N
NOT LONG AGO, THE ORIGIN OF THE SOLAR
SYSTEM SEEMED AN ORDERLY AFFAIR. A cloud
of dust and gas coalesced around a nascent Sun. Close to the newborn
star, volatile substances that vaporize at relatively low temperatures,
like water and other ices, turned into gas and left behind the rocky
planets. Farther out, gas giants formed beyond the “snow line,” where
these volatiles condensed. Earth became the most massive rocky planet
by happenstance, and Jupiter the accidental ruler among the giants.
Shortly after the planets had formed,
a 300 million-year pummeling known as
the Late Heavy Bombardment gave us the
pockmarked faces of the Moon, Mercury,
and other airless worlds. When that ended
some 3.8 billion years ago, all was quiet
save for the odd comet or asteroid that
might sail through the inner system and
hit Earth — like the one that killed the
dinosaurs 66 million years ago. Planetary
scientists thought other solar systems
would look roughly like ours; after all, the
Copernican principle tells us we’re not spe-
cial, and our system likely is average.
Astronomers now know that these for-
mation scenarios are spectacularly wrong.
Planets probably bounced around our Sun
like billiard balls before settling into their
current stately dance. And two decades’
worth of observing planets around other
stars shows that our solar system actually is
quite the oddball. But that leaves the ques-
tion: How did it get this way?
It’s not you, it’s me
Exoplanets were among the first signs that
something was amiss with our solar sys-
tem. Much of this evidence came from
NASA’s Kepler spacecraft, which fixes on
stars and looks for planets
transiting across their faces.
Such transits reveal a planet’s
size and period. When Kepler
launched in 2009, the num-
ber of confirmed exoplanet
systems was in the low hun-
dreds. Kepler has since
pushed that number into
the thousands.
If you plot the periods of
confirmed planets against
their radii, most fall between
one and 10 times Earth’s radius with peri-
ods of 120 days or less. Multiplanet systems
tend to have lots of “super-Earths” — rocky
planets bigger than our own, but no more
than about 17 Earth masses (approximately
the mass of Neptune). Super-Earths are the
most common species among the more
than 3,700 exoplanets discovered so far
— and yet, none exist in our solar system.
Another type of planet that shows up
consistently, though much less often, is the
hot Jupiter. These are gas giants that lie
close to their stars. When astronomers dis-
covered 51 Pegasi b — the first planet found
around a Sun-like star — in 1995, it turned
out to be a hot Jupiter. Now known as
Dimidium, the planet is half Jupiter’s mass
and circles its star in 4.2 days at an average
distance of 4.8 million miles (7.8 million
kilometers). In contrast, Mercury takes
88 days to orbit the Sun and gets no closer
than 29 million miles (46 million km).
Then-current models had no sensible way
to make a Jupiter-sized planet so close to
its star. Dimidium must have moved.
“The standard model, the one everyone
had in their heads, was that the inner disk
was dry and volatile-poor, with planets like
Mercury, which is iron rich and dense. In
the outer system, you’d get
colder, and have lower densi-
ties,” says David Minton of
Purdue University, who
studies planet bombard-
ments and formation. “The
Kepler mission turned that
on its head.”
It’s possible that detection
bias has played a role in
these results. Planet-finding
methods favor worlds with
short periods and high
Planets probably
bounced around
our Sun like
billiard balls
before settling
into their current
stately dance.
24
A ST R O N O M Y • MAY 2018
Jupiter
largely
controlled
the solar
system’s early
evolution. Its inward
migration limited the amount of material available
to form terrestrial planets. Its subsequent outward
migration moved the giant worlds. NASA/JPL/SSI
masses. Scientists have discovered most
exoplanets via Kepler’s transit method, or
by using spectroscopy to observe small
changes in a star’s velocity along the line of
sight. But an alien astronomer looking at
our solar system with a Kepler-like tele-
scope would have to wait at least 24 years
to see two Jupiter orbits, the minimum
needed to confirm its existence.
Still, even accounting for this bias, the
existence of super-Earths and hot Jupiters
showed that models of planet formation at
the very least needed work.
Missing mass and Mars
Besides the exoplanets, other clues showed
up closer to home. When planetary scien-
tists tried building solar systems with
increasingly sophisticated computer
simulations, they kept getting a Mars
that was five to 10 times as massive as
the one we have.
In addition, models of the disk from
which the planets formed assume it had
a relatively uniform density. Astronomer
Alessandro Morbidelli at the Observatoire
de la Côte d’Azur in Nice, France, has
made a career of modeling planet dynam-
ics. He says a uniform density would point
to at least an Earth’s mass of material in
the current asteroid belt. But the actual
mass of the whole thing is only a few