If you track the motion of individual stars in  the ultra-dense star cluster at the very center   of the Milky Way you’ll see that they swing  in sharp orbits around some vast but invisible   mass—that’s the Sagittarius A* supermassive  black hole.

These are perilous orbits,   and sometimes a star wanders just a little too  close to that lurking monster, leading to its   utter destruction in the spectacular phenomenon  known as a tidal disruption event.

Now we’ve never   seen a TDE in the Milky Way, but we’ve seen them  in other galaxies—and we now know how to spot   stellar destructions so extreme that they  reveal properties of the black hole itself.

What happens when a star gets a little too close  to a galaxy’s central supermassive black hole?

Well as   it approaches, the gravitational force on the side  of the star closer to the black hole gets stronger   than the pull on the far side.

This gravitational  differential results in a tidal force that,   from the star’s perspective, feels like it’s being  stretched in the direction of the black hole and   compressed in the perpendicular direction.

If the  star gets within a certain critical distance to   the black hole—the tidal radius—the tidal force  becomes greater than the force of gravity binding   the star together.

The star is smeared out in  a fluid-like, noodley mess in a process called   spaghettification.

Some of this material gets  ejected outwards, while more spirals rapidly into   the black hole, heating to incredible temperatures  before disappearing into the black hole.

These tidal disruption events or TDEs are so rare  that we’ve never seen one in our own Sag A* black   hole.

But we’ve seen around 100 candidate TDEs in  distant galaxies.

Even from Sag A*, and certainly   from other galaxies, a TDE appear to us as just a  flash of light from the galactic center.

But fear   not!

There is a way to get as close as we want to  one of these stellar trainwrecks: by recreating   them in supercomputer simulations.

By tuning those  simulations to our observations we're able to study matter moving in the strange, warped  spacetime around a black hole.

And recently, we’ve   even predicted a type of extreme tidal disruption  never before seen, but we may see very soon.

Throwing virtual stars at virtual black  holes is actually enormously challenging,   and we’ve only just pretty good at it.

To  understand how we do this and why it’s so hard,   let’s start with some of the gravitational  physics behind these tidal disruption events.

We’ve been able to simulate motion  under gravity for some time now.

The   Apollo program did the orbital  mechanics of its spacecraft on   “supercomputers” with the computing power  of a hand-held calculator… from the 90s.

These simulations used Isaac Newton’s theory  of gravity—calculating the gravitational force   between all pairs of objects, seeing  how those forces change velocities   and how those velocities change positions  over small time steps, and then repeating   over and over.

The calculations are quick  enough that modern computers can calculate   the interactions of millions of objects to  simulate entire galaxies, or even universes.

But Newtonian gravity is only an  approximation of how gravity really   works—an approximation valid when gravity is  weak.

That’s fine for most of the universe,   and even for mild tidal disruption events like  the one that may have torn apart a moon to form   Saturn’s rings.

But to properly describe a TDE  near a black hole we need the full theory—and   that’s Einstein’s general theory of relativity.

Rather than describing gravity as a force,   general relativity, or GR describes the  warping of space and time by massive objects,   and explains gravity as the effect of  that warping on the motion of objects.

Although general relativity is incredibly  well-tested and certainly “correct”,   it’s also notoriously complicated.

For one thing,  the nice symmetries that lead to clean circular   or elliptical orbits in Newtonian gravity no  longer apply in regions of space where GR becomes   important.

As a result, orbits around a black  hole can be twisted and tangle over themselves,   especially around rotating black holes.

Within a certain distance from the black hole,   there are no stable orbits whatsoever.

This  gravitational messiness has implications both   for a disrupted star and for the complexity of  simulating these disruptions on our computers.

For the star itself it means  that, within a certain distance,   all material must either fall into  the black hole or be flung back out.

We can’t have nice ring systems of stellar  material in the region close to a black hole And the complexity of these orbits makes  our simulations much more computationally   challenging.

In GR we describe the motion of  objects under gravity in terms of geodesics—paths   taken under pure gravity.

To calculate the motion  of an object in such a region, it’s not enough   to calculate the force at every point—you  need to plot out the geodesics—in fact,   you need to build a map of the warped spacetime.

It gets much harder if you have to consider the   star’s own effect on that spacetime curvature, but  fortunately for these TDEs the star’s mass is so   piddling compared to the black hole that the star can  be neglected.

But we have a different challenge.

Once the star is disrupted, it’s no longer  a single object; it’s a smeared-out fluid.

To simulate this we need relativistic  hydrodynamics.

We need to employ the equations   of fluid motion for material traveling at close  to the speed of light in a warped spacetime.

We   need to consider a variety of complex effects  like friction and the bath of intense radiation   produced by that superheater material which  can drive out the infalling material outwards.

Oh,   and we also need to simulate how that  light escapes the vicinity of the black   hole to make its way to us.

The same warped  spacetime that destroyed the star also twists   the paths of departing light rays in the  phenomenon known as gravitational lensing.

Until relatively recently, we’ve only been able  to simulate piecemeal parts of this phenomenon,   like the early time disruption of a star,  or the formation of a disk.

But these days,   modern state of the art codes can follow a  tidal disruption event from the incoming star,   all the way through its destruction to  the observable flashes of light we see.

Cool, so what do these simulated TDEs actually look like?

Well,   there are different types.

The most  important factor in determining what   happens is just how close the star gets to  the black hole—whether its orbit takes it   within certain critical distances.

The  most well-known such distance is the   event horizon—the surface of no escape.

For a non-rotating black hole this is at   the Schwarzschild radius—around 3 km  for black hole with the Sun’s mass.

If the star’s orbit would take it across that  surface, it is possible for it to drop directly   into the black hole without any part of it  escaping.

This is called a direct capture event.

Now,   depending on the size of the black hole, it may  be tidally disrupted before entering—that’s the   case if the tidal radius is larger than the event  horizon … true for any black hole smaller than a   few million solar masses.

Then the black  hole slurps up the entire spaghettified   stellar noodle.

Larger than that and the star  is still a star as it submerges.

These direct   capture events are typically not produce much of  an electromagnetic signature—the star just isn’t   ravaged enough to shine brightly before vanishing forever.

These should also be quite rare due to the   relatively small target of the black hole event  horizon, so we’ll leave these aside for now.

But the other end of the spectrum is the  partial TDE.

If the star passes close   enough to be significantly stretched but not  enough to be disrupted—so outside the star’s   tidal radius—then the star may shed significant  material from its outer layers, but enough remains   intact that it remains a star—fusion continues  in its core and it settles down upon escape.

But the real fireworks begin with the  full tidal disruption event—also   called the common TDE—because these  are the ones we most commonly observe   due to their brightness.

This is  when the star crosses the tidal   radius but its orbit doesn’t carry it  directly through the event horizon.

Let’s take a look at a relativistic hydrodynamic  simulation of a common TDE to see what should   happen.

This is a simulation of a star three  times the Sun’s mass passing within the tidal   radius of a 100,000 solar mass black hole.

That’s  where the tidal forces are now large enough that   they overcome the star’s self gravity.

Over  about a day, the star completes its flyby,   but it keeps expanding, producing a giant plume  of plasma- the core of the star loses pressure,   and fusion cuts off.

But that giant stream doesn’t  just keep going, not all of it anyway.

That tail   starts to fall back toward the black hole and then  begins to loop back in on itself.

The material   that hasn’t escaped ends up on an elliptical  orbit that can clearly be seen as the snake biting   its tail, and much of the material that didn’t  escape finds stable orbits around the black hole,   forming the start of an accretion disk.

An important feature of this orbit is this violent choke-point at the   closest point of the orbit.

This choke point is  called a “nozzle shock” caused by the tremendous   speeds that the stellar material attains.

It’s  at this point that friction and turbulence and   other forces in the stream of debris cause  potentially observable flares of photons.

A common TDE can be bright enough to outshine the  entire surrounding galax.

But it seems that that   universe is capable of doing even better—even  if we haven’t seen this yet.

Simulations have   revealed the possibility of something  that astrophysicists are calling an   extreme tidal disruption event—an eTDE.

This should happen when a star’s orbit   takes it within the tidal radius of a black  hole and very close to the event horizon,   say 2 to 3 Schwarzchild radii for a non-rotating  black hole.

The result is a vastly more energetic   phenomenon than the common TDE.

We don’t  yet have a confirmed observation of an eTDE,   but recent simulations are helping us to  understand exactly what to look for and how   to look.

Let’s take a look at the relativistic  hydrodynamic simulation of this event.

by Dr.   Taeho Ryu at the Max Planck Institute and  JILA at CU Boulder.

This was just published,   and is the first to follow the eTDE from a star  with realistic stellar structure to the hot   mess that it gets shredded into.

And by the way, the  earlier simulations we saw are also by Dr. Ryu.

This time we’re throwing a 1 solar mass star  at a million solar mass black hole, aiming to   skim just outside the event horizon.

It begins  somewhat similarly to the common TDE.

However,   within seconds we see direct capture of some  of the star, with an emphasis on “some”.

The   rest of it is dragged out into a tail that  whips around the black hole multiple times,   leaving a spiral trail that was not  seen at with the elliptical orbit of the   fallback that made the common TDE.

Very soon an  entirely new phase emerges – within tens of   minutes the windings of the tail merge together,  leaving behind a circularized ring made up of the   last bits that narrowly escaped their demise.

These remaining bits now sit in an expanding,   circular cloud.

Though stable for quite some time,  the innermost portions of the ring nevertheless   begin to fall back toward the central black  hole.

As the inside of the ring fills in the   previously empty cavity, the black hole accretes  the material and a long lived accretion disk is   formed until eventually all of the surrounding  material is ultimately either consumed or ejected.

As the matter gets accreted by the black hole,  much of that gravitational potential energy   is converted to radiation energy, making a  bright burst of light.

In the extreme TDE,   this “luminosity” increases very rapidly until it reaches  what is called the Eddington luminosity.

This is   a sort of upper limit on the radiative  output for a given black hole mass,   where the outward radiative force directly  balances the inward force of gravity;   any more radiation, and the material is blown outward  by the radiative winds.

Being at this   Eddington limit puts the system in a state of  temporary equilibrium, with outward radiation   balancing gravity.

But after  a while, the outgoing radiation drops off and the   black hole’s gravity dominates again, enabling  it to finish gobbling up the stellar material.

These extreme TDEs should be rare because the  star has to make a very close approach.

But   it turns out that for larger supermassive black  holes, these are the only tidal disruption events   that can happen.

That’s because, as the mass of a  black hole grows, both the event horizon and tidal   radius grow.

But the event horizon grows faster,  so that at a few tens of millions of solar masses   the tidal radius is very near or even inside the  event horizon.

Approaching these masses, if the   star gets close enough to be disrupted at all,  it’s also close enough to undergo an extreme TDE.

So we now know what tidal disruption events  should look like if you’re actually close to   one.

But the real value of these simulations  is to predict what they look like from Earth   so we can actually test our ideas.

And  potentially identify our first extreme TDE.

So what do TDEs look like to us?

Because they’re  too far away to resolve the spatial details,   we really just see a point of light growing in  the center of a distant galaxy where we expect   the supermassive black hole to lie.

That  point brightens as the star disrupts and   spreads out and then fades away as the stellar  material either falls into the black hole or   disperses.

The timescale depends on the black  hole mass.

For a million solar mass black hole,   the rise time is around a month and then,  typically, it will have faded away within   a year.

For a common TDE, most of that light  is emitted at visible wavelengths.

Flashes of   light that grow over about a month and then fade  over several months.

We can also see radio flares   as ejected material slams into the diffuse stuff  in the surrounding galaxy, producing shockwaves.

Extreme TDEs are expected to be quite a bit  more extreme, as one would hope.

They should   rise in brightness much more quickly.

An eTDE  around a million solar mass black hole should   brighten over only a few hours instead of  a month and then maintain a more consistent   brightness than a common TDE for up to  a year before finally fading out.

This   is something that the new study revealed because  the simulations found this temporary equilibrium   as outgoing radiation helps resist gravity.

Those simulations also predicted super compact   accretion with a temperature peaking at around a  million Kelvin.

This results in a broad spectrum   of light with most photons being X-rays, and  is around 10 times brighter than a common TDE.

We find common TDEs in our visible-light  surveys of the sky.

They don’t happen very often—every   several thousand years in a given galaxy, and  most commonly around smaller supermassive black   holes.

We’ve never seen one directly in the Milky  Way, but there are observed remnants such as radio   lobes in the galactic center that would indicate  that one occurred a few thousand years ago.

ETDEs should be rarer than common TDEs because  they require a closer approach tend to happen   around larger black holes, which are rarer than  smaller ones.

But Dr. Ryu and team estimate one   per 15,000 years for a 20 million solar mass  black hole.

To spot such a thing we need to   monitor a lot of galaxies, which is something  new large field surveys can do—but we also   need to do it at X-ray wavelengths, which is  much harder than visible light wavelengths.

Fortunately, we already have an instrument  that may be up to the task.

eROSITA is an X-ray   detector on Spektr-RG space observatory and it's  created an X-ray survey of the entire sky—but unfortunately operation was suspended in 2022 before  covering the X-ray energies needed to   detect an eTDEs.

But eROSITA will soon  resume its operations and finish its   survey in the next couple of years.

And then  our first eTDEs will hopefully shine out.

And this will be a rare chance to witness the  motion of matter in the region very close to   a black hole event horizon.

Comparing what  we see with our simulations will let us test   our models of these violent events—how  stars get ripped apart by black holes,   and also our understanding of how gravity really  works in the most extreme regions of spacetime.