Features RSS

Artist's impression of an active galactic nucleus surrounded by a dusty accretion disk

JWST has uncovered a population of point-like red objects in the early universe that may be active black holes shrouded in dust. Recent follow-up observations of one of these objects suggest that it’s an active supermassive black hole that’s disproportionately large compared to the galaxy it inhabits.

Supermassive Black Holes in the Making

illustration of a gas cloud collapsing to form a black hole

The collapse of a primordial gas cloud into a black hole is a possible way to make supermassive black hole seeds in the 104–105-solar-mass range. [NASA/CXC/M. Weiss; CC BY 4.0]

Most galaxies have a central supermassive black hole, and even galaxies in the first billion years of our universe’s history appear to host black holes millions of times as massive as the Sun. How these black holes get so massive in such a small amount of time is an open question; could the small black holes left behind by dying stars accrete enough gas to go from stellar-mass to supermassive in a few hundred million years, or must the seeds of black holes be far larger?

Both scenarios have challenges: it’s not clear how stellar-mass black holes could sustain the accretion rate necessary to transform into supermassive black holes, and we don’t yet have direct evidence for the formation of massive black hole seeds. To learn more about how supermassive black holes come to be, researchers need to amass observations of these objects in the early universe.

Red Source Revealed

Vasily Kokorev (Kapteyn Astronomical Institute) and collaborators followed up on observations of a reddish point source made through the JWST Ultradeep NIRSpec and NIRCam ObserVations before the Epoch of Reionization (UNCOVER) program. Spectra of the source, named UNCOVER ID 20466, reveal it to be at a redshift of z = 8.5, placing it about 600 million years after the Big Bang.

spectrum of the red point source investigated in this work

Two-dimensional spectrum (top) and one-dimensional extracted spectrum (bottom) of the source investigated in this study. Click to enlarge. [Adapted from Kokorev et al. 2023]

The source’s reddish color and point-like nature suggest that it’s a galaxy hosting an active galactic nucleus — a supermassive black hole accreting gas from its surroundings — that is cocooned in dusty gas. This assessment was backed up by its spectrum, which showed a mixture of broad and narrow emission lines. The broad emission lines come from high-density, high-velocity gas near the central black hole, while the narrow lines come from less dense gas farther from the nucleus. This is the most distant active galactic nucleus for which researchers have clearly detected broad emission lines.

How Large Is the Seed?

Plot of observed black hole masses and modeled black hole masses over cosmic time.

Observed black hole masses and ages (symbols) and the black hole masses achievable as a function of time for accretion at the Eddington rate onto a 100-solar-mass seed (red line) and a 104-solar-mass seed (gray line). Click to enlarge. [Kokorev et al. 2023]

The black hole at the center of this distant galaxy has a mass of more than 100 million solar masses, which amounts to at least 30% of the galaxy’s mass being contained in the black hole. This is a much larger proportion than we see for supermassive black holes in the local universe or for typical black holes in the early universe. How did this black hole get so large so quickly, and why is it so large compared to its home galaxy?

Kokorev’s team favors a scenario in which the black hole was seeded by the collapse of a primordial gas cloud into a 104-solar-mass black hole. In this scenario, the black hole can reach its observed mass in the allotted time without surpassing the Eddington limit, the theoretical maximum accretion rate. This also tends to create black holes that are large compared to their host galaxy.

It’s also possible for the black hole’s seed to be smaller, 100 solar masses or so, if it can accrete at super-Eddington rates for hundreds of millions of years — but exactly how that might happen is unclear, especially without the galaxy’s stellar mass growing in tandem. As JWST reveals more of the high-redshift universe, we’ll gain a better understanding of how black hole seeds are sown and sprouted.

Citation

“UNCOVER: A NIRSpec Identification of a Broad-Line AGN at z = 8.50,” Vasily Kokorev et al 2023 ApJL 957 L7. doi:10.3847/2041-8213/ad037a

Artist's impression of a pulsar

The strongest material in the universe isn’t graphene or spider silk or diamonds — it’s the crystalline crust of a dead star’s core. New research explores how to adapt our fluid dynamics models to simulate this exotic material.

A Material Unlike Any Other

Nearly all of the visible matter in our universe is in the form of plasma, which researchers excel at simulating with fluid dynamics models. Solid objects often require a different modeling treatment, since solids have a property that plasmas lack: material strength, or the ability to resist cracking or deformation. Material strength is a critical property for the crusts of neutron stars, which are made of ions arranged in a crystal lattice. Neutron star crust is the strongest material in the universe, and a teaspoon of this superlative matter would weigh 5 tons if brought to Earth’s surface.

This immense strength means that neutron star crusts can’t be modeled with typical fluid dynamics models that don’t take material strength into account. In a recent research article, Irina Sagert (Los Alamos National Laboratory) and collaborators tackle the problem of modeling neutron star crusts with fluid dynamics models, allowing us to study this extreme material with greater accuracy.

Making a Solid Model

Sagert’s team used a smoothed-particle hydrodynamics code called FleCSPH to model waves in a neutron star’s crust. These waves are thought to explain some of the features observed in X-ray flares from neutron stars, and they might affect the gravitational-wave signal produced when neutron stars spiral toward a collision. While previous studies have used smoothed-particle hydrodynamics to explore the behavior of neutron stars, these simulations treated the stars as fluid through and through, including their solid crusts, which made probing these waves impossible. Sagert’s team’s model includes a solid crust atop a fluid core.

plots of rubber rings colliding

Snapshots of the colliding rubber rings. The color scale shows the particle velocity. Click to enlarge. [Sagert et al. 2023]

Before applying their model to neutron stars, the team first modeled several test setups: two rubber rings colliding, compressing, and rebounding off one another; an imploding spherical metal shell; and a cylindrical metal rod striking a solid surface. While these scenarios seem far removed from the crust of a dead star’s core, these tests show the model’s ability to capture the behavior of solids under various stresses. The model passed each test, allowing the team to advance to the main event.

Challenges and Paths Forward

Model of a toroidal oscillation in a neutron star's crust

Model of a toroidal oscillation in a neutron star’s crust. [Sagert et al. 2023]

The team applied their model to the problem of toroidal waves in a neutron star’s crust. This presented several challenges:

  1. The crust makes up just a tiny fraction of a neutron star’s total volume, so most of the computational power goes toward simulating the star’s fluid interior rather than its crust.
  2. In the simplest case, in which the neutron star has no magnetic field, there should be no friction between the crust and the core. Because of the way a smoothed-particle hydrodynamics model performs its calculations, though, there will always be some effective friction between the crust and the core in simulations.
  3. Despite its extreme density and strength, neutron star crust shares something in common with gelatin: it’s much more resistant to being compressed on all sides than it is to being torn apart by shearing. This property means that small numerical fluctuations in the crust’s density can grow large unless suppressed.

The team explored several way of overcoming these challenges, and the resulting model output showed promising agreement with analytical models. The quest to model neutron star crusts isn’t yet over, but Sagert and collaborators see a clear path ahead. Incorporating relativistic physics will open the door to accurate modeling of neutron star mergers and allow researchers to study neutron star collisions and enormous X-ray flares from cracking neutron star crusts more precisely than ever before.

Citation

“Modeling Solids in Nuclear Astrophysics with Smoothed Particle Hydrodynamics,” I. Sagert et al 2023 ApJS 267 47. doi:10.3847/1538-4365/acdc94

A photograph of a brass, mechanical model of the solar system. A small metal Saturn is surrounded by smaller spheres.

Enceladus, one of Saturn’s moons, is currently being tugged around and heated up by another moon named Dione. How the two ended up in this arrangement is a mystery, since to get there, Enceladus must have avoided getting caught up in a resonance with another moon named Tethys. A recent article offers a possible explanation: Enceladus may have blitzed over to its current position in a short-lived burst of speed.

A photograph of a smooth, white moon against a black background.

Enceladus, as seen by the Cassini mission in 2015. [NASA/JPL-Caltech/Space Science Institute]

Dynamic Moons

We typically imagine that moons circle their host planets with clockwork regularity, meaning that they precisely trace out the same path at the same speed for all time. However, true reality cannot be described by a system composed of rigid gears. Instead, moons very slowly change their orbits, wandering around to different separations, eccentricities, and inclinations. Through simulations of their motions over millions of years, astronomers can reveal that what initially looks dull and regular is actually just a snapshot of rich, dynamical behaviors.

Predicting these motions is a challenging task since moons are motivated to wander by forces that are difficult to model. Moons are not homogenous, inflexible solid spheres, meaning they can “squish” and “relax” repeatedly as they get tugged around. This squishing, referred to as tidal interactions, saps energy from the moons’ orbits by converting it to heat that warms up the rock and ice of each moon. Describing this behavior relies on knowing quantities that are difficult to measure, such as the composition and interior structure of the planet and each moon.

Obstacle to Dione

A two panel plot, each of which shows time on the X axis and inclination on the Y. The left plot shows simulations where Enceladus moves at 5x the rate expected from equilibrium tides, while the right plot show simulations that assume a speed 10x the equilibrium. The final inclinations are much larger in the left plot.

An illustration of how Encleadus’s speed affects its response to a resonance. When moving faster (right panel), Enceladus mostly skips over the resonance and remains at a lower inclination compared to the inclination it would have reached had it been moving slower (left panel). Click to enlarge. [Ćuk and El Moutamid 2023]

One of Saturn’s many moons, Enceladus, is currently caught in a resonance with another moon named Dione, meaning it completes nearly exactly two laps around Saturn each time Dione circles the planet once. This configuration greatly enhances tidal forces, and Enceladus is so affected by tides that at least a portion of its icy interior has melted into a subsurface ocean.

But how did the pair of moons reach this configuration? By asserting a guess about the parameters that govern the system’s tidal parameters, astronomers can simulate the motion of the moons over millions of years, then check how well their arrangement matches the present day. This was the technique used by Matija Ćuk (SETI Institute) and Maryame El Moutamid (Cornell University) in their recent article that revealed an interesting conclusion: to have reached its current resonance with Dione, Enceladus must have crossed through a region of parameter space where it risked getting trapped into a resonance with a different moon, Tethys.

A slightly more distant photo of a white moon against a black background. Here, about half the moon is in shadow.

Dione, whose radius is roughly equal to the distance between Boston and Baltimore. This image was taken with the Cassini spacecraft’s narrow-angle camera on June 22, 2017. [NASA/JPL-Caltech/Space Science Institute]

Spurt of Speed

The fact that we see Enceladus in a low-inclination resonance with Dione today implies that it must have blitzed through this other resonance, drifting so quickly that it managed to skate through unaffected. It could not have kept up this speed for long, however, otherwise it would have been swallowed by Saturn by now. Instead, Ćuk and El Moutamid posit that Enceladus must have experienced a short-lived burst of tidal evolution, a time in which it managed to hop over a resonance with Tethys only to get stuck in one with Dione.

Exactly what caused this burst is unclear, as are the effects of complicated resonances with the fluid interior of Saturn itself. Even so, just by knowing this burst happened, astronomers can another chapter to add to the complex history of this fascinating, watery moon.

Citation

“A Past Episode of Rapid Tidal Evolution of Enceladus?” Matija Ćuk and Maryame El Moutamid 2023 Planet. Sci. J. 4 119. doi:10.3847/PSJ/acde80

image of a galaxy cluster

Living in a crowded cluster neighborhood can have a big impact on a galaxy. New research looks to galaxies beyond the local universe to test how living in a galaxy cluster affects galaxies’ stellar populations.

Cluster Influence

spiral galaxy Messier 83

Spiral galaxy Messier 83 is alight with pink and blue star forming regions. Observations suggest that cluster membership can slow a galaxy’s star formation. [NASA, ESA, and the Hubble Heritage Team (STScI/AURA); Acknowledgement: W. Blair (STScI/Johns Hopkins University) and R. O’Connell (University of Virginia)]

Galaxy clusters contain hundreds or thousands of galaxies, and the static-seeming galaxy clusters we see in still images are actually dynamic systems, constantly welcoming new galaxies to the neighborhood. Observations of galaxy clusters in the nearby universe show that when a galaxy joins a cluster, it fundamentally changes the properties of that galaxy. Namely, cluster membership seems to slow the formation of new stars.

That’s what happens in the local universe — does the same story hold for galaxies farther away?

A Higher-Redshift Investigation

Keunho Kim (University of Cincinnati) and collaborators used observations of galaxy clusters spanning 5 billion years of the universe’s history to test this finding farther back in time. The team analyzed South Pole Telescope and Atacama Cosmology Telescope observations of 105 galaxy clusters at redshifts 0.26 < z < 1.13 (about 5.4–10.7 billion years after the Big Bang).

diagram showing the orbit of an infalling galaxy and where certain places on the orbit land the galaxy on a plot relating velocity and radius

Left: Diagram showing the orbit of a galaxy as it falls in to a cluster. Right: Corresponding locations on the position–velocity plot. Click to enlarge. [Kim et al. 2023]

Part of the challenge of studying the effects of cluster membership at such a large distance is that it’s hard to tell which galaxies belong to the cluster; galaxies billions of light-years away from the cluster can get in the way and be mistaken for cluster members. To get around this issue, Kim’s team analyzed not just the on-sky location of the apparent cluster galaxies but also their velocities relative to the center of the galaxy cluster. This analysis revealed not just whether a galaxy belonged to the cluster, but how long since it joined; the slow orbit of an infalling galaxy places galaxies in distinct locations on a position–velocity plot over time.

Slowing Down Star Formation

plot of age of stellar populations versus infall time

Demonstration of how stellar populations become older as time since infall increases. This result holds for galaxies of all luminosities studied. [Kim et al. 2023]

Combining estimates of when the galaxies joined their home clusters with spectroscopically determined ages of their stellar populations, Kim’s team found that the longer galaxies have spent in a cluster, the older their stars are. For the galaxies studied, the difference in average stellar age between those that have been cluster-bound the longest and the late arrivals is about 700 million years.

If the average age of a galaxy’s stars increases with time spent in a cluster, this means that living in a galaxy cluster suppresses star formation — just as was found for local galaxies. This result holds regardless of the luminosity of the galaxy or its redshift, giving us clues about the process responsible for stopping star formation. Given the time frames involved, Kim and collaborators suggest that the process responsible for quenching star formation proceeds slowly, such as the removal of the circumgalactic medium that would normally replenish a galaxy’s star-forming fuel.

Citation

“A Gradual Decline of Star Formation since Cluster Infall: New Kinematic Insights into Environmental Quenching at 0.3 < z < 1.1,” Keunho J. Kim et al 2023 ApJ 955 32. doi:10.3847/1538-4357/acecff

a large sunspot photographed on the Sun

Solar flares are bursts of high-energy radiation that are associated with sunspots. New research uses models to study what happens when sunspots collide and under what conditions these collisions cause solar flares.

magnetic fields of a large sunspot

The color scale in this image shows the strength of the radial magnetic field, with outward-pointing fields in white and inward-pointing fields in black. There is a large sunspot on the left side of the image. [NASA/SDO]

Critical Spots for Research

Sunspots are relatively cool areas of the Sun’s disk (3000–4000K as compared to the 5800K surface) where the solar magnetic field rises above the surface. Sunspots often come in pairs, with one found where magnetic field lines emerge from the solar surface and the other where the field lines plunge back in. When the inward- and outward-pointing magnetic fields of two sunspots meet, the release of magnetic energy can power solar flares and other explosive events.

Not all sunspot collisions produce solar flares, though, and understanding why is a goal of solar physics research. Now, researchers have used fluid dynamics simulations to understand the complex magnetic environment surrounding clashing sunspots.

simulation snapshot showing magnetic field lines

Simulation snapshot showing the magnetic field configuration 14 seconds before the peak of the flare. [Adapted from Rempel et al. 2023]

When Sunspots Collide

Using the three-dimensional MURaM hydrodynamics model, Matthias Rempel (National Center for Atmospheric Research) and collaborators placed sunspots on a collision course, varying the speed of the sunspots and their distance apart to understand how these factors affect the production of solar flares. They started the sunspots at 25,000 kilometers apart, and the programmed trajectories brought the spots as close as 8,500 kilometers. As the sunspots approached and moved past each other, the moving spot broke apart in a process called collisional shearing. This shearing process also occurs in real sunspots, and it may be important to the creation of solar flares.

All of the collisions heated up the solar atmosphere above the sunspots, but only certain simulations resulted in solar flares and ejections of plasma from the solar atmosphere. The higher the velocities of the sunspots and the closer they got together, the more energy was produced, with the distance of closest approach being the biggest determinant of whether or not a solar flare occurred. The team also found that when solar flares did happen, they made good use of the available magnetic energy; 40–50% of the built-up magnetic energy was released in the flares.

Drawing a Line on the Sun

maps of magnetic fields during the simulations

Magnetic fields during the initial state (top left) and closest encounters (remaining panels). The orange lines show the polarity inversion lines. Click to enlarge. [Rempel et al. 2023]

Another way of quantifying these collisions is by their effect on the polarity inversion line: an imaginary line that separates regions of inward-pointing magnetic field from regions of outward-pointing magnetic field. Previous research has shown that close encounters between sunspots that cause complex inversion lines are more likely to be associated with flares and other forms of solar activity. In this study, Rempel’s team found that for collisions with close approaches, the polarity inversion line is physically longer, and these collisions produced solar flares while more distant brushes did not.

The simulated solar flares in this study reached energies corresponding to M-class flares, the second most energetic class of solar flares. Rempel’s team suspects that by cranking up the magnetic flux to the values seen in highly active sunspots, even more powerful flares are possible.

Citation

“Comprehensive Radiative MHD Simulations of Eruptive Flares Above Collisional Polarity Inversion Lines,” Matthias Rempel et al 2023 ApJ 955 105. doi:10.3847/1538-4357/aced4d

zoomed-out view of a spiral galaxy and an elliptical galaxy

Galaxies seem to have less matter than they should. Has the missing matter been found at last in the form of hot, sparse gas?

The Troubling Matter of Normal Matter

image of a spiral galaxy

The space surrounding a galaxy isn’t empty. Instead, it’s filled with tenuous circumgalactic gas invisible to optical telescopes. [NASA, ESA, CXC, SSC, and STScI; CC BY 4.0]

From tiny dwarf galaxies to enormous ellipticals, all galaxies have something in common: they all seem to have less baryonic matter — the stuff that makes up stars, gas, dust, and everything we can see and touch — than we expect. Some galaxies only have a few percent of the matter they should have.

Researchers suspect that the matter isn’t really “missing” but is instead “hidden,” present in a form that’s hard to observe. Recently, a major breakthrough happened when researchers zoomed out from the luminous, starry disks of galaxies to study the tenuous gas of the circumgalactic medium. There, they found immense reservoirs of cool (~104K) gas accounting for as much as 50% of the lost matter. In a recent article, researchers have turned up the heat on the search, seeking out a second, hotter (~106K) component of the circumgalactic medium.

cartoon of the observing setup

A diagram (not to scale!) of the observing setup. X-rays emitted by the quasar pass through the circumgalactic medium of a foreground galaxy. [Kerry Hensley/AAS Nova]

A Critical Alignment

A team led by Fabrizio Nicastro (Italian National Institute for Astrophysics) aimed to track down the remaining missing matter by observing the light from quasars as it passes through galaxies in the foreground. Quasars are the luminous centers of distant galaxies that are powered by supermassive black holes consuming matter. Superheated disks surrounding these active black holes emit enormous amounts of X-rays. As these X-rays travel from the quasar to our telescopes, they can be intercepted by galaxies along their path. If the sought-after hot gas is present in the surroundings of these galaxies, it’ll make itself known by absorbing some of the quasars’ X-ray light.

Nicastro and collaborators analyzed X-ray spectra of three quasars whose lines of sight pass through the circumgalactic medium of a foreground galaxy. None of the individual quasar spectra showed definitive signs of the elusive hot circumgalactic gas, but a signal emerged when the team combined observations of all three quasars. Using several different fitting methods and ways of combining the data, the team detected a signal at a significance ranging from 4.2 to 6.8 sigma.

Missing Mass Found

plot of an X-ray absorption line

Example of an X-ray absorption line identified after stacking the observations from all three quasars. [Adapted from Nicastro et al. 2023]

Based on how much the quasars’ X-ray light was absorbed by the gas in its path, the team estimated the mass of the hot gas surrounding the three galaxies in their sample — and it was a lot. Nicastro and collaborators estimated that 70% of the remaining missing matter can be found in the hot circumgalactic medium, and it might even contain the full amount that is missing.

The discovery of this supply of hot circumgalactic gas has significance beyond solving the missing matter problem: the newfound detection of hot gas can also tell us about how galaxies have exchanged gas with the circumgalactic medium throughout their lives.

Citation

“X-ray Detection of the Galaxy’s Missing Baryons in the Circumgalactic Medium of L* Galaxies,” Fabrizio Nicastro et al 2023 ApJL 955 L21. doi:10.3847/2041-8213/acec70

Hubble image of a star surrounded by a blue nebula

The universe is a surprisingly dusty place. New research takes a look at a rare system where massive stars make dust in their powerful winds.

Stardust Gets Its Start

Cosmic dust is typically born in the atmospheres of evolved stars, where it’s cool enough for carbon and silicon to condense into solid grains. A more dramatic example of dust creation can be seen in the binary system WR 137, where the clashing stellar winds of two massive stars create dust every time the stars approach each other.

Wolf–Rayet star

The dusty ejecta of Wolf–Rayet stars makes for fantastic images. These stars will eventually explode as supernovae. [NASA, ESA, CSA, STScI, Webb ERO Production Team]

Both of the stars in this binary system are extremely rare. One is a Wolf–Rayet star: a once-massive star that has lost its entire hydrogen envelope, leaving its scorching-hot core behind. Evolutionary modeling suggests that the Wolf–Rayet star in WR 137 once clocked in at 60 solar masses, but the combined forces of stellar evolution and mass loss have reduced the star to a mere 4.4 solar masses. The other star is a hot, massive O-type star that is rotating so fast that its atmosphere is being ejected into a disk around the star.

As these stars draw near each other every 13 years, the Wolf–Rayet star’s intense stellar winds (at 4.5 million miles per hour!) pummel the O star’s disk, creating a perfect environment for making dust.

infrared light curves of WR 137

Infrared light curves of WR 137 showing four brightness increases due to dust production. Click to enlarge. [Peatt et al. 2023]

When Stellar Winds Collide

Past observations show an increase in WR 137’s infrared brightness every 13 years. Because dust grains absorb light of many wavelengths and re-emit it in the infrared, this periodic increase in infrared light suggests that there is a periodic increase in dust formation as well. The next flurry of dust formation should happen in 2024, so a team led by Megan Peatt (Embry-Riddle Aeronautical University) seized the opportunity to observe the system at infrared wavelengths using the Stratospheric Observatory For Infrared Astronomy (SOFIA), which has now been decommissioned.

The observations, made in July 2021, February 2022, and May 2022, show a steady increase in infrared emission, marking the increase in dust production as the stars approach each other. In addition to the characteristic spectral lines from the Wolf–Rayet star’s powerful winds, the team also identified a weak emission line around 6.3–6.4 microns (1 micron = 10-6 meter). This feature grew stronger as the system brightened, suggesting that it’s linked to the formation of dust.

Signs of Dust Composition?

zoomed-in spectra showing the change in the 6.2 micron feature

Vertically offset spectra showing the slight migration of the 6.2-micron feature. Click to enlarge. [Peatt et al. 2023]

The precise location of this unidentified emission line appeared to shift to longer wavelengths over time. Peatt’s team suggested that it could reflect a change in the composition of the dust; 6.2-micron emission features could come from a class of molecules called polycyclic aromatic hydrocarbons, which contain rings of carbon atoms bonded to hydrogen atoms. A shift in the emission feature toward longer wavelengths could indicate a shift from hydrogen-rich molecules to hydrogen-poor ones.

This might happen because as the stars draw close, the carbon-rich, hydrogen-poor wind of the Wolf–Rayet star collides and mixes with the hydrogen-rich disk around the O star. As dust production begins, there are ample hydrogen atoms to be wrapped into dust molecules, but as it continues, the proportion of wind material to disk material rises, meaning fewer hydrogen atoms are available. The team emphasizes that this result is speculative, and we’ll need more observations of WR 137 to understand how it makes dust. Hopefully, we’ll learn more about this cosmic dust factory as it nears peak production in 2024!

Citation

“FORCASTing the Spectroscopic Dust Properties of the WC+O Binary WR 137 with SOFIA,” Megan J. Peatt et al 2023 ApJ 956 109. doi:10.3847/1538-4357/acf201

A rendering of a flat spiral galaxy, viewed from an angle. At the center lies an exposed white sphere meant to represent the quasar, which has cleared away all of the gas nearby. A bright jet erupts from the quasar perpendicular to the galaxy in both directions.

Astronomers have long known that the universe has grown more metallic over time: in its younger, purer days, it was composed almost entirely of hydrogen and helium. Recently, however, researchers discovered a galaxy that was notably ahead of the trend and had already amassed a high metal content only a billion years after the Big Bang.

Building Starstuff

Nearly all of the atoms heavier than helium began their lives in a star, the forges of the cosmos responsible for crushing primordial materials into the rich array of elements we see today. These forges run day and night, constantly churning through the universe’s finite supply of hydrogen and helium. Consequently, the overall budget of hydrogen goes down over time, while the proportion of heavier elements (which astronomers call “metals,” regardless of their actual metallic properties) grows. When astronomers look back in time and observe the high-redshift universe, they expect to find mostly pure hydrogen and helium, unpolluted by the starstuff that makes up rocks and people and telescopes.

This prediction usually stands up to observations, and when looking at galaxies with redshifts beyond z = 4 (those born in the first roughly 1.5 billion years after the Big Bang), researchers most often observe clouds of gas with barely any metals. However, a collaboration led by Jianghao Huyan, University of South Carolina, recently discovered a surprising contradiction to this harmonious agreement: their observations of hazy galaxy at z = 4.7 revealed a metal fraction more than two orders of magnitude above the prediction for such a young source.

Mystery Metals

An 8-panel plot, each of which shows wavelength vs. flux for a different absorption line. The red model generally agrees quite well with the black data.

Zoomed-in regions of the measured spectrum centered on different metal absorption lines. The red curve represents the best-fitting model spectrum, while the black histogram is shows their data. [Huyan et al. 2023]

Huyan and colleagues made their discovery when observing a distant quasar named SDSS J002526.84-014532.5 that sits at a redshift of 5.07. Sitting between Earth and this luminous radiation source is a still distant, but slightly closer, galaxy at a redshift of 4.74. As the light from the quasar passed through the wispy gas of the intervening galaxy on the way to our telescopes, specific wavelengths were preferentially absorbed by the molecules and atoms it encountered along the way. By measuring the relative amount of this absorption across many wavelengths, the researchers could back out which elements had tried to block the light’s path, and how dense each species must have been within the gas.

They found that the galaxy possessed a substantial amount of carbon, oxygen, magnesium, and other heavy elements. In fact, just 1.2 billion years after the Big Bang, this galaxy already had a higher relative amount of carbon and oxygen than our own Sun that was born many billions of years later. This was a startling find: models of early galaxy formation expect a significantly smaller fraction of metals, even when accounting for the large uncertainties about the behavior of theorized-but-not-yet-seen first-generation stars.

A 2D plot showing a set of gently downward sloping curves, meant to indicate that metallicity should drop with larger redshifts. The point marking this galaxy lies near the upper right corner of the plot.

Previous measurements and predictions of galaxy metallicity as a function of redshift. The galaxy in question here is marked as the pink triangle that lies far above the model curve. Click to enlarge. [Huyan et al. 2023]

Like many of the most intriguing surprise discoveries, the authors currently have no explanation for what could lead to such a substantial metal enrichment. They concede that it’s possible this particular line of sight may have passed through an anomalously developed region of gas, and that on average, the galaxy as a whole may be as metal-poor as expected. Even in this scenario, however, they cannot explain how that small patch could have been processed to such an extent. Perhaps it is time to revisit the chemical evolution models of early galaxies; perhaps there is something special about this particular galaxy that remains to be uncovered. For now, though, astronomers have another mystery on their hands, and once again the universe has proved ready to challenge our attempts to explain it.

Citation

“Discovery of Super-enriched Gas ∼1 Gyr after the Big Bang,” Jianghao Huyan et al 2023 ApJL 954 L19. doi:10.3847/2041-8213/aceefe

JWST image of Pandora's Cluster

Dark matter, which makes up 85% of the matter in the universe, is thought to interact with normal matter only through gravity. If dark matter and normal matter could interact through collisions, what would that mean for our models of the universe?

maps of the cosmic microwave background anisotropy and local galaxy distribution

Measurements of S8 made from the fluctuations in the cosmic microwave background radiation (top) and the distribution of nearby galaxies (bottom) do not agree. [Top: ESA and the Planck Collaboration; Bottom: M. Blanton and the Sloan Digital Sky Survey]

Clumpiness in Conflict

Our best cosmological model, called lambda-CDM, describes a universe dominated by dark energy, dark matter, and normal matter, in that order. Lambda-CDM has weathered many challenges, but a couple of nagging disagreements have prompted some researchers to propose tweaks or outright rewrites of this leading theory of cosmology.

The first and most famous issue is the Hubble tension: the disagreement between different methods of measuring the expansion rate of the universe. The second issue, and the subject of today’s article, has to do with how clumpy or uniform matter is in our universe. The “clumpiness” parameter, called S8, measured in the nearby universe from observations of galaxy clusters is smaller than what we measure in the distant universe from the cosmic microwave background radiation — the oldest light in the universe, hearkening back to just 380,000 years after the Big Bang. In an attempt to relieve this tension, researchers have explored the effects of loosening one of our fundamental assumptions about dark matter.

Dark Matter Can Have a Little Interaction, as a Treat

In the lambda-CDM model, dark matter is only capable of interacting with normal matter through gravity. This means that dark matter can’t bump into normal matter; a dark-matter particle cruising through your body would have no effect at all. In a recent research article, a team led by Adam He (University of Southern California) suggested that tweaking this property of dark matter could solve our problems with the S8 tension.

plots of posterior distributions

Posterior distributions resulting from different data combinations under the interacting dark matter (IDM) model. [He et al. 2023]

He and collaborators explored the consequences of allowing a small fraction of dark-matter particles to interact with normal matter through collisions. These collisions would allow the two types of matter to exchange heat and momentum. The team used this dark-matter model to analyze observations of galaxy clusters from the Baryon Oscillation Spectroscopic Survey (BOSS) and found that allowing 5–15% of dark-matter particles to collide with normal matter reduced the S8 tension by 30%. Importantly, in the process of lessening the S8 tension, this change doesn’t worsen the Hubble tension — an issue that other models have struggled with.

A Preference for Interacting Dark Matter?

When the team looped in data from the Planck spacecraft, which measured the cosmic microwave background radiation, and the Dark Energy Survey, which has mapped the locations of hundreds of millions of galaxies, they found that the data actually slightly lean toward a scenario in which dark matter and normal matter lightly interact.

Commissioning images from the Euclid space telescope

Commissioning images from the Euclid space telescope. [ESA/Euclid/Euclid Consortium/NASA; CC BY-SA 3.0 IGO]

The results are not conclusive evidence that we need to revamp our understanding of dark matter. He’s team suggests that interacting dark matter lessens the S8 tension because it reduces the clumpiness of matter in the universe on small scales, and upcoming observations of small-scale structure may douse or stoke the interacting-dark-matter fire; the European Space Agency’s Euclid space telescope, which launched in July 2023, will develop a three-dimensional map of the universe, and Rubin Observatory will take a new census of dwarf galaxies, allowing us to map the structure of the universe on the scales required to study this issue further.

Citation

“S8 Tension in the Context of Dark Matter–Baryon Scattering,” Adam He et al 2023 ApJL 954 L8. doi:10.3847/2041-8213/acdb63

side-by-side images of the Sun, a galaxy containing a supernova, and a protoplanetary disk

To meet the challenges posed by our growing collection of data, researchers have devised increasingly sophisticated computing techniques. Today, we’re taking a look at three ways machine-learning techniques have been applied to astrophysical data.

Machine Learning in the Spotlight

Machine learning is a term that describes a collection of techniques in which computers explore data and develop their own algorithms. In astrophysics research, this often takes the form of training computers on a set of known inputs and outputs before introducing data from outside the training set and allowing the computer to derive outputs for those data. For example, researchers could train an algorithm using a set of stellar spectra coupled with known properties of those stars (e.g., spectral type, age, metallicity) and then use the resulting algorithm to classify other stars based on their spectra.

Machine learning and other artificial intelligence techniques are increasingly popular in many fields of science. Here we take a brief look at three recent research articles that describe how machine learning can help us model planet-forming disks, compare observations from different telescopes, and detect fleeting cosmic events.

A Rapid Disk Predictor

A team led by Shunyuan Mao (毛顺元) from the University of Victoria used an artificial neural network to model the interactions between planets and the disks of gas and dust they form in. Planet-forming disks show a wide variety of structures, such as rings and spiral arms, that appear to be linked to the presence, movement, and growth of young planets. By modeling these features, researchers can determine the properties of the planets embedded in protoplanetary disks, but the process can take hours of computing time. Luckily, machine learning appears to offer an easier, faster way to model these disks.

actual versus predicted surface density profile of a gap in a protoplanetary disk

One example of PPDONet’s performance, showing the actual (blue) and predicted (red) density profile of a gap in a disk. [Mao et al. 2023]

Mao’s team has introduced the Protoplanetary Disk Operator Network (PPDONet), which can predict the results of a disk and a planet interacting in less than one second — using a normal laptop. This enormous reduction in computing time is made possible by the team’s machine-learning methods, which recognize when modeling outcomes will be similar to previous runs, jumping ahead and eliminating the need to start every simulation from scratch and iterate through millions of timesteps. The team trained their model on fluid dynamics simulations of disks containing a single planet and found that the model accurately predicts the structure of the disks. The model is publicly available.

Matching Images Between Spacecraft

Researchers wanting to predict solar flares, coronal mass ejections, and other forms of solar activity often base their predictions on images of the Sun taken at extreme-ultraviolet wavelengths. Thanks to spacecraft like the Solar Dynamics Observatory (SDO) and the Solar and Heliospheric Observatory (SOHO), we have decades of solar images to work with, but the differences between telescopes can make it challenging to combine data from different sources into a single prediction — when different observations have different fields of view, spatial and temporal resolution, and noise levels, it’s hard to compare apples to apples.

Demonstration of the different fields of view and spatial resolution of SOHO (left) and SDO (right). [Chatterjee et al. 2023]

To make it possible to work with both SDO and SOHO data sets, Subhamoy Chatterjee (Southwest Research Institute) and collaborators trained a deep-learning model using data from the two spacecraft taken at the same time. The model translated the SOHO images to match the resolution and other characteristics of the SDO images. In an improvement over previous attempts to homogenize solar imaging data, Chatterjee’s team also used Bayesian statistical methods to estimate the uncertainty of the translated images — a critical piece of information for estimating the uncertainty of predictions based on those images.

A Faster Way to Track Down Transients

demonstration of the traditional transient search process

Example of the transient search process. The template image (left) is subtracted from the search image (center), resulting in a difference image (right) that clearly shows a transient source. Click to enlarge. [Adapted from Acero-Cuellar et al. 2023]

Every time we survey the night sky, we find fleeting flashes of light from exploding stars, cosmic collisions, and more. We can learn a lot from studying these events, known as transients, but the process of tracking them down can be time intensive and computationally expensive. A typical method for finding astronomical transients in survey data involves creating reference templates from multiple observations that are then altered to match the seeing conditions and observing setup of the comparison data. The scaled template is then subtracted from the new data, and the resulting image, called a difference image, is scoured for new sources. This method is effective but time consuming, and imaging artifacts, moving stars, and variable stars can all cause false positives. Tatiana Acero-Cuellar (University of Delaware and National University of Colombia) and collaborators suggest that machine learning can make this process faster and eliminate the need for human intervention.

Using data from the Dark Energy Survey, Acero-Cuellar’s team constructed two neural networks to test the possibility of eliminating the difference image altogether. Using one neural network that was trained to use all three images and one that was trained to use all but the difference image, the team found that eliminating the difference image does reduce the network’s ability to identify transients, but only slightly — the accuracy dropped from 96% to 91%. While these neural networks are time-consuming to train, especially when the difference image is not used, putting them into practice requires only a few seconds. This demonstrates the potential for neural networks to eliminate a time-consuming step while retaining a high level of accuracy, which could help us handle the enormous amount of data produced by current and upcoming surveys.

Citation

“PPDONet: Deep Operator Networks for Fast Prediction of Steady-state Solutions in Disk–Planet Systems,” Shunyuan Mao et al 2023 ApJL 950 L12. doi:10.3847/2041-8213/acd77f

“Homogenizing SOHO/EIT and SDO/AIA 171 Å Images: A Deep-learning Approach,” Subhamoy Chatterjee et al 2023 ApJS 268 33. doi:10.3847/1538-4365/ace9d7

“What’s the Difference? The Potential for Convolutional Neural Networks for Transient Detection without Template Subtraction,” Tatiana Acero-Cuellar et al 2023 AJ 166 115. doi:10.3847/1538-3881/ace9d8

1 … 5 6 7 8 9 … 96