Centauri Dreams

What to make of a Jupiter-class planet that orbits its host star at a distance of 13.8 billion kilometers? This is well over twice the distance of Pluto from the Sun, out past the boundaries of what in our system is known as the Kuiper Belt. Moreover, this is a young world still in the process of formation. At nine Jupiter masses, it’s hard to explain through conventional modeling, which sees gas giants growing through core accretion, steadily adding mass through progressive accumulation of circumstellar materials.

Core accretion makes sense and seems to explain typical planet formation, with the primordial cloud around an infant star dense in dust grains that can accumulate into larger and larger objects, eventually growing into planetesimals and emerging as worlds. But the new planet – AB Aurigae b – shouldn’t be there if core accretion were the only way to produce a planet. At these distances from the star, core accretion would take far longer than the age of the system to produce this result.

Enter disk instability, which we’ve examined many a time in these pages over the years. Here the mechanism works from the top down, with clumps of gas and dust forming quickly through what Alan Boss (Carnegie Institution for Science), who has championed the concept, sees as wave activity generated by the gravity of the disk gas. Waves something like the spiral arms in galaxies like our own can lead to the formation of massive clumps whose internal dust grains settle into the core of a protoplanet.

Data from ground- and space-based instruments have homed in on AB Aurigae b, with Hubble’s Space Telescope Imaging Spectrograph and Near Infrared Camera and Multi-Object Spectrograph complemented by observations from the planet imager called SCExAO on Japan’s 8.2-meter Subaru Telescope at Mauna Kea (Hawaii). The fact that the growing system around AB Aurigae presents itself more or less face-on as viewed from Earth makes the distinction between disk and planet that much clearer.

Image: Researchers were able to directly image newly forming exoplanet AB Aurigae b over a 13-year span using Hubble’s Space Telescope Imaging Spectrograph (STIS) and its Near Infrared Camera and Multi-Object Spectrograph (NICMOS). In the top right, Hubble’s NICMOS image captured in 2007 shows AB Aurigae b in a due south position compared to its host star, which is covered by the instrument’s coronagraph. The image captured in 2021 by STIS shows the protoplanet has moved in a counterclockwise motion over time. Credit: Science: NASA, ESA, Thayne Currie (Subaru Telescope, Eureka Scientific Inc.); Image Processing: Thayne Currie (Subaru Telescope, Eureka Scientific Inc.), Alyssa Pagan (STScI).

We benefit from the sheer amount of data Hubble has accumulated when working with a planetary orbit on a world this far from its star. A time span of a single year would hardly be enough to detect the motion at the distance of AB Aurigae, over 500 light years from Earth. The paper on this work – Thayne Currie (Subaru Telescope and Eureka Scientific) is lead researcher – pulls together observations of the system at multiple wavelengths to give disk instability a boost. The authors note the significance of the result, comparing it with PDS 70, a young system with two growing exoplanets, one of whom, PDS 70b, was the first confirmed exoplanet to be directly imaged:

…this discovery has profound consequences for our understanding of how planets form. AB Aur b provides a key direct look at protoplanets in the embedded stage. Thus, it probes an earlier stage of planet formation than the PDS 70 system. AB Aur’s protoplanetary disk shows multiple spiral arms, and AB Aur b appears as a spatially resolved clump located in proximity to these arms. These features bear an uncanny resemblance to models of jovian planet formation by disk instability. AB Aur b may then provide the first direct evidence that jovian planets can form by disk instability. An observational anchor like the AB Aur system significantly informs the formulation of new disk instability models diagnosing the temperature, density and observability of protoplanets formed under varying conditions.

I do want to bring up an additional paper of likely relevance. In 2019, Michael Kuffmeier (Zentrum für Astronomie der Universität Heidelberg) and team looked at a variety of systems in terms of late encounter events that can disrupt a debris disk that is still forming. AB Aurigae is one of the systems studied, as noted in their paper:

Our results show how star-cloudlet encounters can replenish the mass reservoir around an already formed star. Furthermore, the results demonstrate that arc structures observed for AB Aurigae or HD 100546 are a likely consequence of such late encounter events. We find that large second-generation disks can form via encounter events of a star with denser gas condensations in the ISM millions of years after stellar birth as long as the parental Giant Molecular Cloud has not fully dispersed. The majority of mass in these second-generation disks is located at large radii, which is consistent with observations of transitional disks.

Just what effect such late encounter events might have on what may well be disk instability at work will be the subject of future studies, but if we’re using AB Aurigae as a likely model of the process at work, we will need to untangle such effects.

The paper is Currie et al., “Images of embedded Jovian planet formation at a wide separation around AB Aurigae,” Nature Astronomy 04 April 2022 (abstract / preprint). The Kuffmeier paper is “Late encounter-events as a source of disks and spiral structures,” Astronomy & Astrophysics Vol. 633 A3 (19 December 2019). Abstract.

Exoplanet science can look forward to a rash of discoveries involving gravitational microlensing. Consider: In 2023, the European Space Agency will launch Euclid, which although not designed as an exoplanet mission per se, will carry a wide-field infrared array capable of high resolution. ESA is considering an exoplanet microlensing survey for Euclid, which will be able to study the galactic bulge for up to 30 days twice per year, perhaps timed for the end of the craft’s cosmology program.

Look toward crowded galactic center long enough and you just may see a star in the galaxy’s disk move in front of a background star located much further away in that dense bulge. The result: The lensing phenomenon predicted by Einstein, with the light of the background star magnified by the intervening star. If that star has a planet, it’s one we can detect even if it’s relatively small, and even if it’s widely spaced from its star.

For its part, NASA plans to launch the Roman space telescope by 2027, with its own exoplanet microlensing survey slotted in as a core science activity. The space telescope will be able to conduct uninterrupted microlensing operations for two 72-day periods per year, and may coordinate these activities with the Euclid team. In both cases, we have space instruments that can detect cool, low-mass exoplanets for which, in many cases, we’ll be able to combine data from the spacecraft and ground observatories, helping to nail down orbit and distance measurements.

While we await these new additions to the microlensing family, we can also take surprised pleasure in the announcement of a microlensing discovery, the world known as K2-2016-BLG-0005Lb. Yes, this is a Kepler find, or more precisely, a planet uncovered through exhaustive analysis of K2 data, with the help of ground-based data from the OGLE microlensing survey, the Korean Microlensing Telescope Network (KMTNet), Microlensing Observations in Astrophysics (MOA), the Canada-France-Hawaii Telescope and the United Kingdom Infrared Telescope. I list all these projects and instruments by way of illustrating how what we learn from microlensing grows with wide collaboration, allowing us to combine datasets.

Kepler and microlensing? Surprise is understandable, and the new world, similar to Jupiter in its mass and distance from its host star, is about twice as distant as any exoplanets confirmed by Kepler, which used the transit method to make its discoveries. David Specht (University of Manchester) is lead author of the paper, which will appear in Monthly Notices of the Royal Astronomical Society. The effort involved sifting K2 data for signs of an exoplanet and its parent star occulting a background star, with accompanying gravitational lensing caused by both foreground objects.

Eamonn Kerins is principal investigator for the Science and Technology Facilities Council (STFC) grant that funded the work. Dr Kerins adds:

“To see the effect at all requires almost perfect alignment between the foreground planetary system and a background star. The chance that a background star is affected this way by a planet is tens to hundreds of millions to one against. But there are hundreds of millions of stars towards the center of our galaxy. So Kepler just sat and watched them for three months.”

Image: The view of the region close to the Galactic Center centered where the planet was found. The two images show the region as seen by Kepler (left) and by the Canada-France-Hawaii Telescope (CFHT) from the ground. The planet is not visible but its gravity affected the light observed from a faint star at the center of the image (circled). Kepler’s very pixelated view of the sky required specialized techniques to recover the planet signal. Credit: Specht et al.

This is a classic case of pushing into a dataset with specialized analytical methods to uncover something the original mission designers never planned to see. The ground-based surveys that examined the same area of sky offered a combined dataset to go along with what Kepler saw slightly earlier, given its position 135 million kilometers from Earth, allowing scientists to triangulate the system’s position along the line of sight, and to determine the mass of the exoplanet and its orbital distance.

What an intriguing, and decidedly unexpected, result from Kepler! K2-2016-BLG-0005Lb is also a reminder of the kind of discovery we’re going to be making with Euclid and the Roman instrument. Because it is capable of finding lower-mass worlds at a wide range of orbital distances, microlensing should help us understand how common it is to have a Jupiter-class planet in an orbit similar to Jupiter’s around other stars. Is the architecture of our Solar System, in other words, unique or fairly representative of what we will now begin to find?

Animation: The gravitational lensing signal from Jupiter twin K2-2016-BLG-0005Lb. The local star field around the system is shown using real color imaging obtained with the ground-based Canada-France-Hawaii Telescope by the K2C9-CFHT Multi-Color Microlensing Survey team. The star indicated by the pink lines is animated to show the magnification signal observed by Kepler from space. The trace of this signal with time is shown in the lower right panel. On the left is the derived model for the lensing signal, involving multiple images of the star cause by the gravitational field of the planetary system. The system itself is not directly visible. Credit: CFHT.

From the paper:

The combination of spatially well separated simultaneous photometry from the ground and space also enables a precise measurement of the lens–source relative parallax. These measurements allow us to determine a precise planet mass (1.1 ± 0.1 ?_? ), host mass (0.58 ± 0.03 ?_?) and distance (5.2 ± 0.2 kpc).

The authors describe the world as “a close analogue of Jupiter orbiting a K-dwarf star,” noting:

The location of the lens system and its transverse proper motion relative to the background source star (2.7 ± 0.1 mas/yr) are consistent with a distant Galactic-disk planetary system microlensing a star in the Galactic bulge.

Given that Kepler was not designed for microlensing operations, it’s not surprising to see the authors refer to it as “highly sub-optimal for such science.” But here we have a direct planet measurement including mass with high precision made possible by the craft’s uninterrupted view of its particular patch of sky. Euclid and the Roman telescope should have much to contribute given that they are optimized for microlensing work. We can look for a fascinating expansion of the planetary census.

The paper is Specht et al., “Kepler K2 Campaign 9: II. First space-based discovery of an exoplanet using microlensing,” in process at Monthly Notices of the Royal Astronomical Society” (preprint).

A living world around another star will not be an easy catch, no matter how sophisticated the coming generation of space- and ground-based telescopes turns out to be. It’s one thing to develop the tools to begin probing an exoplanet atmosphere, but quite another to be able to say with any degree of confidence that the result we see is the result of biology. When we do begin picking up an interesting gas like methane, we’ll need to evaluate the finding against other atmospheric constituents, and the arguments will fly about non-biological sources for what might be a biosignature.

This is going to begin playing out as the James Webb Space Telescope turns its eye on exoplanets, and methane is the one potential sign of life that should be within its range. We know that oxygen, ozone, methane and carbon dioxide are produced through biological activity on Earth, and we also know that each can be produced in the absence of life. The simultaneous presence of such gases is what would intrigue us most, but the opening round of biosignature detection will be methane. Here I’ll quote Maggie Thompson, who is a graduate student in astronomy and astrophysics at UC Santa Cruz and lead author of a new study on methane in exoplanet atmospheres:

“Oxygen is often talked about as one of the best biosignatures, but it’s probably going to be hard to detect with JWST. We wanted to provide a framework for interpreting observations, so if we see a rocky planet with methane, we know what other observations are needed for it to be a persuasive biosignature.”

The problem of interpretation is huge, given how numerous are the sources of methane. The study Thompson is discussing has just appeared in Proceedings of the National Academy of Sciences, addressing a wide range of phenomena from volcanic activity, hydrothermal vents, tectonic subduction zones to asteroid or comet impacts. There are many ways to produce methane, but because it is unstable in an atmosphere and easily destroyed by photochemical reactions, it needs to be replenished to remain at high levels. Thus the authors look for clues as to how that replenishment works and how to distinguish these processes from signs of life.

Image: Methane in a planet’s atmosphere may be a sign of life if non-biological sources can be ruled out. This illustration summarizes the known abiotic sources of methane on Earth, including outgassing from volcanoes, reactions in settings such as mid-ocean ridges, hydrothermal vents, and subduction zones, and impacts from asteroids and comets. Credit: © 2022 Elena Hartley.

Current methods for studying exoplanet atmospheres rely upon analyzing the light of the host star during a transit as it passes through the planet’s atmosphere, the latter absorbing some of the starlight to offer clues to its composition. To do this well, we need relatively quiet stars with little flare activity. M-dwarfs are great targets for this kind of work because of their small size, so that the transit depth of a rocky planet in the habitable zone will be relatively large and the signal stronger. It’s also useful that small red stars represent as much as 80% of all stars in the galaxy (for a deep dive into this question, see Alex Tolley’s Red Dwarfs: Their Impact on Biosignatures).

Context will be the key in the hunt for biosignatures, with false positives a persistent danger. Outgassing volcanoes should add not only methane but also carbon monoxide to the atmosphere, while biological activity should consume carbon monoxide. The authors argue that it would be difficult for non-biological processes to produce an atmosphere rich in both methane and carbon dioxide with little carbon monoxide.

Thus a small, rocky world in the habitable zone will need to be evaluated in terms of its geochemistry and its geological processes, not to mention its interactions with its host star. Find atmospheric methane here and it is more likely to be an indication of life if the atmosphere also shows carbon dioxide and the methane is more abundant than carbon monoxide and the planet is not extremely rich in water. The paper is an attempt to build a framework for distinguishing not just false positives, but identifying real biosignatures that may be easy to overlook.

Making the process even more complex is the fact that the scope of abiotic methane production on a planetary scale is not fully understood. Even so, the authors argue that while various abiotic mechanisms can replenish methane, it is hard to produce a methane flux comparable to Earth’s biogenic flux without creating clues that signal a false positive. Here we’re at the heart of things; let me quote the paper:

…we investigated whether planets with very reduced mantles and crusts can generate large methane fluxes via magmatic outgassing and assessed the existing literature on low-temperature water-rock and metamorphic reactions, and, where possible, determined their maximum global abiotic methane fluxes. In every case, abiotic processes cannot easily produce atmospheres rich in both CH₄ and CO₂ with negligible CO due to the strong redox disequilibrium between CO₂ and CH₄ and the fact that CO is expected to be readily consumed by life. We also explored whether habitable-zone exoplanets that have large volatile inventories like Titan could have long lifetimes of atmospheric methane. We found that, for Earth-mass planets with water mass fractions that are less than ?1 % of the planet’s mass, the lifetime of atmospheric methane is less than ?10 Myrs, and observational tools can likely distinguish planets with larger water mass fractions from those with terrestrial densities.

Let’s also recall that when searching for biosignatures, terms like ‘Earth-like’ are easy to misuse. Today’s atmosphere is a mix of nitrogen, oxygen and carbon dioxide, but we know that over geological time, the atmosphere has changed profoundly. The early Earth would have been shrouded in hydrogen and helium, with volcanic eruptions producing carbon dioxide, water vapor and sulfur. The oxygenation event that occurred two and a half billion years ago brought oxygen levels up. We thus have to remember where a given exoplanet may be in its process of development as we evaluate it.

So by all means let’s hope we one day find something like a simultaneous detection of oxygen and methane, two gases that ought not to co-exist unless there were a sustaining process (life) to keep them present. An out of equilibrium chemistry is intriguing, because life wants to throw chemical stability out of whack. And by all means let’s accelerate our work in the direction of biosignature analysis to root out those false positives. We begin with methane because that is what JWST can most readily detect.

And as to the question of ambiguity in life detection, JWST is not likely able to detect atmospheric oxygen and ozone, nor will it be a reliable source on water vapor, so its ability to make the call on habitability is limited. Going forward, the authors think that if the instrument detects significant methane and carbon dioxide and can constrain the ratio of carbon monoxide to methane, this will serve as a motivator for future instruments like ground-based Extremely Large Telescopes to follow up these observations. It will take observational tools in combination to nail down methane as a biosignature, but the ELTs should be well placed to take the next step forward.

The paper is Thompson et al., “The case and context for atmospheric methane as an exoplanet biosignature,” Proceedings of the National Academy of Sciences 119 (14) (March 30, 2022). Abstract. See also Krissansen-Totton et al., “Understanding planetary context to enable life detection on exoplanets and test the Copernican principle,” Nature Astronomy 6 (2022), 189-198 (abstract).