we see in the data. This happens on a regular schedule, as the
* Following negotiation with a journal editor who insisted that it was policy that all acronyms be spelt out, it was agreed that WTF stands for ‘Where’s the Flux?’, neatly referring to the central mystery presented by the star.
Is It AlIens? 215
ringing repeats with each swing of the smaller star past the larger.
This is interesting, and such behavior valuable in trying to under-
stand stellar interiors, but something more complicated is going
on with the WTF star.
Daryll found a clue in the data that was available on the web.
Most of the objects in Kepler’s target list had been extensively
studied in preparation for the mission. In many cases, the
properties of any planets found can only be pinned down if the
stars themselves are understood, and effort has already gone
into excluding stars whose inherent variability would have
hidden likely planets. As a result of all this work, Daryll could
tell that this star was brighter in the infrared than stars of its
type usually are.
Unexpected brightness in the infrared usually means that
there is a disc of dust, leftover from planet formation, in orbit
around a star. As the dust absorbs light from the star, it reradiates mostly infrared radiation; hiding a star behind dust therefore
usually results in the system appearing dimmer than normal
when viewed in visible light, but brighter in the infrared. This is
one reason that astronomers studying star formation normally
turn to longer wavelengths, hoping to be able to observe stars
still embedded in their embryonic cocoons. The infrared excess
suggested that the neighbourhood of the WTF star was a dusty
place, and Daryll realized that this might be the key to explaining
its bizarre behaviour.
He suggested that there really was a planet in orbit around the
star, but that the planet was itself surrounded by a dust disc. That
seems sensible enough. Just as planets form from the disc of left-
over dust and gas which surrounds newborn stars, so a newly
formed planet might be surrounded by a disc of leftover material
from which moons might form. Our own Moon probably had a
more violent origin, coalescing from the debris of a collision
216 Is It AlIens?
between the proto-Earth and a Mars-sized object, and the two
moons of Mars seem to be no more than captured asteroids.
Large planets such as Jupiter seem to have formed their systems
of large moons more directly, though, and a large planet with a
dense disc of material passing in front of a star would certainly
block plenty of the star’s light. If the geometry of the passage
changed each time, you might be able to explain the observed
differences each time the WTF star dimmed or flickered.
The attractive thing about this proposition is that it could
explain an almost arbitrary pattern. The disc might have a gap at
its centre, between the planet and its inner edge, just as there’s a
gap between Saturn’s rings and the planet itself. The team flying
the Cassini spacecraft, which took several plunges between rings and planet at the end of its life, called it the ‘Big Empty’, so it
should be no surprise that light would shine through such a gap,
adding to the complexity of the behaviour during an observed
‘blink’. Saturn’s rings also have gaps within them, shaped by
interactions between ring particles and the myriad tiny moons
which surround and shepherd them. Add the same sort of thing
to the WTF system, and you might have a chance of explaining
what’s going on.
By the time speculation had reached this state, with Daryll and
others drawing possible models for the rings, the Planet Hunters
team themselves became involved, most notably Tabby Boyajian,
then a Yale postdoc, who led the professional end of the effort
to solve the mystery of this most unusual star.* The dust disc
explanation felt wrong from the start; every piece of information
we had in the Kepler Input Catalogue pointed to KIC8462852
* Because of Tabby’s efforts in leading the work on the star, it’s sometimes known not as the WTF star but as Boyajian’s star, which I rather like. ‘Tabby’s star’ also gained currency, but I like the authoritative and official sound of using her surname.
Is It AlIens? 217
being a perfectly ordinary, stable, and middle-aged star, while dust
discs are almost exclusively the property of younger objects. Worse
than that, the fact that two of the dips accounted for almost 20 per
cent of the star’s light meant that the obscuring object had to be
enormous. If enough dust existed in a disc to create such a large
dip, it should be stonkingly bright in the infrared—and it wasn’t.
So there’s no dust disc. And the star appears to be perfectly
normal, with nothing in its colour or spectrum marking it out
as one especially likely to behave oddly. The team led by Tabby
checked that there were no signs of camera malfunction. Neigh-
bouring stars appeared to maintain a nice, constant brightness,
and when, driven by some magic combination of desperation
and paranoia, they checked which pixel each observation of our
star landed on there was no obvious pattern that might explain
the observed dips. In the paper we put together announcing the
discovery, and on which seven separate citizen scientists appear
as authors, we fairly reluctantly nailed our colours to a hypoth-
esis that suggested that the dips we observed were the result of a
string of comets.
Comets have a lot to commend them. For starters, they’re less
bright in the infrared than one would expect a dust disc to be,
and that means you can hide enough stuff to cause big dips in
brightness without exceeding the infrared limit set by the obser-
vations. Our comet would need to be broken into bits, so that
each piece could be responsible for an individual dip, but that’s
ok. Breaking up is something that comets tend to do. Comet
Schwassmann–Wachmann 3, for example, survived for sixty-
five years after being discovered by two German observers in
1930, but broke into four pieces in 1995. By 2006 it was in eight
separate pieces and seems to be in the process of crumbling
entirely. Comet Biela, a spectacular sight in the nineteenth cen-
tury, split somewhere around the middle of the century and had
218 Is It AlIens?
disappeared completely by the time of its predicted return in
1859. Both Schwassmann–Wachmann 3 and Biela even produced
short-lived meteor showers, their remnants burning up in the
atmosphere as Earth crossed their orbits.
The most famous of comet breakups was that of Shoemaker–
Levy 9 (SL9), which came too close to Jupiter in the early 1990s.
By the time it was discovered, the giant planet’s gravity had split
it into a string of separate nuclei, each on a collision course with
Jupiter itself. The impacts happened on the far side of the planet
as seen from Earth, but I will never forget the experience of turn-
ing my small backyard
telescope to the planet a few hours later
and seeing clearly the striking bruise left in Jupiter’s atmosphere
by the impact of the first large piece of the comet. I dragged my
parents out of bed so they could take me to the larger telescope at
school, and marvelled at a sight not seen for centuries.
More recent amateur observations have established that aster-
oids and comets hit Jupiter at least a couple of times a year, but
SL9 was special because of the size of the comet and the sheer
drama of the event. For a week or so, impact after impact caused
bruise after bruise in the giant planet’s atmosphere, many of
which remained visible for months following the impact.
These experiences made it seem sensible to us that a comet
might have happened to break up just as Kepler started observing this particular patch of sky. People who actually understand
comets disagreed. A typical comet nucleus is a small thing. That
visited by the European Space Agency’s Rosetta probe and its
famous bouncing lander, Philae, is just a few kilometres across.*
Our comet would have to be the size of Ceres, the largest body in
* Churyumov–Gerasimenko, since you ask, but commonly known as Chewy-
Gooey until it was pointed out that Churyumov and Gerasimenko, its discoverers, might not be amused.
Is It AlIens? 219
the asteroid belt. When discovered, Ceres was large enough to be
considered a planet, but the increasing flood of discoveries
quickly relegated it to being just an asteroid, a nineteenth- century parallel to the plight of Pluto in the twenty-first. Both are now
technically classified as dwarf planets (much to the chagrin of a
loud and very vocal minority of the planetary science community
and the wider cacophony of shouty people online).
So we had either found the largest comet known, and done so
just as it started to disintegrate, or we had no idea any more what
the WTF star was up to. Others had ideas, and we heard a lot
about one of them in particular. Jason Wright from Penn State
and his colleagues thought that our discovery fitted perfectly
with a research programme they had underway, and the title of
their paper was certainly eye-catching. It’s called ‘The Ĝ search
for extraterrestrial civilizations with large energy supplies IV.
The signatures and information content of transiting megastruc-
tures’. It’s that last word that does it; mention finding alien ‘megastructures’ and the world and its dog starts to pay attention.
Specifically, the word ‘megastructures’ turned out to be catnip
for journalists. It sounds just technical enough to make the story
appropriately sciency, while not being so technical that it puts
people off. The paper Jason and friends published (in the
Astrophysical Journal no less—the premier US venue for astro-
nomical research) spends most of its time talking about how one
might, if so inclined, use data to distinguish a transiting alien
space station from the signature of an ordinary planet. The logic
is that any sufficiently advanced alien civilization would want to
make use of as much energy as possible, so rather than idling
away on the surface of a planet like ours would seek to surround
their star with fleets of orbiting solar panels. Often called in sci-
ence fiction a Dyson sphere, a spherical shell surrounding a star
would be unstable. It’s best to think of many individual orbiting
220 Is It AlIens?
Figure 28 Artist’s impression of a Dyson swarm; what it might look like if an alien civilization surrounded their star with solar panels.
spacecraft arrayed into a much larger ‘megastructure’—what the
physicist Freeman Dyson called a swarm (Figure 28). Either way
it would be a spectacular feat of engineering on the grandest of
scales, but as the authors of the paper pointed out, clusters of
swarming alien spacecraft would make a pretty good explanation
for exactly what we see in the WTF star’s blinking.
So Planet Hunters volunteers may have been responsible for
the discovery of alien intelligence, the most significant moment
in astronomical history. Have they really? The press wrote the
story up as if astronomers had seen green tentacles waving back
from a passing spaceship. That fuss made the star famous, and
ultimately led to an appeal on the web to fund Jason and Tabby’s
efforts to keep an eye on their new favourite star.
Screaming ‘Aliens!’—or in this case, having the press scream
‘ALIENS!’ on your behalf—turns out to be a good way not only to
attract those who might want to donate to your research, but
Is It AlIens? 221
also to get other astronomers to notice what you’re doing. There
can’t have been a department anywhere in the world that didn’t
discuss the star, if only in idle chat by the coffee machine, but
conversation led to action at Harvard, where the observatory
keeps a stack of historical images of the night sky. Most exist in
the form of photographic plates, enormous things that could be
strapped to the end of a telescope and exposed to record dim
starlight. The Harvard observatory has spent a lot of time and
energy scanning these things, turning relics sitting in an archive
into useful, digital data, and it was quickly realized that the WTF
star appears in more than ten historical plates, dating back to the
late nineteenth century and stretching forwards to 1970 or so.
These historical records revealed the startling fact that the star
has been gradually fading over the course of the century. This
result started an enormous row among the handful of experts on
such data, who disagreed about how long-term storage and the
process of digitizing the plates might have affected the results,
but more recent, careful analysis of the Kepler data seems to confirm the observed trend. The star is fading slowly and seemingly
inexorably, regardless of the dramatic sudden dips that had
drawn attention. My scientific instinct tells me that we’re looking
for an explanation that ties together both unusual behaviours—
slow fade and sudden dips. Having one star behave oddly for two
different reasons seems like a stretch, and so I reckon we’re
searching for just one answer.
Clearly the slow fading of the star has implications for any
alien civilization too. Perhaps they are still constructing their
star-circling space station. In a note we published in the Journal of Brief Ideas,* with tongues firmly in cheek, Brooke Simmons and
* This is a real thing—you can find our paper here:
org/ideas/424bb64cf38eb9d7db0dae57dec3d28d>.
222 Is It AlIens?
I calculate their progress, assuming that the end goal is a full Dyson sphere which completely captures the star’s light. Assuming
construction doesn’t slow down (or speed up), they’ve got about
700 years left. We also noted in the paper that this probably
meant that elections on any worlds responsible probably
occur less than once a millennium, it being hard to fund infra-
structure projects anywhere if they last longer than a single
&nb
sp; electoral cycle.
By the end of 2017, though, there was still no clear consensus
as to what was going on. New infrared observations suggest a
surrounding dust cloud might be responsible for the slow fade,
but not the dips. The leading hypothesis in my mind is that the
star has recently swallowed a planet. Such an event, as modelled
by exoplanet astronomers whose imagination knows no bounds,
would apparently cause the star to brighten and then to slowly
fade. Any remaining rubble, left over from the inevitable disinte-
gration of the planet, could be responsible for dramatic dips—
the explanation accounts for both halves of the puzzle, but more
evidence is clearly needed.
Specifically, we need data taken during one of the dips by tele-
scopes larger than Kepler. A worldwide network of robotic telescopes has been employed to keep an eye on the star, and plenty
of other professional and amateur observers have joined in too.
For a couple of years, nothing happened. And then, one other-
wise unremarkable day in May 2017, the star dipped once more.
This threw Tabby and her colleagues into a frenzy of activity.
Just as we’d relied on the black market in telescope time to get
that initial spectrum for the Voorwerp, so the team started call-
ing, begging, and pleading for people to observe the star. Some of
this activity happened quietly, as applications for what’s known
as ‘Director’s Discretionary Time’ (available slots in the personal
gift of the observatory director) and programmes which allow
Is It AlIens? 223
one to observe ‘Targets of Opportunity’ went in, but it also con-
sisted of frantic Twitter activity, with Tabby and others posting
the latest data that showed the dip in progress and asking for
help.
That meant we—the world—watched as the star dipped,
recovered, and then dipped again. Debates broke out about
whether the dip was the same shape as on previous occasions.
Spectra were obtained and slowly the star returned to its normal
brightness. What do the results tell us? Well, the mystery remains,
but we know one thing—there is no alien megastructure orbit-
ing this particular star. We know that because observations were
obtained during these 2017 dips in brightness which showed that
The Crowd and the Cosmos: Adventures in the Zooniverse Page 26