Revealing The Milky Way’s Compact Remnant Underworld

When we look up at the Milky Way, arching from horizon to horizon over our skies, what we are seeing are the brightest stars in our Galaxy. Just using our eyes, we see about 2,000 or so of them from dark skies, of the 200 billion that reside in our spiral disc home. With a camera or special equipment - many more of these stars become visible, allowing us to take a look further into and around our Galaxy. 

In any case, we notice that most of the stars we observe reside in a central band - known as the Galactic Plane, and the density of stars reduces as we move above and below this central, narrow region of brightness. That’s because our Galaxy is a spiral galaxy, and we are seeing it from within, and edge on. We note in its centre, a bigger, brighter bulge exists, and everything rotates around this region. If we had a bird’s eye view from above, we would see the large, swirling spiral arms as they wound, and wrapped around this central region. 

But not all astrophysical objects are big enough or bright enough to give off light that we can see here on Earth. In fact - some objects remain almost invisible or actually are invisible because, well, all light that falls upon them never can escape. Literally. 

Their very names conjure up all sorts of sci-fi fantasies and possibilities in our minds. There’s even a relatively small Hollywood genre dedicated to using them in films. These are the stories of neutron stars and black holes, which are in fact, very real and observable astrophysical objects. Scientists tend to refer to them as compact remnants - because they are massive, they are tiny, and they are the remnant by-product of when progenitor massive stars die in spectacular cosmic fireworks, known as supernova explosions. In a nutshell, they are the zombie remains of the dead stars in our Galaxy - the Galactic Underworld.

Since these compact remnants are born from progenitor stars that (for the most part) live in the central, Galactic Plane - one would expect that the highest density of these objects would also be found in this region. But a new study, one that models these compact remnants from their progenitor phase, through their supernovae events and into their compact remnant phase has found some interesting new results, which have been published in the journal, Monthly Notices of the Royal Astronomical Society. 

It’s the first map of its kind - a chart of the corpses of once massive suns that have since collapsed into black holes and neutron stars – revealing a graveyard that stretches three times the height of the Milky Way. The study also outlines how almost a third of these objects have been flung out from the galaxy altogether.

“These compact remnants of dead stars show a fundamentally different distribution and structure to the visible galaxy,” said David Sweeney, a PhD student at the Sydney Institute for Astronomy of the University of Sydney, and lead author of the paper.

“The ‘height’ of the galactic underworld is over three times larger than the Milky Way itself,” he added. “And an amazing 30% of them have been completely ejected from the galaxy.”

The new study, which features astrophysicists from Sydney University, Monash University and the Australian Research Council Centre of Excellence for Gravitational Wave Discovery (OzGrav) found that this underworld does not conform to the visible galaxy we see with our eyes and instruments, and in fact, the spatial distribution of these remnants extends to a much larger halo of bodies, eerily and silently orbiting around, with the rest of the Galaxy. 

The reason? Well, those powerful supernovae events - can sometimes give the newborn remnant a big velocity kick, that launches the object away from its original location in any random direction. Additionally, the historical changes our Galaxy has experienced (through mergers with other smaller galaxies, and the evolution of stellar orbits over time) can also change where stars live and die (or become zombies, as they do in these cases). 

The researchers found, that their models even produced a different spatial distribution for neutron stars, vs. that of black holes, related to the smaller kick velocities that black holes receive during these violent explosions. 

To conceptualise how these compact remnants come to be, one can follow their formation models from the progenitor phase through to the time they become either a neutron star or black hole - with the evolutionary transition passing via the supernova stage. By studying the light from these bright explosions, astronomers have been able to develop a categorisation of supernovae that features three main branches, and further sub-branching. 

The first, known as a Type Ia supernova involves another compact remnant (albeit, less massive), known as a white dwarf. White dwarfs are formed as the remnant cores of stars that are like our Sun (or up to eight times the solar mass), where instead of dying in a massive explosion, they gently puff out their outer layers, leaving behind the hot cinder core remnant. 

In these scenarios, the white dwarf star either merges with another white dwarf (triggering the explosion) or more commonly, the white dwarf steals materials from its own binary companion, building up its mass. When this occurs, enough material builds up on the start to trigger a new thermonuclear explosion, obliterating the star in this first type of supernova category. 

The second type, known as Type II supernovae is what produces neutron stars and black holes. In this model, the progenitor star is so massive - usually, about 10+ times that of our Sun, that as the outer layers blow off into space during the explosion, the core of the star collapses inwards under the pull of gravity, squeezing electrons and protons together to form neutrons (hence, a neutron star). In some cases (when the mass of the star is 25+ times that of the Sun), the collapsing process doesn’t halt, crushing the core down into a singularity, and thus forming black holes.

There’s also an in-between for these two scenarios -  when the mass of the progenitor is ~8 - 10 times that of the Sun (i.e. too big to just puff off its outer layers and form a white dwarf, and too small to undergo the full core-collapse model), creating another type of supernovae category. The electron-capture supernovae. 

Electron-capture supernovae cases feature a core that is made of lots of oxygen, neon and magnesium atoms (when stars are big enough, they fuse hydrogen and helium into heavier elements in a process known as nucleosynthesis). As the cores of these objects get denser over time through the continual fusion of elements, the magnesium and neon atoms absorb their own electrons. These electrons are important as they provide supportive, outward-stabilising pressure against gravity for the core, and so, as they disappear, this pressure becomes weaker - causing the core to collapse in on itself. 

Each of these three categories, and the different objects that are produced from them, is an important factor in the modelling that is presented in this latest paper - because the different types of supernovae events can provide a different kind of natal kick on the compact remnant, thus affecting the overall spatial distribution of these objects around the Galaxy. 

“The hardest problem I had to solve in hunting down their true distribution was to account for the ‘kicks’ they receive in the violent moments of their creation,” said Sweeney. “Supernova explosions are asymmetric, and the remnants are ejected at high speed – up to millions of kilometres per hour – and, even worse, this happens in an unknown and random direction for every object.”

Although billions of compact remnants must have been formed since the galaxy was young, these exotic carcasses were flung out into the darkness of interstellar space by the supernova that created them, hence slipped beyond the sight and knowledge of astronomers – until now. By carefully recreating the full lifecycle of the more ancient dead stars, the researchers have constructed the first detailed map showing where their corpses lie.

“One of the problems for finding these ancient objects is that, until now, we had no idea where to look,” said Sydney Institute for Astronomy’s Professor Peter Tuthill, a co-author of the paper. “The oldest neutron stars and black holes were created when the galaxy was younger and shaped differently, and then subjected to complex changes spanning billions of years – so it has been a major task to model all this to find them.”

Newly-formed neutron stars and black holes conform to today’s galaxy, so astronomers know where to look. But the oldest neutron stars and black holes are like ghosts still haunting a house demolished long ago, so they are harder to find.

“It was like trying to find the mythical elephant’s graveyard,” said Professor Tuthill, referring to a place where, according to legend, old elephants go to die alone, far from their group. “The bones of these rare massive stars had to be out there, but they seemed to shroud themselves in mystery.”

To map out the spatial distribution of compact remnants, Dr Sweeney and colleagues developed modelling that considered the evolution of the Milky Way over time. At first, the team mapped out the stellar population by modelling the position, velocity, metalicity, and mass of stars that would form the progenitors. These populations were also then considered from the perspective and locations of four distinct regions of the Milky Way - the thin disc, the thick disc, the larger stellar halo, and the triaxial bulge. 

This stellar distribution was then aged in their models, allowing the very largest mass objects to eventually undergo their supernovae events, and evolve into a compact remnant. Where appropriate, the stars that underwent a core-collapse supernova in the modelling, also experienced a natal kick from the event, sending the newly formed remnant in a random direction - which over time, created the new spatial distribution of these compact remnants. 

The modelling also considered the speed distribution of the natal kicks the remnants receive, classifying them into two categories - a lower velocity portion, which was attributed to the electron-capture supernovae, and a high-velocity portion, attributed to the core-collapse supernovae model. Additionally, as black holes have a higher mass relative to neutron stars, their natal kicks were several times smaller. 

These values were then added to each of the remnants that the model produced, applying them isotropically. At far enough distances from the Galactic Centre (~32,600 light-years), this allowed sufficient escape velocity from the Milky Way’s gravitational well, sending these exotic objects into intergalactic space. 

As a result of all these factors, the Galactic Underworld becomes a much more spatially distributed, and diffuse structure, relative to the visible components of the Galaxy. In the model, the scale height of the Milky Way is approx. 1,100 light-years, but according to the spatial distribution of compact remnants that have received a variety of different kick velocities, this puffs out to 4,110 light-years at the 50th percentile point (this point was selected due to the increasing uncertainty at larger radii). 

Interestingly, the probability density varied for the two different types of compact remnants, with black hole remnants more likely to be located near the galactic centre thanks to their larger masses having a greater resisting effect on the kick velocity, whereas neutron star remnants becoming much more spread out at greater distances. 

The models also showed details of the Galactic mass loss, induced by the stronger kick velocities imprinted on the compact remnants, giving them enough force to overcome the escape velocity of the Galaxy and leave the Milky Way altogether. Since its creation 12.5 billion years ago, the models report that the Milky Way has lost approx. 0.4% of the stellar mass in this manner. 

“It’s a little like in snooker,” said Sweeney. “If you know which direction the ball is hit, and how hard, then you can work out where it will end up. But in space, the objects and speeds are just vastly bigger. Plus, the table’s not flat, so the stellar remnants go on complex orbits threading through the galaxy. Finally, unlike a snooker table, there is no friction – so they never slow down. Almost all the remnants ever formed are still out there, sliding like ghosts through interstellar space.”

The intricate models they built – together with University of Sydney Research Fellow Dr Sanjib Sharma and Dr Ryosuke Hirai of Monash University – encoded where the stars were born, where they met their fiery end and their eventual dispersal as the galaxy evolved.

The final outcome is a distribution map of the Milky Way’s stellar necropolis.

“It was a bit of a shock,” said Dr Sharma. “I work every day with images of the visible galaxy we know today, and my expectation was that the galactic underworld would be subtly different, but similar in broad strokes. I was not expecting such a radical change in form.”

The research team aged the models through time - incorporating the ongoing evolution of the Milky Way itself, as well as the continual motion of the compact remnants, now imparted by their natal kicks - finding that the spatial distribution of these compact remnants could be used to determine the localised density of these objects around certain locations within the Galaxy. 

For example, by thinking about our location (well, our Sun’s location) in the Galaxy, and the spatial distribution density of these compact remnants - can we work out how close we might be to the nearest neutron star or black hole? 

It turns out that (according to the modelling reported in the paper) the probable nearest neutron star should be about 62 light-years away, and the nearest black hole, about 69 light-years away. Astronomically speaking, that’s relatively close to us. 

Observationally, the nearest neutron star (or pulsar - which is a form of neutron star that has radio beams emanating from its magnetic poles) is about 424 light-years away, and the nearest black hole is reported to be 1,500 light-years away. 

The discrepancy can be explained by a number of factors. Firstly, the modelling might have input values that include high uncertainties, or unmodelled parameters that could affect these outcomes. Additionally, neutron stars that are not pulsars are very hard to observe - they are tiny objects that don’t generate any fusion light, so we rely on detecting them in x-ray bands. And for the neutron stars that are pulsars - their radio beams need to be pointing in our direction for us to notice them, otherwise, we won’t know they are there as pulsars. 

And as for black holes - these objects are invisible unless they have a binary companion which we can observe orbiting them (usually, giving off x-ray light, or having the binary move around an invisible object from our perspective). There is also an additional method of detecting isolated, singular black holes which involves surveying the sky to search for gravitationally-lensed background stars - but to date, these numbers don’t account for the expected population of black holes.

Whilst sci-fi stories might intricately detail the destruction of the Solar System by wondering massive remnants like black holes and neutron stars that stray into our neighbourhood - probability says that this is an unlikely scenario - even when considering the spatial distribution of these remnants that this latest research has found. After all, the Solar System has existed for 4.5 billion years without undergoing such events. 

But objects of this nature need not get too close to have an effect on us here on Earth. A good example is that of the cousins of pulsars, with objects known as magnetars. These objects are (as the name implies) extremely magnetic, young neutron stars - with magnetic fields on the order of trillions of times as powerful as your regular fridge magnet. 

Since they’re relatively young, they are also fairly wild and unstable - often unleashing powerful blasts of high-energy radiation and particles into space. These can traverse great distances, until they eventually come into contact with the upper layers of Earth’s atmosphere, triggering a cascade of particles that results in an increased abundance of Carbon-14 (¹⁴C) on the surface. Interestingly enough, this surge of ¹⁴C is absorbed by Earth’s trees which hold a global record of these variations over time. 

In this latest modelling, approx. 34 magnetars would have formed over the last 10,000 years, with about 50% of these within 33,000 light-years of Earth. The expected nearest magnetar (if considering the spatial density distribution) should, according to this research, be about 14,000 light-years away. 

What the scientists found is that this distribution and population of magnetars is too low to have had the measured effects of ¹⁴C in Earth’s trees, and so, another source for the isotope appearing in tree ring samples should be considered.