Sky & Telescope - September 2020 - 36

Relative brightness

Science in Progress

Exomoon
transit?

1.001
1.000
0.999
0.998
0.997
0.996
0.995

Oct. 29

Oct. 30

Oct. 29

Oct. 30

p AN EXOMOON? Hubble Space Telescope observations of the star Kepler-1625 show a clear dip when a planet passed in front of the star in 2017
(left, central dip). Depending on how they remove instrumental effects from the data, astronomers see (or don't see) signs of a second, much smaller
dip after the planet Kepler-1625b transited. The right-hand graph is one version of the correction, with averaged data in blue.

stars to host moons, because there is no safe orbit for a moon
to reside in. Too far from the planet, and it gets stripped away
by the star's gravity. Too close, and it gets ripped to shreds by
the planet's gravity.
Given that Kepler was mostly able to detect planets close to
their host stars, this explains why Kipping and Teachey haven't
been able to ﬁnd any moons within the Kepler data so far.
That said, one exoplanet among the 284 analyzed did
stand out from the rest - one that, against all odds, just
might host a massive exomoon.

At Last, an Exomoon Candidate
That object, known as Kepler-1625b I, orbits a planet about
the size of Jupiter around a star roughly 8,000 light-years
away. Because the satellite is likely the size of Neptune, the
team nicknamed it "Neptmoon."
But Kipping knew to be cautious - especially after so
many false positives throughout his career. "You can really
get wrapped up in your own little world sometimes when
you're staring at these data sets," he says. "So it's important
to think about the bigger context and to remember
instances when things didn't work out."
To that end, Kipping and Teachey threw
a host of preliminary tests at the system,
aiming to rule out the presence of a
moon. They checked whether the signal

One exoplanet among
the 284 analyzed did
stand out from the rest
- one that, against all
odds, just might host a
massive exomoon.

SEPTEMBER 2 02 0 * SK Y & TELESCOPE

A . TE ACHE Y E T A L. / ASTR ONOMICAL JOUR NAL 2020

perturb the planet's orbit, causing the planetary transits to
diverge from a steady, clocklike recurrence. Transits might
occur earlier or later than expected, or change in duration.
The issue is that these effects are minuscule. Some of
the largest moons in our solar system are roughly the size
of Pluto and Mercury - too small to easily detect around
other stars. Exomoon transits also wouldn't occur in reliable
patterns. "A moon can actually play hide-and-seek and hide
behind the planet or be tilted enough to miss sometimes,"
Kipping says.
In addition, there might be a dearth of exomoons - at
least around planets close to their stars. After Kipping looked
at roughly 60 exoplanets only to ﬁnd zero exomoons, he
decided to adopt a slightly different strategy, in which he and
graduate student Alex Teachey (both now at Columbia University) combined different light curves together from 284
systems to look for an ensemble effect.
The study took three years, and by 2017 the team was convinced that Kepler planets residing from 0.1 to 1 astronomical unit did not host moons - at least not ones larger than
the Galilean satellites of Jupiter.
"We were really surprised not to ﬁnd anything," Kipping says. Then Stephen Kane
(University of California, Riverside)
explained why. He found that it's nearly
impossible for planets so close to their

Sky & Telescope - September 2020

Table of Contents for the Digital Edition of Sky & Telescope - September 2020