Rethinking The Snow Line

Facebooktwittergoogle_plusredditpinterestlinkedinmail
This image of the planet-forming disc around the young star V883 Orionis was obtained by ALMA in long-baseline mode. This star is currently in outburst, which has pushed the water snow line further from the star and allowed it to be detected for the first time. The dark ring midway through the disc is the water snowline, the point from the star where the temperature and pressure dip low enough for water ice to form.
This planet-forming disc around the young star V883 Orionis was obtained by the European Southern Observatory’s  Atacama Large Millimeter/submillimeter Array (ALMA), a prime site for radio astronomy. The star is a state of “outburst,” which has pushed the water snow line further from the star and allowed it to be detected for the first time. The dark ring midway through the disc is the water snowline, the point from the star where the temperature and pressure dip low enough for water ice to form.  ALMA (ESO/NAOJ/NRAO)/L. Cieza

In every planet-forming disk there’s a point where the heat from a host star needed to keep H2O molecules as vapor peters out, and the H2O be becomes a solid crystal.  This is the snow line, and it looms large in most theories of planet formation.

Most broadly, planets formed inside the snow line will generally be rocky and small — a function of the miniscule dust grains that begin the planet forming process.  But outside the snow line the grains get coated by the icy H2O and so are much bigger, leading to gas and ice giant planets.

The existence of water snow lines (and for other molecules, too) is nothing new, but an image of a water snow line would be.  And now an international team led by  Lucas Cieza of Universidad Diego Portales in Santiago, Chile, has found the water vapor/ice line around a very young star 1,350 light-years away. The results were published in the journal Nature this week.

Using a high-precision radio astronomy array in Chile’s Atacama Desert, the team had been looking into whether the massive bursts of young stars might be caused by a theorized collapse into them of fragments of the disk.  But instead they detected and imaged the water snow line instead.

The image itself is an achievement, but what makes the finding especially intriguing is that the snow line was found at an entirely unexpected and enormous distance from the star — more than 42 astronomical units, or forty-two times the distance from our sun to Earth.

That would it was warm enough for H2O to remain a vapor roughly as far out as the orbit of the dwarf planet Pluto around the sun.  A more typical early star snow line is expected to be around 3 AU, an region between the orbits of Mars and Jupiter.

Brenda Matthews, an astronomer at the National Research Council of Canada not involved in the study, wrote in an accompanying column that the snow line finding challenges some traditional models of planet formation.

“The fact that the location of the snow line can evolve with time has strong implications for planet formation,” she wrote. A rapid heating and cooling of the planet-forming disk “would confound models that predict the slow formation of rocky planets within the snow line, and rapid gas-giant formation outside it.”

An artist's illustration shows the water snow line detected around the young star V883 Orionis — the delineation between where the hot star vaporizes all water, leaving rocky dust and debris, and where ice and snow exist in the disk. Credit: A. Angelich (NRAO/AUI/NSF)/ALMA (ESO/NAOJ/NRAO)
An artist’s illustration shows the water snow line detected around the young star V883 Orionis — the delineation between where the hot star vaporizes all water, leaving rocky dust and debris, and where ice and snow exist in the disk. A. Angelich (NRAO/AUI/NSF)/ALMA (ESO/NAOJ/NRAO)

Asked his view of those challenges to the traditional planet forming models Cieza gave our Jupiter and, the widely accepted version of how it formed at 5 AU, as an example:

“If the water snow-line stays far out — let’s say at 10 AU — for very long periods of time, then one would think that forming Jupiter at 5 AU would be very difficult given the importance of water ice for the planet formation process.”

‘That is just a speculation,” Cieza wrote.  “What we really need is planet formation models that take into consideration the very variable accretion rates of proto-stars and the drastic changes in the location of the water snow-line. That is one of the main conclusion of our paper.”

He also described the potential difficulty of delivering the proper amount of water to rocky planets like our own.

“The current thinking is that terrestrial planets form inside the snow line,” he wrote. “We are used to thinking that the Earth is full of water because 70% of its surface  is covered by oceans. However, water represents only 0.02% of the total mass of the planet. Strictly speaking, the Earth is an extremely dry planet because it formed inside the water snow-line, where the temperature in the protoplanetary disk was too hot for water ice to exist and water vapor can not be accreted efficiently by small rocky planets.

The result:  “If the water snow-line had stayed at 10 AU for very long periods of time, perhaps the Earth wouldn’t had received enough water-rich comets and asteroids to form the ocean and we wouldn’t have any life on Earth.”

Cieza and others on the team are not saying that all planet-forming disks necessarily had very distant snow lines at some point, but that some did and that needs to be taken into account.

This view shows several of the ALMA antennas and the central regions of the Milky Way above. In this wide field view, the zodiacal light is seen upper right and at lower left Mars is seen. Saturn is a bit higher in the sky towards the centre of the image. The image was taken during the ESO Ultra HD (UHD) Expedition.
This view shows several of the ALMA antennas and the central regions of the Milky Way above.  The V883 Orionis viewing was done in four different array configurations with baselines ranging from 30 feet apart to almost 8 miles apart.  (ESO)

As explained by Cieza and co-author Zhaohuan Zhu of  Princeton University, the ALMA image does not measure H2O per se, but rather it measures the grain size present in the disk.  It is well established that inside the snow line — where water is in vapor form — the dust grains are small.  But beyond the snow line — where the water is in the form of ice and snow — the dust grains get bigger because the ice adheres to the grains.

So the dust grain size is a stand-in for the phase of the H2O, and can be captured in an image.

The fact that the snow line is so far out also made it possible to make an image. For technical reasons,  current telescopes cannot distinguish characteristics such as grain size that are close to the suns.  ALMA has previously detected the “snow line” for carbon monoxide for a different star at 12 AU, but it cannot get measurements of any molecules for the intriguing “hot Jupiters” that orbit within one AU of their stars.

The heat and luminosity from V883 Orionis –which was detected in 1993 — is pretty remarkable.  The star has only 30 percent more mass than our sun, but it is sending out 400 times more heat and other radiation.  This is all because of those solar bursts that seem to accompany the early evolution of many (or is it most?) stars and their disks.

Zhu, a NASA Hubble Postdoctoral Fellow, said that the early stellar outbursts are expected to last only about 100 years — which makes the detection of  the one at V883 Orionis quite special.  But the short time frame makes it difficult to see the bursts as common, potentially planet-defining phenomenon.

“So far maybe 20 to 30 outbursts have been found,” Zhu said.  “So we aren’t at all sure whether every young star should have an outburst.  It’s very controversial in the community.”

As theorized, the outbursts are caused by material from the disk falling onto the star.  How exactly that might happen is not understood, and was actually the focus of the observations that led to the discovery of the 42 AU water snow line.

Young and highly variable stars like V883 Orionis are included in a class of objects first identified in 1937, with the detection of suddenly very luminous FU Orionis.  V883 Orionis was identified as a FU Orionis, or FUors, star in 1993 based on similarities between its spectrum and that 1937 prototype.

Joel Green, a research scientist at the Space Telescope Science Institute in Baltimore, has studied that original FUor star’s more recent behavior and found that it was still consuming disk material at a remarkable rate.  Since the feeding frenzy started 80 years ago, the star has eaten the equivalent of 18 Jupiters, Green’s team concluded.

In a release last month from NASA’s Jet Propulsion Laboratory, Green said that  “By studying FU Orionis, we’re seeing the absolute baby years of a solar system. Our own sun may have gone through a similar brightening, which would have been a crucial step in the formation of Earth and other planets in our solar system.”

And if it did go through a similar and substantial brightening, what does that say about the snow line in our very early solar system?  And about how and where our planets might have been formed?

 

Facebooktwittergoogle_plusredditpinterestlinkedinmail
Marc Kaufman
Marc Kaufman is the author of two books about space: "Mars Up Close: Inside the Curiosity Mission” and “First Contact: Scientific Breakthroughs in the Search for Life Beyond Earth.” He is also an experienced journalist, having spent three decades at The Washington Post and The Philadelphia Inquirer. While the “Many Worlds” column is supported and informed by NASA’s Astrobiology Program, any opinions expressed are the author’s alone.

To contact Marc, send an email to marc.kaufman@manyworlds.space.

Leave a Reply

Your email address will not be published. Required fields are marked *