Stellar Façades
28 NOVEMBER 2019 • SKY & TELESCOPE
get their shape, the researchers did see vast bubbles of plasma
on the surface boiling up from the interior and spanning
more than one-quarter of the star’s diameter. Just one of
those bubbles, known as convection cells, would fi ll about
80% of the space between Earth and the Sun.
This confi rms what astronomers suspected from theoreti-
cal calculations: Weakening gravity on the surfaces of dying
stars leads to oversized convection cells. But “there’s so much
physics going on that we still don’t know,” says Paladini. “The
next step... is to see how long they last on the surface.”
Such details could help astronomers better understand how
energy churns about in aging stars.
Pi^1 Gruis offers a potential glimpse into the Sun’s future.
But stars that are signifi cantly heftier evolve in a differ-
ent way. That’s partly what drew Keiichi Ohnaka (Catholic
University of the North, Chile) to aim the VLTI at Antares,
the dazzling red star that marks the heart of the constellation
Scorpius, the Scorpion.
Lying about 550 light-years away, Antares is roughly 12
times as massive as the Sun — heavy enough to one day
explode in a supernova. “We know these dying stars are
losing mass, but we still don’t know yet how,” says Ohnaka.
Some force must be working against the star’s gravity, but
the exact physics is unknown. This phenomenon, he says, “is
quite important to understanding how material is recycled in
the universe.”To chip away at this mystery,
Ohnaka and colleagues used the
VLTI to chart gas motions on and
near Antares. By measuring Dop-
pler shifts in the light from the
gas, the team found enormous gas
clumps racing towards and away
from us at speeds of up to 20 km/s
(45,000 mph). The splotchy
images revealed that this chaotic
environment extends more than
330 million kilometers from the
star’s surface — a distance equal
to about 70% of the star’s radius.
The pattern of gas motion
strongly resembles patterns gener-
ated on the star’s surface from
convection. That seems to suggest
some link between convection
and mass loss — but in computer
simulations of supergiant stars,
convection stirs up only gas that’s
snuggled up to the star.
“Convection exists, but it can’t
shoot up so high,” he says. It
appears that the computer simu-
lations are missing some element.
But it’s not clear yet what that is.Gonna Need a Bigger Interferometer
Getting to this point has been a long road. “For the longest
time, it was a technical art to actually make this work,”
Schaefer says. But now, she says, we have arrays producing
science on a routine basis.
The biggest limitation is that interferometers can see
only fairly bright, relatively nearby stars. To widen their net,
astronomers are going to need arrays of telescopes spread
across a kilometer or more of terrain. “If we have these, then
we will start being able to make images of our neighbors — of
Tau Ceti, of Epsilon Eridani — and really see Sun-like stars,”
Baron says. Although an engineering challenge, when it
comes to physics, “there’s no real obstacle to it,” he adds.
Since 2013, John Monnier (University of Michigan) has
been leading the charge for a ground-based interferometer
known as the Planet Formation Imager (PFI). As its name
implies, PFI would be able to see the infrared glow of giant
planets forming around young stars.
That kind of resolution demands about a dozen 3-meter-
wide telescopes separated by up to 1.2 kilometers, which
would be like having one telescope as wide as Meteor Crater
in Arizona. “We can do it right now,” says Monnier. “It’s
mostly the cost.” He estimates the price tag at around a half
billion dollars, though his team is looking at ways to knock
that down by a few hundred million.
“We’re just in this transition in our fi eld... the next gen-ν
βδπα στεμθ η ζκλpuANTARES Interferometric observations reveal that the
red supergiant in the heart of Scorpius has gigantic convec-
tive cells on its surface.SCORPIUS:^AKIRAFUJII;ANTA
RES^ SURFACE:^ES
O^
/^ K.^ O
HNAKAιρ