Features RSS

SPRITEs

In recent years, astronomers have developed many wide-field imaging surveys in which the same targets are observed again and again. This new form of observing has allowed us to discover optical and radio transients — explosive or irregular events with durations ranging from seconds to years. The dynamic infrared sky, however, has remained largely unexplored … until now.

Infrared Exploration

SPIRITS outcome

Example of a transient: SPIRITS 14ajc was visible when imaged by SPIRITS in 2014 (left) but it wasn’t there during previous imaging between 2004 and 2008 (right). The bottom frame shows the difference between the two images. [Adapted from Kasliwal et al. 2017]

Why hunt for infrared transients? Optical wavelengths don’t allow us to observe events that are obscured, such that their own structure or their surroundings hide them from our view. Both supernovae and luminous red novae (associated with stellar mergers) are discoverable as infrared transients, and there may well be new types of transients in infrared that we haven’t seen before!

To explore this uncharted territory, a team of scientists developed SPIRITS, the Spitzer Infrared Intensive Transients Survey. Begun in 2014, SPIRITS is a five-year long survey that uses the Spitzer Space Telescope to conduct a systematic search for mid-infrared transients in nearby galaxies.

In a recent publication led by Mansi Kasliwal (Caltech and the Carnegie Institution for Science), the SPIRITS team has now detailed how their survey works and what they’ve discovered in its first year.

luminosity gap

The light curves of SPRITEs (red stars) lie in the mid-infared luminosity gap between novae (orange) and supernovae (blue). [Kasliwal et al. 2017]

Mystery Transients

Kasliwal and collaborators used Spitzer to monitor 190 nearby galaxies. In SPIRITS’ first year, they found over 1958 variable stars and 43 infrared transient sources. Of these 43 transients, 21 were known supernovae, 4 were in the luminosity range of novae, and 4 had optical counterparts. The remaining 14 events were designated “eSPecially Red Intermediate-luminosity Transient Events”, or SPRITEs.

SPRITEs are unusual infrared transients that lie in the luminosity gap between novae and supernovae, and they have no optical counterparts. They all occur in star-forming galaxies.

Search for the Cause

What’s the physical origin of these phenomena? The authors explore a number of possible sources, including obscured supernovae, stellar mergers with dusty winds, collapse of extreme stars, or even weak shocks in failed supernovae.

M83

Spitzer image of M83, one of the closest barred spiral galaxies in the sky. SPIRITS 14ajc was discovered in one of M83’s spiral arms. [NASA/JPL-Caltech]

In one case, SPIRITS 14ajc, the SPRITE’s spectrum shows signs of excited molecular hydrogen lines, which are indicative of a shock. Based on the data, Kasliwal and collaborators propose that the shock might have been driven into a molecular cloud after it was triggered by the decay of a system of massive stars that either passed closely or collided and merged.

The other SPRITEs may all have different origins, however, and in general the infrared photometric data isn’t sufficient to identify which model fits each transient. Future technology, like spectroscopy with the James Webb Space Telescope, may help us to better understand the origins of these elusive transients, though. And future surveying with projects like SPIRITS will help us to discover more SPRITE-like events, expanding our understanding of the dynamic infrared sky.

Citation

Mansi M. Kasliwal et al 2017 ApJ 839 88. doi:10.3847/1538-4357/aa6978

OGLE-2016-BLG-1195Lb

What do we know about planet formation around stars that are so light that they can’t fuse hydrogen in their cores? The new discovery of an Earth-mass planet — orbiting what is likely a brown dwarf — may help us better understand this process.

Planets Around Brown Dwarfs?

brown dwarf

Comparison of the sizes of the Sun, a low-mass star, a brown dwarf, Jupiter, and Earth. [NASA/JPL-Caltech/UCB]

Planets are thought to form from the material in protoplanetary disks around their stellar hosts. But the lowest-mass end of the stellar spectrum — brown dwarfs, substellar objects so light that they straddle the boundary between planet and star — will have correspondingly light disks. Do brown dwarfs’ disks typically have enough mass to form Earth-mass planets?

To answer this question, scientists have searched for planets around brown dwarfs with marginal success. Thus far, only four such planets have been found — and these systems may not be typical, since they were discovered via direct imaging. To build a more representative sample, we’d like to discover exoplanets around brown dwarfs via a method that doesn’t rely on imaging the faint light of the system.

gravitational microlensing

A diagram of how planets are detected via gravitational microlensing. The detectable planet is in orbit around the foreground lens star. [NASA]

Lensed Light as a Giveaway

Conveniently, such a method exists — and it’s recently been used to make a major discovery! The planet OGLE-2016-BLG-1195Lb was detected as a result of a gravitational microlensing event that was observed both from the ground and from space.

The discovery of a planet via microlensing occurs when the light of a distant source star is magnified by a passing foreground star hosting a planet. The light curve of the source shows a distinctive magnification signature as a result of the gravitational lensing from the foreground star, and the gravitational field of the lensing star’s planet can add its own detectable blip to the curve.

OGLE-2016-BLG-1195Lb

microlensing

The magnification curve of OGLE-2016-BLG-1195. The peak in the curve in (a) shows the main microlensing by the lens star. An additional blip just after the peak, shown in detail in inset (b), shows the additional lensing by the planet. [Shvartzvald et al. 2017]

A team of scientists led by Yossi Shvartzvald (NASA Postdoctoral Fellow at the Jet Propulsion Laboratory) have now presented the discovery of planet OGLE-2016-BLG-1195Lb, which was made using both ground-based (the Korea Microlensing Telescope Network) and space-based (Spitzer) observations of a microlensing event. The combination of these observations allowed the team to determine a number of properties of the system.

The team’s models indicate that the host is a ~0.072 solar-mass (~74 Jupiter-mass) star, which — if it has the same metallicity as the Sun — likely lies just below the hydrogen-burning mass limit. A ~1.3 Earth-mass planet is orbiting it at a projected separation of ~1.11 AU. The system lies in the galactic disk, roughly 13,700 light-years away.

Looking to the Future

This discovery confirms that the protoplanetary disks of ultracool dwarfs do, in fact, contain enough mass to form terrestrial planets. In addition, the find represents a remarkable technical achievement. OGLE-2016-BLG-1195Lb is the lowest-mass planet ever detected using gravitational microlensing, which bodes well for continued and future microlensing campaigns with high cadences and high detection sensitivity. With luck we’ll soon be able to expand our sample of planets discovered around these unusual hosts, allowing us to build statistics and better understand how and where these planets form.

Citation

Y. Shvartzvald et al 2017 ApJL 840 L3. doi:10.3847/2041-8213/aa6d09

exomoon

Many of the exoplanets that we’ve discovered lie in compact systems with orbits very close to their host star. These systems are especially interesting in the case of cool stars where planets lie in the star’s habitable zone — as is the case, for instance, for the headline-making TRAPPIST-1 system.

But other factors go into determining potential habitability of a planet beyond the rough location where water can remain liquid. One possible consideration: whether the planets have moons.

Supporting Habitability

Moon stability

Locations of equality between the Hill and Roche radius for five different potential moon densities. The phase space allows for planets of different semi-major axes and stellar host masses. Two example systems are shown, Kepler-80 and TRAPPIST-1, with dots representing the planets within them. [Kane 2017]

Earth’s Moon is thought to have been a critical contributor to our planet’s habitability. The presence of a moon stabilizes its planet’s axial tilt, preventing wild swings in climate as the star’s radiation shifts between the planet’s poles and equator. But what determines if a planet can have a moon?

A planet can retain a moon in a stable orbit anywhere between an outer boundary of the Hill radius (beyond which the planet’s gravity is too weak to retain the moon) and an inner boundary of the Roche radius (inside which the moon would be torn apart by tidal forces). The locations of these boundaries depend on both the planet’s and moon’s properties, and they can be modified by additional perturbative forces from the host star and other planets in the system.

In a new study, San Francisco State University scientist Stephen R. Kane modeled these boundaries for planets specifically in compact systems, to determine whether  such planets can host moons to boost their likelihood of habitability.

TRAPPIST-1 moon stability

Allowed moon density as a function of semimajor axis for the TRAPPIST-1 system, for two different scenarios with different levels of perturbations. The vertical dotted lines show the locations of the six innermost TRAPPIST-1 planets. [Kane 2017]

Challenge of Moons in Compact Systems

Kane found that compact systems have a harder time supporting stable moons; the range of radii at which their moons can orbit is greatly reduced relative to spread-out systems like our own. As an example, Kane calculates that if the Earth were in a compact planetary system with a semimajor axis of 0.05 AU, its Hill radius would shrink from being 78.5 times to just 4.5 times its Roche radius — greatly narrowing the region in which our Moon would be able to reside.

Kane applied his models to the TRAPPIST-1 system as an example, demonstrating that it’s very unlikely that many — if any — of the system’s seven planets would be able to retain a stable moon unless that moon were unreasonably dense.

Is TRAPPIST-1 Really Moonless?

Moon transit

Image of the Moon as it transits across the face of the Sun, as viewed from the Stereo-B spacecraft (which is in an Earth-trailing orbit). [NASA]

How do these results fit with other observations of TRAPPIST-1? Kane uses our Moon as an example again: if we were watching a transit of the Earth and Moon in front of the Sun from a distance, the Moon’s transit depth would be 7.4% as deep as Earth’s. A transit of this depth in the TRAPPIST-1 system would have been detectable in Spitzer photometry of the system — so the fact that we didn’t see anything like this supports the idea that the TRAPPIST-1 planets don’t have large moons.

On the other hand, smaller moons (perhaps no more than 200–300 km in radius) would have escaped detection. Future long-term monitoring of TRAPPIST-1 with observatories like the James Webb Space Telescope or 30-meter-class ground-based telescopes will help constrain this possibility, however.

Citation

Stephen R. Kane 2017 ApJL 839 L19. doi:10.3847/2041-8213/aa6bf2

UKIDSS UDS field

Submillimeter galaxies — i.e., galaxies that we detect in the submillimeter wavelength range — are mysterious creatures. It’s only within the last couple decades that we’ve had telescope technology capable of observing them, and we’re only now getting to the point where angular resolution limits allow us to examine them closely. A new study has taken advantage of new capabilities to explore the properties of a sample of 52 of these galaxies.

Dusty Star Formation

Submillimeter galaxies are generally observed in the early universe. Though they’re faint in other wavebands, they’re extremely luminous in infrared and submillimeter — their infrared luminosities are typically trillions of times the Sun’s luminosity. This is thought to be because these galaxies are very actively forming stars at rates of hundreds of times that of the Milky Way!

submm galaxies

Example 10” × 10” true-color images of ten submillimeter galaxies in the authors’ ALMA-identified sample. [Simpson et al. 2017]

Submillimeter galaxies are also extremely dusty, so we don’t see their star formation directly in optical wavelengths. Instead, we see the stellar light after it’s been absorbed and reemitted by interstellar dust lanes — we’re indirectly observing heavily obscured star formation.

Why look for submillimeter galaxies? Studying them can help us to learn about galaxy and star formation early in our universe’s history, and help us to understand how the universe has evolved into what we see locally today.

Submillimeter Struggles

Due to angular resolution limitations in the past, we often couldn’t pin down the exact locations of submillimeter galaxies, preventing us from examining them properly. But now a team of scientists has used the Atacama Large Millimeter/submillimeter array (ALMA) to precisely locate 52 submillimeter galaxies identified by the Submillimeter Common-User Bolometer Array (SCUBA-2) in the UKIDSS Ultra Deep Survey field.

The precise locations made possible by ALMA allowed the team — led by James Simpson (University of Edinburgh and Durham University) — to identify the multi-wavelength properties of these galaxies in a pilot study that they hope to extend to many more similar galaxies in the future.

Lessons from Distant Galaxies

What did Simpson and collaborators learn in this study?

  1. redshift

    Photometric redshift distribution of the ALMA-identified submillimeter galaxies in the authors’ sample (grey). [Simpson et al. 2017]

    For the set of galaxies for which the team could measure photometric redshifts, the median redshift was z ~ 2.65 (though redshifts ranged up to z ~ 5).
  2. Submillimeter galaxies are cooler and larger than local far-infrared galaxies (known as ULIRGs). The authors therefore argue that it’s unlikely that ULIRGs are evolved versions of submillimeter galaxies.
  3. Estimates of dust mass in these galaxies suggest that effectively all of the optical-to-near-infrared light from colocated stars is obscured by dust.
  4. Estimates of the future stellar mass of these galaxies suggest that they cannot evolve into lenticular or spiral galaxies. Instead, the authors conclude, submillimeter galaxies must be the progenitors of local elliptical galaxies.

Citation

J. M. Simpson et al 2017 ApJ 839 58. doi:10.3847/1538-4357/aa65d0

spiral disk

The young star Elias 2-27 is surrounded by a massive disk with spectacular spiral arms. A team of scientists from University of Cambridge’s Institute of Astronomy has now examined what might cause this disk’s appearance.

Top: ALMA 1.3-mm observations of Elias 2-27’s spiral arms, processed with an unsharp masking filter. Two symmetric spiral arms, a bright inner ellipse, and two dark crescents are clearly visible. Bottom: a deprojection of the top image (i.e., what the system would look like face-on). [Meru et al. 2017]

Top: ALMA 1.3-mm observations of Elias 2-27’s spiral arms, processed with an unsharp masking filter. Two symmetric spiral arms, a bright inner ellipse, and two dark crescents are clearly visible. Bottom: a deprojection of the top image (i.e., what the system would look like face-on). [Meru et al. 2017]

ALMA-Imaged Spiral Arms

With the dawn of new telescopes such as the Atacama Large Millimeter/submillimeter Array, we’re now able to study the birth of young stars and their newly forming planetary systems in more detail than ever before. But these new images require new models and interpretations!

Case in point: Elias 2-27 is a low-mass star that’s only a million years old and is surrounded by an unusually massive disk of gas and dust. Recent spatially-resolved ALMA observations of Elias 2-27 have revealed the stunning structure of the star’s disk: it contains two enormous, symmetric spiral arms, as well as additional features interior to the spirals.

What caused the disk to develop this structure? Led by Farzana Meru, a group of Institute of Astronomy researchers has run a series of simulations that explore different ways that Elias 2-27’s disk might have evolved into the shape we see today.

Modeling a Disk

Meru and collaborators performed a total of 72 three-dimensional smoothed particle hydrodynamics simulations tracking 250,000 gas particles in a model disk around a star like Elias 2-27. They then modeled the transfer of energy through these simulated disks and produced synthetic ALMA observations based on the outcomes.

Left: Synthetic ALMA observations of disks shaped by an internal companion (top), an external companion (middle), and gravitational instability within the disk (bottom). Right: Deprojections of the images on the left. Scale is the same as in the actual observations above. The external companion and the gravitational instability scenarios match the actual ALMA observations of Elias 2-27 well. [Adapted from Meru et al. 2017]

Left: Synthetic ALMA observations of disks shaped by an internal companion (top), an external companion (middle), and gravitational instability within the disk (bottom). Right: Deprojections of the images on the left. Scales are the same as in the actual observations above. The external companion and the gravitational instability scenarios match the actual ALMA observations of Elias 2-27 well. [Adapted from Meru et al. 2017]

By comparing these synthetic observations to the true ALMA observations of Elias 2-27, the authors hoped to determine which of three possible scenarios could produce the disk shape we see: 1) a companion (a planet or star) internal to the spiral arms, 2) a companion external to the spirals, or 3) gravitational instabilities operating within the disk.

Gravity or a Companion?

Meru and collaborators find that two scenarios produce observations that are very similar to what ALMA imaged. In the first, the disk is so massive that it becomes gravitationally unstable. Self-gravity of the disk then forms the spiral structures. In the second scenario, the arms are formed by a planetary companion of up to ~10–13 Jupiter masses orbiting Elias 2-27 outside of the spiral arms, at a large distance roughly in the range of 300–700 AU.

Though the possible companion inside the spiral arms is ruled out, the scenarios of a gravitational instability or an external companion remain plausible. If the former is true, then Elias 2-27 would be one of the first examples of an observed self-gravitating disk. If the latter is true, then Elias 2-27’s disk likely fragmented recently, forming the giant planet that shapes the disk. This would be the first evidence for a disk that has fragmented into planetary-mass objects.

Future deep near-infrared imaging may offer the chance to distinguish between these scenarios by allowing us to search for the heat from the possible companion.

Citation

F. Meru et al 2017 ApJL 839 L24. doi:10.3847/2041-8213/aa6837

TRAPPIST-1

The recent discovery of seven Earth-sized, terrestrial planets around an M dwarf star was met with excitement and optimism. But how habitable are these planets actually likely to be? A recent study of these planets’ likely climates may provide an answer to this question.

An Optimistic Outlook

In February of this year, the TRAPPIST-1 system was announced: seven roughly Earth-sized, transiting, terrestrial planets all orbiting their host ultracool dwarf star within a distance the size of Mercury’s orbit. Three of the planets were initially declared to be in the star’s habitable zone — and scientists speculated that even those outside the habitable zone could potentially still harbor liquid water — making the system especially exciting.

planet d

In Wolf’s simulations, the surface temperature (solid lines) of TRAPPIST-1d grows to more than 380K in just 40 years. [Adapted from Wolf 2017]

The planets were labeled as “temperate” because all seven have equilibrium temperatures that are under ~400K. Since liquid water requires a surface temperature of 273-373K, this certainly seems promising!

Finding Realistic Temperatures

But there’s a catch: equilibrium temperatures are not actual measurements of the planet’s surface temperature, they’re just very rudimentary estimates based on how much light the planet receives. To get a better estimate of the real temperature of the planet — and therefore assess its habitability — you need advanced climate modeling of the planet that include factors like the greenhouse effect and planetary albedo.

planet f

In Wolf’s simulations, the surface temperature of TRAPPIST-1f plummets rapidly even when modeled with dense carbon dioxide atmosphere (purple line). The bottom panel shows the corresponding rapid growth of sea-ice on the surface oceans for the different atmospheric models. [Wolf 2017]

To that end, scientist Eric Wolf (University of Colorado Boulder) has conducted state-of-the-art 3D climate calculations for the three center-most planets — planets d, e, and f — in the TRAPPIST-1 system. Wolf assumed traditional terrestrial-planet atmospheres composed of nitrogen, carbon dioxide, and water, and he examined what would happen if these planets had large water supplies in the form of surface oceans.

Runaway and Snowball Planets

Wolf’s climate model indicates that the closest-in of the three planets, planet d, would undergo thermal runaway even in the best case scenario. In just 40 years of the simulation, the planet’s surface temperature exceeds 380K, suggesting it couldn’t continue to sustain liquid water. Wolf argues that planet d and the two planets interior to it, b and c, all lie inside of the traditional liquid water habitable zone — they are hot, dry, and uninhabitable.

Next, Wolf models the outermost of the three center planets, planet f. Even when planet f is modeled with a dense carbon dioxide atmosphere, it can’t avoid its fate of becoming completely ice-covered within roughly 60 years. Wolf concludes that planets f, g and h all lie outside of the traditional habitable zone defined by the maximum carbon dioxide greenhouse limit.

planet e

Equilibrium solutions for TRAPPIST-1e with various atmospheric conditions. Top panel: mean surface temperature. Middle panel: sea-ice coverage. Bottom panel: habitable surface area. [Wolf 2017]

Goldilocks?

Lastly, Wolf turns to planet e, the central planet in the system. This planet, he finds, is the most viable candidate for a robustly habitable world. The simulations show that planet e can maintain habitable surface conditions for a variety of atmospheric compositions.

While astrobiologists eyeing TRAPPIST-1 may be disappointed that at second glance the planets are not quite as inhabitable as they first seemed, it is promising to see that the habitability of the central planet holds up reasonably well to some more realistic testing. Either way, future examinations of all seven of these planets should help us learn more about terrestrial, Earth-sized planets.

Citation

Eric T. Wolf 2017 ApJL 839 L1. doi:10.3847/2041-8213/aa693a

quasar

Quasars — active and luminous galactic centers — can be difficult to find at some high redshifts due to their camouflaging color. A team of scientists has now come up with a way to detect these distant monsters in spite of their disguise.

Quasar Camouflage

color contamination

The color track of quasars between 5 ≤ z ≤ 6 in the commonly used i – z and r – i bands. Each dot on the red line marks a 0.1 difference in redshift. The contours show the colors of M dwarfs, from early type to late type. Quasars at a redshift of 5.3 ≤ z ≤ 5.7 are clearly contaminated by M dwarfs, making them difficult to identify. [Adapted from Yang et al. 2017]

One of the key ways we can study the early universe is by building a large sample of high-redshift quasars. In particular, we believe that reionization of the universe is just completing around z ~ 6. Quasars near this redshift are crucial tools for probing the post-reionization epoch and exploring the evolution of the intergalactic medium, quasar evolution, and early supermassive black hole growth.

But quasars at this redshift are difficult to detect! The problem is contamination: quasars at this distance are the same color in commonly used optical bands as cool M-dwarf stars. As a result, surveys searching for quasars have often just cut out that entire section of the color space in order to avoid this contamination.

This means that there’s a huge gap in our sample of quasars around z ~ 5.5: of the more than 300,000 quasars known, only ~30 have been found in the redshift range of 5.3 ≤ z ≤ 5.7.

color–color selection criteria

The addition of new color–color selection criteria using infrared bands (bottom two plots) allows the authors to differentiate quasars (blue) from M dwarfs (grey), which isn’t possible when only the traditional optical color–color selection criteria are used (top plot). [Adapted from Yang et al. 2017]

A New Approach

In a recent publication led by Jinyi Yang (Peking University, China and Steward Observatory, University of Arizona), a team of scientists has demonstrated a new technique for finding these missing quasars. The team uses this technique to perform the first systematic survey of luminous quasars at a redshift near z = 5.5.

Instead of relying only on the conventional color space in broad optical bands, Yang and collaborators selected candidates by also looking at their colors in near-infrared and mid-infrared bands. The team used observations from the Sloan Digital Sky Survey, the UKIRT Infrared Deep Sky Surveys — Large Area Survey, the VISTA Hemisphere Survey, and the Wide Field Survey Explorer.

Examining the quasar candidates in these color spaces allowed the authors to more clearly differentiate between the M dwarfs and the quasars, so that they could select only the candidates that clearly fell in the regions dominated by quasars in all three spaces. Yang and collaborators then performed spectroscopic follow-up on their candidates to confirm them.

Gap Quasars Uncovered

new quasars found

The authors found 21 new high-redshift quasars (red), including 15 in the range of 5.3 ≤ z ≤ 5.7. [Adapted from Yang et al. 2017]

The team found a total of 21 new quasars from their main sample, with 15 new quasars discovered specifically in the redshift range of 5.3 ≤ z ≤ 5.7. This nearly doubles the number of known quasars at z ~ 5.5!

This initial success has more applications in the future; upcoming surveys will provide an even larger sample to examine for z ~ 5.5 quasars. The team demonstrated that their pipeline can be applied to such surveys by testing it on some preliminary data from the UKIRT Hemisphere Survey. In just this initial test they already discovered another z ~ 5.5 quasar, demonstrating that they’ll have little difficulty finding more once the complete data set is released.

Citation

Jinyi Yang et al 2017 AJ 153 184. doi:10.3847/1538-3881/aa6577

Photograph of a space telescope above the earth.

Every year, astronomers submit over a thousand proposals requesting time on the Hubble Space Telescope (HST). Currently, humans must sort through each of these proposals by hand before sending them off for review. Could this burden be shifted to computers?

A Problem of Volume

HST25 Submissions

Astronomer Molly Peeples gathered stats on the HST submissions sent in last week for the upcoming HST Cycle 25 (the deadline was Friday night), relative to previous years. This year’s proposal round broke the record, with over 1200 proposals submitted in total for Cycle 25. [Molly Peeples]

Each proposal cycle for HST time attracts on the order of 1100 proposals — accounting for far more HST time than is available. The proposals are therefore carefully reviewed by around 150 international members of the astronomy community during a six-month process to select those with the highest scientific merit.

Ideally, each proposal will be read by reviewers that have scientific expertise relevant to the proposal topic: if a proposal requests HST time to study star formation, for instance, then the reviewers assigned to it should have research expertise in star formation.

How does this matching of proposals to reviewers occur? The current method relies on self-reported categorization of the submitted proposals. This is unreliable, however; proposals are often mis-categorized by submitters due to misunderstanding or ambiguous cases.

As a result, the Science Policies Group at the Space Telescope Science Institute (STScI) — which oversees the review of HST proposals — must go through each of the proposals by hand and re-categorize them. The proposals are then matched to reviewers with self-declared expertise in the same category.

With the number of HST proposals on the rise — and the expectation that the upcoming James Webb Space Telescope (JWST) will elicit even more proposals for time than Hubble — scientists at STScI and NASA are now asking: could the human hours necessary for this task be spared? Could a computer program conceivably do this matching instead?

PACMan comparison

Comparison of PACMan’s categorization to the manual sorting for HST Cycle 24 proposals. Green: proposals similarly categorized by both. Yellow: proposals whose manual classifications are within the top 60% of sorted PACMan classifications. Red: proposals categorized differently by each. [Strolger et al. 2017]

Introducing PACMan

Led by Louis-Gregory Strolger (STScI), a team of scientists has developed PACMan: the Proposal Auto-Categorizer and Manager. PACMan is what’s known as a Naive Bayesian classifier; it’s essentially a spam-filtering routine that uses word probabilities to sort proposals into multiple scientific categories and identify people to serve on review panels in those same scientific areas.

PACMan works by looking at the words in a proposal and comparing them to those in a training set of proposals — in this case, the previous year’s HST proposals, sorted by humans. By using the training set, PACMan “learns” how to accurately classify proposals.

PACMan then looks up each reviewer on the Astrophysical Data System (ADS) and compiles the abstracts from the reviewer’s past 10 years’ worth of scientific publications. This text is then evaluated in a similar way to the text of the proposals, determining each reviewer’s suitability to evaluate a proposal.

How Did It Do?

Swaps

Comparison of PACMan sorting to manual sorting, specifically for the HST Cycle 24 proposals that were recategorized by the Science Policies Group (SPG) from what the submitter (PI) selected. Of these swaps, 48% would have been predicted by PACMan. [Strolger et al. 2017]

Strolger and collaborators show that with a training set of one previous cycle’s proposals, PACMan correctly categorizes the next cycle’s proposals roughly 87% of the time. By increasing the size of the training set to include more past cycles, PACMan’s accuracy can be improved up to 95% (though the algorithm will have to be retrained any time the proposal categories change).

PACMan’s results were also consistent for reviewers: it found that nearly all of the reviewers (92%) asked to serve in the last cycle were appropriate reviewers for the subject area based on their ADS publication record.

There are still some hiccups in automating this process — for instance, finding the reviewers on ADS can require human intervention due to names not being unique. As the scientific community moves toward persistent and distinguishable identifiers (like ORCIDs), however, this problem will be mitigated.

Strolger and collaborators believe that PACMan demonstrates a promising means of increasing the efficiency and impartiality of the HST proposal sorting process. This tool will likely be used to assist or replace humans in this process in future HST and JWST cycles.

Citation

Louis-Gregory Strolger et al 2017 AJ 153 181. doi:10.3847/1538-3881/aa6112

Are planetary rings as common in our galaxy as they are in our solar system? A new study demonstrates how we might search for ringed exoplanets — and then possibly finds one!

Saturns Elsewhere?

J1407b

Artist’s illustration of the super ring system around exoplanet J1407b. This is the only exoplanet we’ve found with rings, but it’s not at all like Saturn. [Ron Miller]

Our solar system is filled with moons and planetary rings, so it stands to reason that exoplanetary systems should exhibit the same features. But though we’ve been in the planet-hunting game for decades, we’ve only found one exoplanet that’s surrounded by a ring system. What’s more, that system — J1407b — has enormous rings that are vastly different from the modest, Saturn-like rings that we might expect to be more commonplace.

Have we not discovered ringed exoplanets just because they’re hard to identify? Or is it because they’re not out there? A team of scientists led by Masataka Aizawa (University of Tokyo) has set out to answer this question by conducting a systematic search for rings around long-period planet candidates.

transit light curve

The transit light curve of KIC 10403228, shown with three models: the best-fitting planet-only model (blue) and the two best-fitting planet+ring models (green and red). [Aizawa et al. 2017]

The Hunt Begins

Why long-period planets? Rings are expected to be unstable as the planet gets closer to the central star. What’s more, the planet needs to be far enough away from the star’s warmth for the icy rings to exist. The authors therefore select from the collection of candidate transiting planets 89 long-period candidates that might be able to host rings.

Aizawa and collaborators then fit single-planet models (with no rings) to the light curves of these planets and search for anomalies — curves that aren’t fit well by these standard models. Particularly suspicious characteristics include a long ingress/egress as the planet moves across the face of the star, and asymmetry of the transit shape.

After applying a series of checks to eliminate false positives, the authors are left with one candidate: KIC 10403228.

Rings or Not?

schematics of ringed exoplanet

Schematics of the two best-fitting ringed-exoplanet models for KIC 10403228, and the possible parameters of the system. The planet crosses the disk of the star from left to right with a grazing transit. [Adapted from Aizawa et al. 2017]

Next, the authors apply a wide range of ringed-exoplanet models to KIC 10403228’s light curve. They find two different scenarios that fit the data well: one in which the ring is significantly tilted with respect to the orbital plane, and another in which it’s only slightly tilted.

The authors conclude by testing a variety of other scenarios that could explain the anomalies in the light curve instead. They find that two other scenarios are plausible: 1) the star is in an eclipsing binary system, with the second star surrounded by a circumstellar disk, and 2) the star is part of a hierarchical triple, and the transits are caused by a binary star system as it orbits KIC 10403228.

Though Aizawa and collaborators aren’t able to rule either of these other two scenarios out, they suggest that follow-up spectroscopy or high-resolution imaging may help distinguish between the different scenarios. In the meantime, their methodology for systematically searching for ringed exoplanets has proven worthwhile, and they plan to extend it now to a larger data set. Perhaps we’ll soon find other Saturn-like planets in our galaxy!

Citation

Masataka Aizawa (逢澤正嵩) et al 2017 AJ 153 193. doi:10.3847/1538-3881/aa6336

quasar

J1120+0641

The dust continuum (top) and the [CII] emission (bottom) maps for the region around J1120+0641. [Adapted from Venemans et al. 2017]

A team of scientists has used the Atacama Large Millimeter/submillimeter Array (ALMA) to explore the host galaxy of the most distant quasar known. Their observations may help us to build a picture of how the first supermassive black holes in the universe formed and evolved.

Faraway Monsters and Their Galaxies

We know that quasars — the incredibly luminous and active centers of some distant galaxies — are powered by accreting, supermassive black holes. These monstrous powerhouses have been detected out to redshifts of z ~ 7, when the universe was younger than a billion years old.

Though we’ve observed over a hundred quasars at high redshift, we still don’t understand how these early supermassive black holes formed, or whether the black holes and the galaxies that host them co-evolved. In order to answer questions like these, however, we first need to gather information about the properties and behavior of various supermassive black holes and their host galaxies.

A team of scientists led by Bram Venemans (Max-Planck Institute for Astronomy, Germany) recently used the unprecedented sensitivity and angular resolution of ALMA — as well as the Very Large Array and the IRAM Plateau de Bure Interferometer — to examine the most distant quasar currently known, J1120+0641, located at a redshift of z = 7.1.

A High-Resolution Look

The team’s observations of the dust and gas emission from the quasar’s host galaxy revealed a number of intriguing things:

  1. no rotational motion

    The red and blue sides of the [CII] emission line are shown here as contours, demonstrating that there’s no ordered rotational motion of the gas on kpc scales. [Adapted from Venemans et al. 2017]

    The majority of the galaxy’s emission is very compact. Around 80% of the observed flux came from a region of only 1–1.5 kpc in diameter.
  2. Despite the fact that the 2.4-billion-solar-mass black hole at the galaxy’s center is accreting at a high rate, the heating in the galaxy is dominated not by the black hole’s accretion, but by star formation.
  3. There’s no sign of the expected structure of a rotating disk on kpc scales.
  4. The authors estimate a dynamical mass of the host galaxy of 43 billion solar masses — and the black hole at the galaxy’s center makes up ~6% of that. This ratio is roughly 10x higher than the black-hole-to-bulge mass ratio in local early-type galaxies.
  5. In the very central region, the black hole accounts for around 20% of the galaxy’s dynamical mass, and gas and dust likely accounts for most of the remainder. This doesn’t leave much room for massive stars in the center of the galaxy.

ALMA’s capabilities have enabled these first efforts to spatially resolve the host galaxy of the most distant quasar known, resulting new and unexpected information. The authors now look hopefully to the future, when even longer baselines of ALMA may allow us a still-higher-resolution look at this distant quasar, possibly providing answers to some of the questions it has raised.

Citation

Bram P. Venemans et al 2017 ApJ 837 146. doi:10.3847/1538-4357/aa62ac

1 … 73 74 75 76 77 … 96