The Crowd and the Cosmos: Adventures in the Zooniverse

by Lintott, Chris

  eventually disperse, their stars lost to intergalactic space or once

  more part of the main galaxy. There’s not much hope in finding

  definitive proof of a merger just from shape.*

  Nor is the colour of a galaxy likely to be much good. The colour

  tells you what’s happening right now, and the great burst of star

  formation that accompanies a really dramatic merger most likely

  lasts only a few hundred million years. Galaxies which have sur-

  vived mergers will not easily be distinguished from those which

  have evolved through less dramatic means, which makes it diffi-

  cult to try and understand the effect of mergers on galaxies. Such

  events look spectacular, but we’d like to understand how signifi-

  cant they really are.

  For example, it’s possible that most stars form during the dra-

  matic starbursts that follow a collision. It’s also possible that, like a firework display, these collisions are spectacular but ultimately

  have little long-term effect; in this scenario, most stars form

  because of other processes, and galaxies would look much the

  same in a universe where mergers were much less likely.

  How can we distinguish between these possibilities? What

  we really need is the chance to do a direct experiment. I’d love

  to assemble a vast, intergalactic laboratory (along with, of

  course, a few billion years’ worth of funding). In it, I would

  assemble two populations of galaxies. The two populations

  would begin the experiment in identical states, but in one tank

  gravity would allow the galaxies to merge, just as happens in

  * You might just be able to do something by looking at the outskirts of the galaxy and hoping to detect the faint leftover scattered debris, which can, in some cases, persist for billions of years. New instruments, using special lenses first developed for photographing high-speed motorsport of all things, are useful, but such observations are time-consuming and as yet have only produced data for a small number of systems.

  108 Into the ZoonIverse

  the real Universe. In the other, an ever-watchful PhD student

  could be tasked with intervening to keep the galaxies apart

  from each other. (I have no idea how they would do this, but as

  we’re imagining a laboratory in which there are two tanks at

  least a few hundred billion light years across, I think we’re

  allowed a little magic.) After ten billion years or so, we could

  then compare the resulting mix of galaxies in each tank, and see

  what effect merging really produced.

  Of course, such an experiment is impossible for several very

  good reasons. We can attempt something similar in a supercom-

  puter—and people do—but it is a tricky problem. Keeping track

  of the galaxy-scale details of the merger and the small-scale pro-

  cesses of star formation that determine how the galaxies look at

  the same time is a difficult computational challenge, and requires

  shortcuts to be made.

  As a result, while some might be satisfied with the results of

  simulations, I prefer to look out into the Universe for my experi-

  ments. Somehow we need to assemble a tankful of galaxies that

  have managed to avoid merging. These will be rare, but they are

  out there, and they reveal their presence via their shape—

  exactly the kind of thing that the Galaxy Zoo volunteers can

  help with.

  I often remember Patrick Moore’s claim that the Milky Way

  resembled nothing more than two fried eggs, ‘clapped back to

  back’, with a thin disc surrounding a central bulge, represented

  by the eggs’ yolks. This gives you a good way to think about spiral

  galaxies,* and central bulges are so common that at least one

  astronomer has proposed that both elliptical and spiral galaxies

  should really all be seen as nothing more than bulges, some of

  which happen to have grown discs.

  * It is, however, a lousy way to serve eggs.

  Into the ZoonIverse 109

  One way of producing bulges is via a galaxy merger, as the dis-

  ruption kicks stars out of the main disc and into the bulge. Even

  a small collision should add substantially to a bulge, which is

  why it was surprising when, among the hundreds of thousands

  of galaxies searched by Galaxy Zoo a small number of unusual

  objects started to appear in the classifications. These special sys-

  tems have no visible bulge, and are therefore guaranteed to be

  merger-free. We can put them in our second tank, and compare

  them to galaxies with more normal histories.

  Nature and our volunteers had provided a way of doing the

  experiment I imagined above. In charge of the experiment was

  Brooke Simmons, a Californian who ended up working with me

  in Oxford before winning a prestigious Einstein Fellowship from

  NASA. She took the latter back to California, craving decent

  Mexican food and more sunlight than Oxford could provide (our

  department provided a special lamp for her to cope with the

  winter which replicated the exact spectrum of natural sunlight; it

  was apparently cruel otherwise to keep an American in English

  conditions*).

  Brooke’s speciality is understanding how the supermassive

  black holes at the centre of galaxies grow, and so our first experi-

  ment was to try and see if mergers contribute significantly to

  their growth. Most people expected that they would; even galax-

  ies which have experienced many mergers have a single, massive

  black hole at the centre, not a whole cluster of little ones, so

  merging must happen.

  Further evidence that galaxies’ central black holes grow

  through mergers comes from the best-known bulgeless galaxy,

  NGC 4395, which had been studied a decade earlier and was

  * It’s amazing to see how many people who scoffed at this provision now gather around its light in the winter months, myself included.

  110 Into the ZoonIverse

  found to have a puny black hole, more than ten times less mas-

  sive than we would otherwise expect for a galaxy of its size.

  No mergers, no black hole growth, it seemed. But that result is

  only from one system. Now, with many more merger-free sys-

  tems to play with, we could do the experiment properly.

  That means measuring the mass of each galaxy, and then the

  mass of the central black hole. Estimating the mass of a galaxy

  isn’t too difficult. We know how a star’s colour and brightness

  depend on its mass, and a galaxy is just an assemblage of stars.

  (Yes, there’s dark matter and dust and other things, but we’re

  concerned with the stellar mass here; for most large galaxies the

  ratio of total mass to stellar mass is pretty constant.) The bright-

  ness of the galaxy tells you how many stars are there, and the

  colour tells you what sort of stars they are. Know those two

  things, and you can get a pretty good handle on the mass of the

  galaxy.

  The black hole mass is trickier. Measuring the mass of some-

  thing that’s invisible, and which is in any case tiny on galactic

  scales, isn’t easy, but we can get there via an indirect route. What

  we do is look right at the centre of the galaxy, expecting to see the bri
ght glow of hot material as it falls down into the black hole.

  Such accretion activity is most easily seen in the x-ray region of

  the spectrum, but in the visible what you see is a star-like point of light. For each galaxy, Brooke took all the light that could possibly belong to such a source, and assumed that it belonged to the

  material falling into the black hole.

  That gives us a guess at the rate at which the black hole is grow-

  ing. It’s a start, but we want to know how massive the black hole

  itself is. Luckily, how massive a black hole is turns out to be tied

  to the rate at which it consumes material. The idea that black

  holes do anything but voraciously devour everything around them

  might be surprising; it certainly belies the fearsome reputation

  Into the ZoonIverse 111

  they have in science fiction, where they’re always ‘lurking’ at the

  centre of a galaxy, rather than just hanging out in space minding

  their own business.*

  This has always seemed unfair to me, and I’m almost tempted

  to found the Society for the Promotion of Friendly Black Holes.

  Our promotional material will make much of the careful man-

  ners these exotic beasts exhibit. It turns out there is a maximum

  rate at which, under normal circumstances, black holes will con-

  sume fuel. This slightly surprising result is due to the dramatic

  effect on material that, falling into the black hole has. It heats up and shines brightly. The presence of this radiation creates a pressure, pushing outwards and preventing more material from fall-

  ing in; the whole thing becomes a self-regulating system with a

  maximum rate of accretion known as the Eddington limit.

  Using this piece of information, we can convert the minimum

  observed rate at which material is falling into the black hole into

  an estimated mass, and finally make the comparison we want to.

  I’ve probably given too much detail here, though there are

  many more gory specifics I could have included (we have, for

  example, only really calculated an estimate of the maximum

  mass). I do, however, want to make the point that this is the meat

  and potatoes of modern astrophysical research. First, we care-

  fully assembled a sample of interesting, distinct galaxies. Once

  they were found, we selected a comparison sample, and then we

  measured their properties, being careful to make sure we had as

  much understanding as possible of what the likely errors were

  * A star in a distant orbit around a black hole is in no more danger of being sucked into its cavernous maw than the Earth is of falling into the Sun. It’s true that if you get too close, you’ll fall inevitably into the black hole itself, but that’s hardly the black hole’s fault. Star trekking travellers of the future need not worry about being consumed by a black hole, they need simply to study astronaviga-tion properly.

  112 Into the ZoonIverse

  and how they’d affect the result. It’s painstaking work; each of

  the images of each of the galaxies had to be calibrated by hand,

  and the whole effort probably took Brooke something like six

  months.

  That careful work paid off, though, and these weird galaxies

  now told us something about the Universe and its history that we

  didn’t know before. It turns out that the bulgeless galaxies have

  black holes which are pretty much the mass one would expect

  from a normal galaxy. In other words, despite never having ever

  had a significant merger, living instead lives of splendid isolation, they manage to grow large supermassive black holes.

  This research isn’t finished. I’ve included in this book a brand

  new image of one of our bulgeless systems from the Hubble Space

  Telescope (Figure 17), which is allowing us to measure its properties with new accuracy, and thanks to help from some friends

  we’ve been able to find merger-free galaxies in one of the big

  Figure 17 A bulgeless spiral galaxy, discovered by Galaxy Zoo volunteers and observed by the Hubble Space Telescope. Hubble’s sharp resolution allows us to peer into the heart of the galaxy.

  Into the ZoonIverse 113

  supercomputer simulations and to compare the properties of

  such systems in this artificial Universe to those in the real one.

  If those simulations are reliable, they confirm a hard truth that

  astronomers have faced for years. I often end my talks with a

  bleak version of the far future of our Universe, destined (we

  think) to become an nearly empty void, a vast sea of space

  expanding forever into yet more nothingness.* What seems to

  pack a greater emotional punch, though, is this: the Universe is

  already past its best.

  More stars are dying each year than are being born and galax-

  ies are shutting down in a process that’s been going on for billions of years, since the youthful, exuberant peak of star formation

  that took place shortly† after the Big Bang.

  What’s causing this cosmic shutdown? There are lots of ideas

  around, starting with the simple idea that we’re out of fuel. Stars

  need cold gas to form, and in some galaxies at least the reservoir

  seems to have been exhausted. Maybe galaxies normally rely on

  a flow of material from outside to keep the stellar factory going,

  and that’s disrupted by close encounters with other galaxies.

  Maybe it’s falling into a large cluster of galaxies that triggers that process. Maybe a collision between two galaxies causes a burst of

  star formation so dramatic that it uses up all of the available gas.

  Or maybe if a large galaxy like the Milky Way consumes too

  many smaller systems then the effect is the same. Or maybe we

  need to look back at the central black holes; the complex physics

  and twisted magnetic fields that exist around them can fire jets of

  material moving nearly at the speed of light out into the galaxy,

  heating or expelling the gas it encounters. Maybe if you form too

  * I do like to send an audience home happy.

  † Shortly here means a few billion years; astronomical timescales are hard to get used to.

  114 Into the ZoonIverse

  many massive stars the resulting burst of violent supernovae

  which marks their end can similarly heat the surrounding gas.

  And on and on and on. Understanding exactly what’s going on is

  a difficult task, but it might be important in understanding our own

  Milky Way. Galaxies which are still vigorously making stars are, on

  average, bright blue, lit up by their new creations. Those where star formation has ceased—we would say ‘quenched’—are red. The

  Milky Way, at least according to work by a bunch of Australian

  researchers, is neither red nor blue, but green, which means that it

  seems to be undergoing this transition right now. If they’re right,

  then we happen to catch our galaxy at an unusual point in its his-

  tory and must look outward to understand what’s happening.

  Trying to distinguish between so many possible causes makes

  carrying out a clever experiment like the one we managed with

  the bulgeless galaxies near impossible. Instead, we use the sheer

  scale of the Sloan survey and the classifications provided by

  Galaxy Zoo, pile them all together, and look at which galaxies

  have which properties.

  This task fell to my first P
hD student,* Becky Smethurst, now

  an independent research fellow at Christ Church here in Oxford.

  Before she arrived, I was terrified of taking on the responsibility

  of supervising a PhD student. Your PhD supervisor sets the direc-

  tion and tone of your research, so while the student is ultimately

  responsible it seemed very easy for me to completely ruin some-

  one’s career. Luckily (apart from an unfortunate incident which

  resulted in American Express bombarding me with advertising

  for Taylor Swift concerts, the less said about which the better)

  Becky and I got on well and she proved more than smart enough

  to deal with my blunders.

  * Oxford insists on awarding not a PhD, but a DPhil, thus leaving everyone involved explaining their degree for the rest of their career.

  Into the ZoonIverse 115

  Her task wasn’t easy, and Becky constructed a sophisticated

  apparatus of modern statistics and computer-based analysis to

  look at this problem of quenching. The result of all that effort?

  Well, to almost no one’s surprise it turned out to be complicated.

  Different galaxies seem to go through the transition from blue to

  red in different ways. Some, mostly elliptical galaxies, seem to

  have shut down their star formation rapidly, perhaps the result of

  a spectacular merger. Others, including most spiral systems and

  the Milky Way itself, quench more slowly. Some systems that

  sustain growing black holes show clear signs of recent and dra-

  matic quenching, so it’s clear that they can play a role too.

  Where a galaxy lives also makes a difference, with those that

  find themselves in more crowded environments undergoing a

  more dramatic shutdown than their relatively isolated cousins.

  In other words, we shouldn’t carry round a picture of a galaxy as

  an isolated system (the ‘island universe’ that a galaxy was once

  thought to be) but we should rather think of them as interacting

  with their surroundings. Galaxies at the centre of a large cluster—

  the nearby Virgo Cluster, say, which contains more than a thou-

  sand galaxies and which weighs in at more than a million billion

  solar masses—have a very different life from those living in more

  rarefied parts of the Universe.

  These results—Becky’s work on quenching and that led by