weird. "The signal was so bright that it saturated the electronics in the telescope," Lorimer says. "It's very unusual for a radio source to do that." The radio source glowed for about five milliseconds and then dropped off.

"I remember when I saw the first graph of what the burst looked like," says Matthew Bailes, an astronomer at Swinburne University in Australia who was on Lorimer's team. "I was so excited I couldn't sleep."

For about five years, this signal, dubbed the Lorimer Burst, remained an anomaly.

They are from outside our galaxy, maybe as far as billions of light years away

Some thought it was instrumental interference. In fact, a study published in 2015 found that microwave ovens at the Parkes telescope facility produce a comparable signal.

But starting in 2012, astronomers detected several more bursts using other telescopes, confirming that the signals really came from space.

And not just anywhere in space. They are from outside our galaxy, maybe as far as billions of light years away – according to initial measurements of a phenomenon called the "dispersion effect".

When radio waves travel through the Universe, they interact with electrons in plasma along the way. Those interactions cause a delay, the length of which depends on the radio signal's frequency. Higher-frequency radio waves arrive a hair faster than low-frequency ones. By measuring this delay, astronomers can calculate how much plasma the signals had to go through, which gives an estimate for the distance travelled.

Finding something that's a million times brighter than anything you've ever seen is kind of exciting

Radio signals from other galaxies are not new. It is just that none have been so strong.

For example, a black-hole-powered object called a quasar produces prodigious amounts of energy, including radio waves. But quasars in other galaxies are so far away that their signals are weak. They would be easily swamped by even the signal from a mobile phone placed on the Moon, Bailes says.

Fast radio bursts, on the other hand, stand out. "Finding something that's a million times brighter than anything you've ever seen is kind of exciting," he says.

It is especially exciting because fast radio bursts could be a sign of strange and new physics.

One of the more provocative ideas for what they could be involves "cosmic strings": defects in the fabric of space and time that stretch across the Universe.

If there were tiny black holes forming in the early Universe, they would be evaporating now

Some of these strings might be superconducting and carry electrical current. According to the hypothesis, suggested in 2014, the strings might occasionally snap, exploding in a burst of electromagnetic radiation.

Or, the origin of the bursts could be exploding black holes, Pen says.

A black hole's gravity is so strong that it prevents even light from escaping, but in the 1970s Stephen Hawking realised that a black hole could radiate energy if it evaporated away into nothingness. If there were tiny black holes forming in the early Universe, they would be evaporating now – possibly exploding in a burst of radio waves.

In February 2016, astronomers made what seemed like a breakthrough.

A team led by Evan Keane of the Square Kilometre Array Organisation at Jodrell Bank in the UK analysed one particular burst, detected in April 2015.

The observations were consistent with at least one dramatic scenario: a massive collision between two neutron stars

This burst, according to their analysis, originated in a galaxy of old stars six billion light years away. For the first time, researchers had identified a burst's host galaxy – a crucial discovery.

"The host galaxy is the killer," says Bailes, who was a member of Keane's team. "It's the clincher. Once you get a host galaxy, you know how far away they are." That then lets you precisely measure the energy of the burst, and you can start narrowing down theories for what they could be.

In this case, the observations were consistent with at least one dramatic scenario: a massive collision between two neutron stars in orbit around each other. An answer to these mysterious bursts seemed near. "I got very excited by the results," Lorimer says.

But within weeks, Edo Berger and Peter Williams of Harvard University had cast doubt on this story.

Keane's conclusions had relied on the discovery of what appeared to be a fading radio signal that followed the burst. It was because the source of this signal came from the galaxy six billion light years away that the researchers thought the burst did so too.

Berger and Williams, however, argued that the afterglow had nothing to do with the burst.

Neutron stars do not collide often enough – by orders of magnitude – to explain all the fast radio bursts

When they looked again at the supposed afterglow, using the Very Large Array telescope in the US, they discovered that the afterglow was not an afterglow at all. It was an independent phenomenon caused by the dimming and brightening of the galaxy, thanks to a supermassive black hole gobbling gas and dust at the centre.

In other words, instead of being the source of the fast radio burst, this flickering galaxy just happened to align in front of or behind the burst in the telescope's field of view. And if the burst did not originate in the galaxy, perhaps it was not caused by colliding neutron stars after all.

There is another problem with the colliding neutron star scenario. "The rate of these fast radio bursts are much higher than the rates expected of neutron star mergers," says Maxim Lyutikov of Purdue University in the US. Neutron stars do not collide often enough – by orders of magnitude – to explain all the fast radio bursts.

Soon, yet another discovery would make this explanation even less likely.

In March 2016, a team of astronomers reported an astounding result. They used the Arecibo Observatory in Puerto Rico to study a burst that had first been detected in 2014, and found that the burst lit up again and again, for a total of 11 times over 16 days.

"That was the single biggest piece of information since the first one was discovered," Pen says. "That eliminated the vast majority of models that have been proposed."

They could be, as Lyutikov calls them, "pulsars on steroids"

Until this finding, all fast radio bursts were isolated one-off occurrences. That meant they could be generated during self-destructive, cataclysmic events that could only occur once, such as the explosion of a black hole or the collision between two neutron stars.

But that cannot account for the bursts if they can sometimes be produced over and over again in quick succession. Whatever it is that produces the bursts has to survive the process and potentially generate a fresh burst.

That narrows things down considerably.

One possibility Lyutikov has explored is that the bursts come from young pulsars, neutron stars that can spin as fast as once every millisecond. Over time, pulsars lose rotational energy and slow down. Some of that energy could go into producing bursts of radio-wave radiation.

It is unclear how exactly a pulsar could produce fast radio bursts, but they have been known to release short bursts of radiation. One is the Crab pulsar, which, at less than 1,000 years old, is relatively young and one of the most powerful pulsars known.

Magnetars are plentiful in the Universe

Younger pulsars spin faster and are more energetic. They could be, as Lyutikov calls them, "pulsars on steroids". Although the Crab is not energetic enough to produce fast radio bursts anymore, it might have been in the years immediately after its birth.

Alternatively, the power source of fast radio bursts might not be a neutron star's rotation. Instead its strong magnetic field, which can be a thousand trillion times more powerful than Earth's, could be responsible.

Such highly magnetic neutron stars, called magnetars, can produce bursts through a process similar to the one that generates solar flares. As the magnetar rotates, the magnetic fields in its corona, the wispy outermost layer of its atmosphere, rearrange themselves and become unstable. Eventually, the field lines snap like a whip, unleashing a torrent of energy that accelerates charged particles, which then emit radio bursts.

"Magnetars are plentiful in the Universe," Bailes says. "They have this kind of sporadic thing that they do, which could perhaps explain the fast radio bursts."

These neutron-star ideas are less exotic, involving relatively well-known phenomena, which might make them more likely explanations. "The ones I hear about seriously or discuss with people for any length of time involve a neutron star," Bailes says.

It may turn out to be a window into new physics

However, he admits there is a potential bias in the field. Many astronomers who are studying fast radio bursts also study neutron stars, so they lean on their own expertise.

Indeed, unconventional ideas remain. For instance, some researchers have suggested that the bursts occur when pulsars crash into asteroids.

And maybe there are multiple answers, each explaining different kinds of fast radio bursts. Maybe some bursts repeat while others do not, keeping the door open for colliding neutron stars or other cataclysmic scenarios.

"It may turn out to be something simple," Lyutikov says. "But indeed it may turn out to be a window into new physics, and into new astrophysical events and phenomena."

Regardless of what the bursts are, they could prove to be a boon for cosmology.

For example, they can be used to measure the amount of stuff in the Universe.

During their cosmic journey, the radio waves encounter intergalactic plasma, which delays the signal depending on its frequency – the same dispersion effect astronomers used to estimate their distance. But this delay can also reveal how many electrons lie between our galaxy and the burst.

In the past, it was kind of a fringe activity that people did in their spare time

"Encoded in the radio waves is a census of the Universe's electrons," Bailes says. That gives an estimate of the amount of normal matter in the cosmos, which scientists can then use to test models for how the entire Universe came to be.

According to Pen, what makes the bursts most promising is that they act as cosmic lasers. The bursts zip through space in one direction at high intensities, providing laser-like precision to make measurements down to small scales.

"It's the most precise measurement you can make of structure far away, because you suddenly have a precise probe along the line of sight," he says.

For example, Pen explains, the bursts can reveal the structure of plasma and magnetic fields near the source of the burst. Plasma can make the bursts twinkle, like how Earth's atmosphere causes stars to twinkle. Measuring this twinkling allows astronomers to probe the plasma down to scales of a few hundred kilometres.

Because of this potential – and especially because of the mystery – interest in fast radio bursts has swelled in the last couple of years.

"The field is poised to take off," Lorimer says. "In the past, it was kind of a fringe activity that people did in their spare time."

Some researchers have suggested that the bursts occur when pulsars crash into asteroids

Now, astronomers are ramping up their efforts. They are hunting for more bursts, and monitoring known ones to see if they erupt again. Researchers are enlisting multiple telescopes around the world to pinpoint exactly where in the Universe they're coming from.

For instance, radio telescopes like the array called Chime, for Canadian Hydrogen Intensity Mapping Experiment, will be able to survey large swathes of the sky and discover hundreds of bursts in the next few years.

With more data comes better understanding. And perhaps fast radio bursts will be a mystery no longer.

Join over five million BBC Earth fans by liking us on Facebook, or follow us on Twitter and Instagram.

If you liked this story, sign up for the weekly features newsletter called "If You Only Read 6 Things This Week". A handpicked selection of stories from BBC Future, Earth, Culture, Capital, Travel and Autos, delivered to your inbox every Friday.