Features RSS

infrared Hubble Ultra-Deep Field

Fifteen years ago, the Hubble Space Telescope gazed intently at the infrared glow of galaxies in a tiny fraction of the sky. New research shows how this patch of space has changed since then.

Ultra-Deep and Ultra-Famous

visible-light Hubble Ultra-Deep Field

The Hubble Ultra-Deep Field at visible wavelengths. [NASA, ESA, S. Beckwith (STScI), and the HUDF Team]

The Hubble Ultra-Deep Field is perhaps one of the most recognizable images of our universe. Assembled from observations made in 2003–2004, the visible-light Ultra-Deep Field showcases 10,000 galaxies that stretch back to less than a billion years after the Big Bang.

In 2008–2009 and 2012, Hubble revisited the region, this time piecing together infrared portraits that revealed even more distant galaxies that were absent from the visible-light view. The most distant galaxies in the infrared images appear as they were when the universe was just 2–5% of its current age. What can we learn from comparing these archival images to new images of the same patch of sky?

Paying Another Visit

With Hubble happily still operational (knock on wood!), Matthew Hayes (Stockholm University) and collaborators turned the telescope’s infrared camera toward the Ultra-Deep Field once again in 2023, aiming to find differences between the new images and the original infrared deep-field images from 2008–2009 and 2012.

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

Illustration of an active galactic nucleus. [NASA/SOFIA/Lynette Cook]

In particular, the team hoped to find evidence for faint active galactic nuclei: accreting supermassive black holes at the centers of galaxies. As an active galactic nucleus gulps down varying amounts of gas from its surroundings, its brightness changes like a flickering flame — and these brightness changes are potentially detectable in the set of Ultra-Deep Field images. By cataloging active galactic nuclei in the early universe, researchers hope to pin down how the number of supermassive black holes has changed with cosmic time. This information can help determine how black holes form and evolve.

Spotted: Black Holes

Hayes’s team used two methods to search for sources with varying brightness — potential active galactic nuclei — in the new and archival Hubble images:

  1. Subtracting one image from another to identify objects that appear in only one image
  2. Comparing the brightness of the centers of galaxies between images
photometric variability of active galactic nuclei in the Hubble ultra-deep field

Demonstration of the photometric variability of the two active galactic nuclei. The source at z = 2.0 is shown in the top row, and the source at z = 3.2 is shown in the bottom row. Click to enlarge. [Hayes et al. 2024]

In total, they spotted 71 objects whose brightness varied significantly over the time period. Of the eight objects of interest presented in this work, two are active galactic nuclei at redshifts of z = 2.0 and 3.2 (about 2 to 3 billion years after the Big Bang). Three other objects are likely active galactic nuclei at redshifts beyond z = 6 (less than a billion years after the Big Bang) that couldn’t be definitively cataloged. A further three objects appear to be supernovae, one of which is perched on the edge of a disk galaxy and two of which have no apparent host galaxy.

Hayes’s team used these results to place a lower limit on the number density of black holes during the epoch of reionization, when radiation from the first stars and galaxies transformed the universe from opaque to transparent. They found that the number density of black holes in this time period is similar to the present day value, providing a critical test of black hole formation models. These are just the first results from this campaign, so you can look forward to more ultra-deep-field findings in the future!

Citation

“Glimmers in the Cosmic Dawn: A Census of the Youngest Supermassive Black Holes by Photometric Variability,” Matthew J. Hayes et al 2024 ApJL 971 L16. doi:10.3847/2041-8213/ad63a7

optical image of Cygnus X-1

Astronomers recently tracked a famous X-ray binary system through a change in its accretion state. What does this transition tell us about how black holes accrete gas?

Accretion Questions

illustration of an X-ray binary system

An artist’s impression of an X-ray binary, in which a compact object accretes material from a companion star and emits X-rays during intermittent outbursts [ESO/L. Calçada; CC BY 4.0]

X-ray binaries contain a star and a compact object — either a black hole or a neutron star. As the compact object ensnares gas from its stellar companion, a number of X-ray-bright features can emerge: the gas collects in a super-hot accretion disk and in a tenuous structure called the corona, and transient outflowing jets can appear.

Astronomers have discovered hundreds of X-ray binaries in the Milky Way, but there are still many open questions about the accretion process: What’s the origin of the corona, and how is it structured? Does it sit high above the compact object, or does it hover just above the surface of the disk? What’s the connection between the disk, the corona, and the jets?

One way to potentially answer these questions is to track an X-ray binary as it undergoes a state transition, shifting from producing more low-energy X-rays (“soft state”) to more high-energy X-rays (“hard state”). State transitions are thought to occur when a binary changes how it’s accreting gas, so observing a binary across state transitions can reveal whether the binary’s geometry changes in different accretion modes. Luckily, one of the best-studied X-ray binaries in our galaxy recently gave researchers a chance to study a state transition with a powerful observatory.

From Hard to Soft

Cygnus X-1 is an X-ray binary containing a 41-solar-mass supergiant star and a 21-solar-mass black hole. Over decades of monitoring, scientists have witnessed Cygnus X-1 repeatedly transition between soft and hard states. Phase switches happen randomly, and a phase can last weeks or years.

illustration of the favored coronal geometry

Side view of the favored coronal geometry; a wedge-shaped corona (blue) lies parallel to the accretion disk (yellow). The arrows show the direction of the black hole’s spin. [AAS Nova/Kerry Hensley]

Since the launch of the new Imaging X-ray Polarimetry Explorer (IXPE) spacecraft in 2021, Cygnus X-1 has held steady in a hard state. When researchers examined Cygnus X-1’s hard-state behavior with IXPE, they found that the X-ray emission was unexpectedly strongly polarized. In other words, the orientation of the X-rays as they traveled through space was more orderly than expected. Based on these observations, a “lamppost” model — in which the corona is situated above the black hole’s poles — is now disfavored. Instead, researchers favor a model in which the corona lies parallel to the accretion disk.

Soft-State Insights

polarization degree and angle for Cygnus X-1 hard and soft states

Left: Polarization degree and polarization angle for the hard (blue) and soft (red) states. Right: Polarization degree and angle as a function of energy for the soft state. Click to enlarge. [Steiner et al. 2024]

In April 2023, Cygnus X-1 transitioned out of its long-lived hard state, giving researchers their first opportunity to study the system’s soft-state behavior with IXPE. James Steiner (Center for Astrophysics ∣ Harvard & Smithsonian) and collaborators analyzed five epochs of IXPE data spread over two months. They found that while Cygnus X-1’s X-ray emission in the soft state is less polarized — 2% polarization compared to 4% in the hard state — the two states were otherwise similar; the polarization angle is parallel to the outflowing jet, and the degree of polarization increases with the temperature of the gas.

Using a fully relativistic spectral model, Steiner’s team found that the corona likely lies parallel to the accretion disk, just as in the hard state. While there are many similarities between the hard and soft states of this system, the team suggested that the majority of the polarized light in each state comes from a different source.

In the hard state, X-ray photons become polarized when they scatter off the corona. In the soft state, a substantial fraction of the X-ray photons from the accretion disk are bent back toward the disk by the black hole’s immense gravity, and they become polarized when they are reflected off the surface of the disk. In other words, X-rays from the accretion disk undergo gravitational lensing — showing that the same process that bends the light from distant galaxies is at work in a system billions of times less massive!

Citation

“An IXPE-led X-Ray Spectropolarimetric Campaign on the Soft State of Cygnus X-1: X-Ray Polarimetric Evidence for Strong Gravitational Lensing,” James F. Steiner et al 2024 ApJL 969 L30. doi:10.3847/2041-8213/ad58e4

Hubble image of globular cluster NGC 6387

Millisecond pulsars are some of the most extreme objects in the universe. Surveys of pulsars in globular clusters aim to find these fast-spinning stars and understand why some of them fly solo when they’re expected to be paired up.

Dynamic Duos and Solo Acts

composite image of the Crab Nebula

A composite X-ray, optical, and infrared image of the Crab Nebula, which is energized by the pulsar at its center. [X-ray: NASA/CXC/SAO; Optical: NASA/STScI; Infrared: NASA-JPL-Caltech]

When massive stars expire, they can leave behind their compressed cores as neutron stars, which pack about twice the mass of the Sun into a sphere that could nestle into the narrowest stretch of the English Channel. Pulsars are neutron stars that spin exceptionally quickly and have strong magnetic fields, leading them to produce beams of radio emission that sweep across the sky as the star spins.

The most extreme pulsars are those that complete each rotation in less than 10 milliseconds. Millisecond pulsars are thought to reach their record-setting speeds with the help of a friend, starting out as slower rotators that are “spun up” by accreting material from a binary companion. While that theory explains the origins of millisecond pulsars that inhabit binary systems, it can’t account for those that drift through space alone. To find out where these solo rotators come from, we’ll need to find them first — and researchers are homing in on the best places to look.

diagram comparing the sizes and designs of the Arecibo, FAST, and RATAN-600 radio telescopes

Comparison of the sizes and designs of FAST (center), Arecibo (top), and RATAN-600 (bottom; the world’s largest-diameter single radio dish). [Cmglee; CC BY-SA 4.0]

Seeking Single Pulsars

Of the thousands of known pulsars in the Milky Way, a small but significant fraction can be found in globular clusters: spherical groupings of tens of thousands to millions of stars on the outskirts of the galaxy. Recent observations suggest that globular clusters might be among the best places to find isolated millisecond pulsars.

Using the Five-hundred-meter Aperture Spherical radio Telescope — FAST, the largest filled-aperture radio dish in the world — Dejiang Yin (Guizhou University) and collaborators went pulsar hunting in the globular cluster NGC 6517. Previous searches spotted nine pulsars in NGC 6517, only one of which was in a binary system, suggesting that this cluster may be home to other isolated pulsars.

Best Place to Look

From the new observations, Yin’s team picked out the characteristic pulsed signals from eight millisecond pulsars, all of which appear to lack binary companions. With these new additions, NGC 6517 is now the most pulsar-rich globular cluster accessible to FAST and and the third most of all Milky Way globular clusters. (After this research article was submitted, the team found three more pulsars in their data, so expect to hear more about NGC 6517’s pulsar population soon!)

Why is NGC 6517 home to so many isolated millisecond pulsars? Its density may be responsible: NGC 6517 is one of the most densely packed globular clusters in the Milky Way, which could mean that binary systems containing a millisecond pulsar are more likely to be split apart by close encounters with other stars. Other proposed isolated millisecond pulsar formation mechanisms like neutron star mergers might also be common in dense clusters.

plot of estimated number of pulsars as a function of central escape velocity

The estimated number of pulsars in a globular cluster is correlated with the escape velocity of the cluster. Click to enlarge. [Yin et al. 2024]

Yin’s team performed statistical modeling of NGC 6517 and other pulsar-hosting globular clusters in the Milky Way. They found that the estimated number of pulsars within a cluster scales with the cluster’s escape velocity, leading them to suggest clusters with high escape velocities — Messier 2, Messier 92, Liller 1, NGC 6388, NGC  2808, Messier 54, and Messier 75 — as the best places to search for more isolated millisecond pulsars.

Citation

“FAST Discovery of Eight Isolated Millisecond Pulsars in NGC 6517,” Dejiang Yin et al 2024 ApJL 969 L7. doi:10.3847/2041-8213/ad534e

cosmic microwave background anisotropies and a supernova in a spiral galaxy

In the early 20th century, astronomer Vesto Slipher made the first radial velocity measurements of what were then called spiral nebulae. Nearly all of Slipher’s spiral nebulae — what we now know to be galaxies beyond our own Milky Way — were receding. This simple observation laid the foundation of what is today a complex undertaking: the measurement of the expansion rate of our universe.

Expanding Our Understanding

How can we even measure the universe’s expansion? The rate of expansion can be expressed as a value called the Hubble constant, which bears the unusual units of kilometers per second per megaparsec (km/s/Mpc). (The Hubble constant gets its name from Edwin Hubble, who made one of the first measurements of its value in the local universe; Georges Lemaître beat him to the punch by two years.)

Researchers have different approaches for measuring this constant in the nearby universe and in the more distant, early universe. The expansion rate in the nearby universe can be estimated by combining two measurements: 1) the distance to other local galaxies, and 2) how quickly those galaxies are moving away from us. The most precise measurement of the expansion of the early universe, on the other hand, comes from the Planck spacecraft: researchers extracted a value of the Hubble constant from measurements of the oldest light in the universe, known as the cosmic microwave background.

A Cosmic Conundrum

Here’s the catch: the expansion rate measured from the cosmic microwave background can’t be compared directly to the rate measured in the local universe. That’s because the expansion of the universe is accelerating — the expansion rate today is considerably larger than when the universe was young. To compare the early-universe rate to the present-day rate, researchers use the leading cosmological model, ΛCDM, to extrapolate the early-universe value to the present day.

Researchers have produced more than a thousand estimates of the Hubble constant over the past 60 years, and today the local-universe and early-universe expansion rates have both been measured precisely — but the rates do not agree. The extrapolated early-universe expansion rate is around 67 km/s/Mpc, while the present-day value is pinned at around 74 km/s/Mpc. This mismatch is called the Hubble tension, and the solution is unknown. Are our measurement methods at fault? Or are we due for an overhaul of our leading theory of cosmology? In today’s post, we’ll take a look at five recent research articles that tackle the Hubble tension from different angles — proposing ways to alleviate it or staunchly reinforcing its existence.

What If Our Measurements Weren’t Good Enough?

The most precise measurement to date of the local-universe expansion rate hinges upon Hubble Space Telescope observations of Cepheid variable stars. Astronomer Henrietta Swan Leavitt identified the importance of Cepheids in 1912, when she showed that a Cepheid’s pulsation period is directly tied to its intrinsic luminosity. By comparing the apparent brightness of a Cepheid to its intrinsic luminosity, astronomers can measure the distance to Cepheids — and the galaxies they call home — out to about 100 million light-years.

Picking out individual Cepheids in distant galaxies, even within the local universe, is a challenge — the light from multiple stars in crowded stellar neighborhoods can overlap, stymieing measurements of single stars. This could mean that Hubble’s measurements of Cepheid variables are less reliable for more distant stars, skewing estimates of the Hubble constant. Because JWST has better resolution than Hubble, especially at the near-infrared wavelengths necessary for studying stars in far-off, dusty environments, it can provide a valuable test of the conclusions drawn from Hubble data.

plot of magnitude versus period for Cepheid variable stars

Comparison of the Hubble and JWST period and magnitude data for Cepheid variable stars in the galaxy NGC 4258. The JWST data are visibly less noisy. The inset images at the lower right show the degree of stellar crowding in the Hubble data. [Adapted from Riess et al. 2024]

A team led by Adam Riess (Space Telescope Science Institute and Johns Hopkins University) used JWST to measure the pulsation period and brightness of more than a thousand Cepheid variable stars, yielding precise measurements of the distances to these stars. This analysis showed that while the Hubble measurements are noisier than the JWST measurements, they’re no less accurate. The team ruled out at a significance of 8.2σ the possibility that inaccurate measurements of Cepheids in distant, crowded galaxies are responsible for the Hubble tension.

What If Our Assumptions Are Incorrect?

In addition to using Cepheid variable stars, many estimates of the Hubble constant rely on measurements of Type Ia supernovae, which occur when the remnant core of a low- to intermediate-mass star — a white dwarf — attains a mass of roughly 1.4 solar masses and explodes. The peak brightness of a Type Ia supernova greatly outshines a Cepheid variable, extending the cosmic distance scale out to more than a billion light-years.

plot of Hubble constant values and nonrelativistic matter density parameters calculated using different priors for the Type Ia supernova absolute magnitude

Derived values of the local-universe Hubble constant, H0, for different Type Ia supernova magnitude priors and redshift cutoffs. The blue and green shaded areas show the approximate values of the expansion measured in the local universe and extrapolated from Planck data, respectively. While the choice of supernova luminosity distribution affects the derived Hubble constant, it has little effect on Ωm0, which relates to the density of matter in the universe. See original figure set for error bars. [Adapted from Chen et al. 2024]

Because Type Ia supernovae are theorized to occur at the same limiting mass, these explosions were thought to all have the same maximum luminosity, allowing us to use them for reliable distance measurements. However, there is now evidence that not all Type Ia supernovae have the same brightness. Some are brighter than average, which may result from the collision of two white dwarfs. Some are fainter than average, which may happen when the crust of a white dwarf with a mass less than 1.4 solar masses ignites, triggering the explosion of the entire star. Other factors, like the local abundance of elements heavier than helium, might moderate a supernova’s maximum luminosity as well.

This means that instead of being limited to a single value, the maximum luminosities of Type Ia supernovae follow a distribution that is not yet well known. Yun Chen (Chinese Academy of Sciences) and collaborators examined the impact of different Type Ia supernova luminosity distributions on estimates of the Hubble constant. The team tested five luminosity distributions, three of which are Gaussians and two of which are “top hats”: equal likelihood within a certain range of luminosities and zero chance outside that range. The result? Wide-ranging values of the Hubble constant, some of which agree with the early-universe value and others of which agree with the present-day value. Given the huge impact of the underlying luminosity distribution, Chen’s team calls for a better understanding of the intrinsic properties of Type Ia supernovae.

What If We Use Another Measurement Technique?

Southern Ring Nebula from JWST

A view of the Southern Ring Nebula from JWST. [NASA, ESA, CSA, STScI]

The quest to measure the Hubble constant benefits from trying a variety of measurement techniques, and George Jacoby (NSF’s NOIRLab) and collaborators have demonstrated a unique way to measure the Hubble constant: using observations of planetary nebulae. Planetary nebulae form when stars similar in mass to the Sun — up to about 8 solar masses — lose their atmospheres at the end of their lives. High-energy photons from the exposed stellar core ionize the expelled atmospheric gas, creating a beautiful but short-lived nebula.

Individual planetary nebulae can have a wide range of luminosities; we can’t use single planetary nebulae to measure the distance to another galaxy as we can with Cepheid variable stars or Type Ia supernovae. However, when looking at all of the planetary nebulae in a galaxy, a remarkably consistent pattern emerges: for the brightest planetary nebulae, the number of nebulae as a function of luminosity has the same shape regardless of the properties of the galaxy in question. Using this method, researchers can measure the distances to galaxies out to about 130 million light-years.

planetary nebula luminosity function

The planetary nebula luminosity function for NGC 1385. [Jacoby et al. 2024]

Jacoby’s team used archival data from the Very Large Telescope to measure the planetary nebula luminosity function for 16 galaxies. Their resulting measurement of the Hubble constant — 74.2 km/s/Mpc — is consistent with other local-universe measurements but with larger uncertainties. To measure the Hubble constant from planetary nebulae with enough precision to compare against the Type Ia supernova method, the team recommends targeted observations of a larger sample of galaxies containing at least 50 bright planetary nebulae.

What If We Could Measure the Hubble Constant from Gravitational Waves?

All of the Hubble-constant measurement techniques discussed so far have relied on electromagnetic radiation from one source or another. When the Laser Interferometer Gravitational-wave Observatory (LIGO) and Virgo detectors made the first direct measurement of gravitational waves in 2015, it opened a whole new window into the workings of the universe — and an entirely new way to measure its expansion rate.

Tonghua Liu (Yangtze University) and coauthors demonstrated how gravitational wave observations could enable measurements of the expansion rate out to a redshift of z = 5, or a little more than a billion years after the Big Bang. That time period is currently inaccessible to other measurement methods. Here’s how that would work: first, imagine a binary pair of neutron stars. As these objects circle one another, they expend energy in the form of gravitational waves and sink closer together, hastening their inevitable collision. As the universe expands, the expansion imprints a phase shift in the gravitational waves from the neutron star pair. This effect is tiny, amounting to just a one-second phase shift over 10 years of monitoring a neutron star binary at a redshift of z = 1. Tiny — but not impossible for future gravitational wave observatories to measure!

The phase shift and redshift of a single neutron star binary provide a direct measurement of the expansion rate of the universe at the binary system’s redshift. By measuring phase shifts for a large sample of neutron star systems across a wide range of redshifts, astronomers could measure how the expansion rate of the universe changes across cosmic time — without invoking any assumptions from a particular cosmological model.

plot of simulated acceleration parameters as a function of redshift

Acceleration parameters from simulated DECIGO data. After collecting a large sample of individual acceleration parameter measurements, a machine-learning algorithm can piece together the underlying trend that best fits the observations. [Liu et al. 2024]

Liu’s team simulated phase-shift measurements from a proposed space-based observatory, the Deci-hertz Interferometer Gravitational-wave Observatory (DECIGO), which is designed to detect the low-frequency gravitational waves produced by neutron star binaries years before they merge. Researchers expect that DECIGO will detect a million neutron star binaries, about 10,000 of which would be accompanied by a detectable electromagnetic signal that allows for an estimate of the binary’s redshift. Liu’s team showed that the expansion rate could be measured to within less than a percent, providing a valuable comparison to existing estimates. In addition to measuring the expansion rate, this method also yields a measurement of the universe’s curvature.

What If Dark Energy Is Responsible?

The universe is a mysterious place: according to the leading theory of cosmology, the matter that we see and interact with every day makes up just a tiny fraction — about 5% — of the contents of the universe. Dark matter, a hypothetical form of matter that interacts with everyday matter only through gravity, makes up another 27%. The majority of the matter–energy density of our universe comes from the most mysterious quantity of all: dark energy. Dark energy is thought to provide the outward pressure responsible for the accelerating expansion of the universe, but the exact cause of this pressure remains unknown.

probability distribution function for values of the hubble constant for different models of oscillating dark matter

Probability distribution of the expansion rate (shown here as the rate divided by 100) for different descriptions of oscillating dark matter. The black line shows the result from ΛCDM for comparison. [Rezaei 2024]

The persistence of the Hubble tension has led some researchers to propose alternative theories of cosmology, many of which adjust the properties and identities of dark matter and dark energy. Recently, Mehdi Rezaei (Hamedan Research Center for Applied Meteorology) investigated the impact of oscillating dark energy on the Hubble tension. If dark energy were to oscillate between accelerating and decelerating the expansion of the universe at different points in cosmic time, it could explain the mismatch between the extrapolated early-universe expansion rate and the present-day rate. Using eight different descriptions of oscillating dark energy that have been presented in previous research, Rezaei found that these prescriptions reduced the Hubble tension from its current 5σ severity to 2.14–2.56σ.

In addition to making progress on the Hubble tension, oscillating dark matter could solve another problem of cosmological importance called the coincidence problem. Essentially, the coincidence problem boils down to the fact that in the present-day universe the energy densities of dark matter and dark energy are of the same order of magnitude — but in the distant past and distant future these quantities were/will be way out of balance. In the oscillating dark energy scenario, the energy densities of dark matter and dark energy ebb and flow, and they would have been equal at several points in the universe’s history. While oscillating dark energy alleviates the Hubble tension and the coincidence problem, it doesn’t fit certain observations as well as ΛCDM does — and the search for a definitive solution to the Hubble tension goes on!

Citation

“JWST Observations Reject Unrecognized Crowding of Cepheid Photometry as an Explanation for the Hubble Tension at 8σ Confidence,” Adam G. Riess et al 2024 ApJL 962 L17. doi:10.3847/2041-8213/ad1ddd

“Effects of Type Ia Supernovae Absolute Magnitude Priors on the Hubble Constant Value,” Yun Chen et al 2024 ApJL 964 L4. doi:10.3847/2041-8213/ad2e97

“Toward Precision Cosmology with Improved Planetary Nebula Luminosity Function Distances Using VLT-MUSE. II. A Test Sample from Archival Data,” George H. Jacoby et al 2024 ApJS 271 40. doi:10.3847/1538-4365/ad2166

“Model-Independent Way to Determine the Hubble Constant and the Curvature from the Phase Shift of Gravitational Waves with DECIGO,” Tonghua Liu et al 2024 ApJL 965 L11. doi:10.3847/2041-8213/ad3553

“Oscillating Dark Energy in Light of the Latest Observations and Its Impact on the Hubble Tension,” Mehdi Rezaei 2024 ApJ 967 2. doi:10.3847/1538-4357/ad3963

illustration of two neutron stars spiraling toward each other

In May of last year, an international collaboration of astronomers and physicists switched on a set of detectors and began the most intensive search for gravitational waves ever attempted. They did not have to wait long for a new discovery that challenges our understanding of the extreme universe: just days into their campaign, the scientists caught the signal from the smallest black hole ever detected by their observatories, one that is so light it couldn’t have formed according to our typical models. After a year of analysis, the scores of scientists involved in this remarkable find have just published their work in The Astrophysical Journal Letters.

How To Warp Spacetime

Although gravitational waves, or ripples in spacetime caused by collisions between massive objects, have permeated the universe since its very beginning, scientists have only been able to directly measure them since 2015. Since then, we’ve learned much about these waves and the objects that source them, and we’ve built up a pretty clear picture of what it takes to wrinkle space itself.

photograph of the LIGO Hanford gravitational wave detector

A gravitational wave detector. Ripples in spacetime cause the two arms to change lengths by tiny amounts, and scientists detect passing waves by carefully measuring these distortions. [Caltech/MIT/LIGO Lab]

In short, you need a pair of enormously heavy objects to smash into each other at high speeds. Two types of objects will do the trick: neutron stars, or the exotic remains of supernova explosions, and the more massive and more infamous black holes.

Neutron stars, as best we understand from theoretical models and observations taken with radio and X-ray telescopes, can weigh up to about 3 times the mass of the Sun and are formed when a large star explodes. Black holes, on the other hand, are typically heavier. They weigh more than 5 times the mass of the Sun, and they’re thought to form when a large star collapses in on itself.

Putting these two facts together, it’d be very strange to identify an object that’s between 3 and 5 solar masses: anything in that gap would be too heavy to be a neutron star, but too light to be a black hole. And yet, that’s exactly what the international collaboration found: the collision between a 1.4-solar-mass neutron star and a 3.6-solar-mass black hole, making the latter an inhabitant of the so-called “mass gap.”

A Rule-Breaking Merger

The inferred masses of each component of a handful of gravitational wave mergers. The “mass gap” is marked in light grey, and the likely mass of the primary component of this event, GW230529, lies nearly entirely within it. [A. G. Abac et al 2024]

Their merger, named in the traditionally catchy parlance of astronomers as GW230529_181500, was a real surprise. Though the collaboration has a total of four gravitational wave detectors at its disposal, only one saw the signal: the newest, named KAGRA, is not sensitive enough to spot an event this faint, while VIRGO was undergoing upgrades at the time and one of the two LIGO sites had been switched off only an hour and a half too soon to catch the show. This lack of detector backup, combined with the weirdness of the implied progenitor, left the team with a lot of work to do to prove that what they were seeing was real.

In the end, the event passed all of their tests. The wave was detected by three independent analysis pipelines, and although the choices that go into modeling the collision can affect the inferred mass, most reasonable options all point to the same conclusion: the observed gravitational wave really does seem to have come from an impossibly light black hole and its merger with a typical neutron star. This proof-by-contradiction that small black holes exist implies that there are probably plenty of them, which happily means we can look forward to additional head-scratching detections of similar pairs in the future.

Citation

“Observation of Gravitational Waves from the Coalescence of a 2.5–4.5 M⊙ Compact Object and a Neutron Star,” A. G. Abac et al 2024 ApJL 970 L34. doi:10.3847/2041-8213/ad5beb

Hubble image of the Flame Nebula

visualization of a turbulent flow

Visualization of the plume from a candle transitioning from a smooth flow near the wick to a turbulent flow higher up. [Gary Settles; CC BY-SA 3.0]

Turbulence — the culprit behind bumpy airplane rides, breaking waves, and billowing clouds — happens throughout the universe. Modeling realistic turbulent plasmas is difficult and computationally expensive, but new methods make simulating turbulence easier, faster, and more flexible than ever before.

A Chaotic Challenge

In the accretion disks around supermassive black holes, the clouds of molecular gas where stars are born, and even the atmosphere of the Sun, turbulence plays an important role in transferring energy and mixing plasmas. Because turbulent plasmas are so common, researchers across astronomical fields must wrestle with the challenge of modeling these chaotic systems.

The traditional way to model a turbulent system is to do so numerically: churning through sets of equations to track the behavior of a plasma across tiny time increments. But as models become more complex, this comes at a higher computational cost, requiring an increasingly large number of hours for computers to handle the math.

representation of a simulated three-dimensional turbulent magnetic field

A representation of a typical three-dimensional turbulent magnetic field generated by BxC. [Adapted from Maci et al. 2024]

Synthetic Magnetic Fields

Synthetic models that use relatively simple algorithms to generate realistic plasmas provide an alternative. Recently, Daniela Maci (KU Leuven) and collaborators introduced a new synthetic model of turbulent magnetic fields. The team’s model builds on an existing synthetic model called BxC, a Python-based code that speedily produces three-dimensional turbulent magnetic fields.

To generate realistic turbulence, BxC starts with a random field of vectors based on white noise. From there, this vector field is tweaked and transformed to adjust its statistical properties. In practice, this means that not only do BxC’s synthetic fields look turbulent, they also reproduce the expected statistical properties of a turbulent field, setting BxC apart from other models. This makes BxC an excellent jumping-off platform for creating realistic turbulent magnetic fields, and Maci’s team showed how the statistical properties of the modeled field depend on the model’s inputs.

Paths Forward

simulation results showing turbulence that develops on top of a background magnetic field

Examples of simulated turbulence that develops on top of a background magnetic field. Click to enlarge. [Adapted from Maci et al. 2024]

To make the synthetic fields more applicable to astrophysical plasmas, Maci’s team expanded upon the original code in a couple of key ways. First, they recognized that turbulence often develops on top of a background magnetic field, and they developed a way for turbulence to be applied while maintaining the underlying structure of the field. This applies to magnetic field structures like loops or flux tubes, both of which are seen in the atmosphere of the Sun. Second, they developed a way to incorporate anisotropy, or a difference in the strength of the turbulence with regard to direction.

This work demonstrates the power and flexibility of synthetic models for representing turbulent magnetic fields — and all of these features are available orders of magnitude faster than they would be from a traditional numerical model. The team anticipates making future additions to their model, all of which would give the user more customization options and make the model applicable to more astrophysical scenarios.

Citation

“BxC Toolkit: Generating Tailored Turbulent 3D Magnetic Fields,” Daniela Maci et al 2024 ApJS 273 11. doi:10.3847/1538-4365/ad4bdf

photograph of NGC 3621, the host galaxy of SN 2024ggi

Less than a year after SN 2023ixf captivated astronomers worldwide, another nearby supernova burst onto the scene. At a distance of just 22 million light-years, SN 2024ggi provides another excellent opportunity to study the behavior of red supergiant stars in their final years.

Live Fast, Die Spectacularly

The discovery of a new nearby supernova marks the beginning of a cosmic chase. After astronomers pinpoint a rapidly brightening point of light in a distant galaxy — a new supernova! — they hunt through archival data to learn more about the star that exploded. Using this method, researchers have tracked down dozens of supernova progenitor stars and have learned that core-collapse supernovae usually arise from red supergiant stars with masses in the range of 8 to 18 solar masses.

Supernova hunters have been busy recently: while the dust was still settling from SN 2023ixf — the nearest bright supernova in nearly a decade — the Asteroid Terrestrial-impact Last Alert System discovered the supernova SN 2024ggi on 11 April 2024 in the galaxy NGC 3621. This marks the first time a supernova has been recorded in this galaxy, which is just barely more distant than SN 2023ixf’s host galaxy. What can archival observations tell us about this supernova’s progenitor star?

Archival Analysis

optical and infrared pre-explosion images of the progenitor star of SN 2024ggi

Images of SN 2024ggi’s progenitor from Hubble (top row) and Spitzer (bottom row). [Xiang et al. 2024]

A team led by Danfeng Xiang and Jun Mo (Tsinghua University) turned to observations from the Hubble and Spitzer space telescopes to track down SN 2024ggi’s progenitor. In Hubble images from 1994 to 2003, the team spotted an extremely red star at the location of SN 2024ggi. The star’s reddish hue is likely due to dust that formed around the star in cast-off stellar material.

In Spitzer images of NGC 3621 from 2004 to 2019, the same star is faintly visible but crowded by other stars. By carefully removing the light from the star’s close neighbors, the team was able to monitor the star’s brightness over many years. They found that its brightness varied at both infrared and optical wavelengths with a period of about 378 days. This type of variability is common in red supergiant stars and is likely due to radial pulsations.

Surprisingly Dust Free

Xiang and Mo’s team also plotted the star’s spectral energy distribution, or how its energy output is spread across different wavelengths of light. They used two classes of models to interpret the spectral energy distribution: one model that included a veil of dust around the star and one that was dust free.

observed and modeled spectral energy distributions for the progenitor star of SN 2024ggi

Observed spectral energy distribution of SN 2024ggi’s progenitor star (black squares). The lines show the best-fit spectral energy distributions for the model with dust (red) and without dust (gray). [Adapted from Xiang et al. 2024]

These models suggested that the star had a mass around 13 solar masses and its temperature was a cool 3290K. Surprisingly, given the star’s extremely red color, the dusty and dust-free models both fit the data well, suggesting that the dust shell around the star was thin.

The thinness of the dust shell implies a low mass-loss rate for the star in the decades before its explosion, but researchers studying the subsequent supernova found the star’s mass-loss rate just before its demise to be much higher. This might mean that while SN 2024ggi’s progenitor shed mass at a modest rate during most of its life, it shed far more mass in its final years. This finding adds to the growing body of evidence that red supergiants undergo stellar tantrums before their ultimate explosions.

Citation

“The Red Supergiant Progenitor of Type II Supernova 2024ggi,” Danfeng Xiang et al 2024 ApJL 969 L15. doi:10.3847/2041-8213/ad54b3

Illustration of a supermassive black hole accreting gas from its surroundings

How do supermassive black holes get so massive? New simulations show how black holes might have grown rapidly in the early universe.

deep-sky image of galaxies

A deep-sky image from JWST. The inset shows the galaxy JADES-GS-z14-0, which is currently the most distant known galaxy. This image shows how the galaxy looked less than 300 million years after the Big Bang. [NASA, ESA, CSA, STScI, B. Robertson (UC Santa Cruz), B. Johnson (CfA), S. Tacchella (Cambridge), P. Cargile (CfA)]

Black Holes as Far as Our Telescopes Can See

When JWST first examined galaxies in the early universe, it discovered something extraordinary: galaxies just a few hundred million years after the Big Bang are home to black holes, and these black holes are massive.

How these black holes grew to millions or billions of times the mass of the Sun so quickly is an open question, but most theories fall into two general categories: these early black holes either sprouted from stellar-mass pips that bulked up at a prodigious rate, or they got their start as more massive black holes that grew at a more modest rate. Other, more speculative origin stories invoke exotic forms of dark matter or supermassive stars. All options face serious challenges, and many researchers turn to simulations to find a way forward.

Captured by Clusters

Yanlong Shi (California Institute of Technology and University of Toronto) and collaborators used simulations to test a way for young black holes to become supermassive in short order. The simulations began with 100 million solar masses of gas collected in a cloud about 330 light-years across, representative of a dense star-forming clump in an early galaxy.

simulation snapshots showing the different stages of black hole growth

Gas density (colored areas), star locations (cyan dots), and selected black hole locations (black stars) during three stages of the simulated evolution. These images show how the growing black holes are captured by tight star clusters and transported to the center of the cloud. From there, an accretion disk forms and channels gas to the black holes. Click to enlarge. [Shi et al. 2024]

Into this cloud the authors randomly scattered black hole seeds with masses between 100 and 10,000 solar masses. At first, the black hole seeds grew randomly as they encountered and devoured dense clumps of gas. But at some point, the black hole seeds were captured by nearby massive star clusters. From there, the black holes went along for the ride, trapped within the gravitational wells of the star clusters that migrated toward the center of the cloud. The black holes reached the center of the cloud much faster than if they were not shepherded by star clusters.

Once the black holes reached the center of the cloud, their growth kicked into a whole new gear. Now attracted by a deep gravitational potential well, gas from throughout the cloud sank toward the central black holes and swirled together in an accretion disk. The magnetic field threaded throughout the disk prevented the disk from fragmenting and forming new stars. In about a million years, the two fastest-growing black holes had grown to more than 2 million solar masses.

Connecting the Little Red Dots

plots showing the movement and growth of star clusters and black holes from the simulations

Another illustration of the movement and growth of the black holes. The leftmost image shows the capture of the black holes by the star clusters and the motion of these captured black holes to the center of the cloud. The images on the right show the masses and mass ratios of selected star clusters and black holes over time. Click to enlarge. [Shi et al. 2024]

This scenario appears to provide a path forward for the formation of supermassive black holes in the early universe. In particular, it shows how the capture of a black hole by a star cluster rapidly puts the black hole into prime position to accrete large amounts of gas from its surroundings. The magnetically stabilized accretion disk then provides a way for the black holes to accrete faster than the Eddington limit — the theoretical limit beyond which the outward pressure of radiation generated by accretion overpowers the inward pull of gravity.

The team’s simulations may yield a clue to the solution of another cosmic mystery. Near the end of the simulation, the assemblage of young stars and massive black holes has similar properties to the unusual “little red dot” galaxies spotted by JWST just 700 million years after the Big Bang.

While Shi’s team hopes to explore the observational implications of their model results further, the predictions will likely be difficult to test, requiring telescopes to pierce the dense, dusty gas that obscures the inner galactic regions where black holes grow.

Citation

“From Seeds to Supermassive Black Holes: Capture, Growth, Migration, and Pairing in Dense Protobulge Environments,” Yanlong Shi et al 2024 ApJL 969 L31. doi:10.3847/2041-8213/ad5a95

illustration of a flaring magnetar

Astronomers have discovered two fast radio bursts with eerily similar signals. What’s the likeliest cause of these twin bursts?

Mysterious Bursts

Artist's impression of a pulsar

Artist’s impression of a neutron star, the core of a massive star that has exploded as a supernova. [ESO/L. Calçada; CC BY 4.0]

Fast radio bursts are powerful flashes of radio emission that last anywhere from less than a millisecond to a few seconds. Astronomers have detected roughly 1,000 fast radio bursts from outside our galaxy and a single burst from within the Milky Way. Exactly what causes these bursts is up for debate, and supernovae, colliding black holes, and neutron stars — the extremely dense city-size remnants of massive stars — are all in the running.

As researchers attempt to parse the similarities and differences among the growing sample of radio bursts into a coherent theory of their origins, a curious pair of bursts has arrived on the scene. What can we learn from a nearly identical pair of bursts?

plot of normalized flux for two fast radio bursts

The flux profiles of the 2018 (red) and 2021 (blue) bursts. The flux has been normalized to the peak intensity of each burst. Note that the times on the horizontal axis have been normalized to the time between the burst components. [Bera et al. 2024]

Twins from Extragalactic Space

In 2018, the fast radio burst FRB 20181112A arrived at Earth after a journey of billions of light-years, announcing itself with a prominent 0.2-millisecond flash of radio waves and a second, subtler signal just one millisecond later. In 2021, FRB 20210912A was discovered, looking oddly familiar: it too was marked by an initial bright component and a later faint component.

Apurba Bera (Curtin University) and collaborators examined this pair of bursts to understand how deep their similarities go. Using data from the Australian Square Kilometre Array Pathfinder, Bera’s team showed that the relative emission timescales — for example, the ratio of the width of the bright signal to the width of the faint signal — are remarkably similar.

The absolute timescales of the bursts are slightly different, which the team suggests could be due to their different travel times. The 2018 burst arrived from a redshift of z = 0.4755, but the 2021 burst’s redshift is unknown. If the 2021 burst occurred at a redshift of z = 1.35 — possible, given a previous estimate of z = 1.18 ± 0.24 — the absolute timescales would be identical as well.

Possible Origins

normalized flux and position angle for two fast radio bursts

Normalized flux density and polarization position angle (PA) for the 2021 (top) and 2018 (bottom) bursts. The “A” panels show the total intensity (I), the intensity of linearly polarized light (L), and the intensity of circularly polarized light (V). Click to enlarge. [Adapted from Bera et al. 2024]

As for other similarities, both bursts are strongly polarized, meaning that the light waves were oriented in a nearly uniform direction as they traveled through space. In both cases, the polarization direction changed between the first part of the burst and the second.

Bera and coauthors note that the properties of the twin burst pair match what they’d expect for a magnetized neutron star rotating with a period of 1.1 milliseconds. This is faster than the most rapidly rotating neutron star known, and it’s just below the hypothetical limit on neutron star rotation. This may point to the existence of a sub-class of fast radio bursts arising from neutron stars spinning near the maximum rate.

The majority of fast radio bursts appear just once, but a few dozen are known to repeat. If the 2018 and 2021 bursts studied here fall into the latter category, it could help solve the mystery of their origin: the slow but steady spin-down of a rapidly rotating neutron star would leave a detectable mark on the radio signal.

Citation

“The Curious Case of Twin Fast Radio Bursts: Evidence for Neutron Star Origin?” Apurba Bera et al 2024 ApJL 969 L29. doi:10.3847/2041-8213/ad5966

An aerial photograph looking down on two large telescopes.

It seems natural that stars would spin in the same direction as, and nicely aligned with, their planets. Recent work, however, shows that might not be the case if the star in question is particularly hot.

Toppled-Over Stars

When we picture planets moving around their host stars, we usually imagine them as perfect spheres traveling along perfect circles aligned within a perfect plane. Reality, unsurprisingly, is messier than this ideal image. Some objects, like Mars, maneuver on orbits that look more like ovals than circles, while others, most famously Pluto, drift above and below everyone else on orbits that are tipped over relative to the bulk of the solar system. The closer you look, the more jumbled everything appears: while that’s maybe disappointing from an aesthetic point of view, astronomers thrive in this chaos and latch onto every irregularity they can in order to explain how a given system came to be.

An illustration of the many angles used to describe a star’s orientation. The star’s true obliquity is marked by Ψ. [Louden et al. 2024]

One property astronomers love to measure is the angle between a star’s spin axis and the orbits of its planets, also known as its obliquity. Our Sun sets a good example as a mostly-aligned host star: its north pole is tipped over a modest 7 degrees from the rest of the solar system, a slant larger than that of the leaning tower of Pisa but still respectively close to upright. About a decade ago, however, astronomers began to realize that not all stars were as well behaved. While in general stars are more likely to be aligned with their planets’ orbits than not, a good fraction of the hotter stars seemed to be nearly randomly oriented without any care for what’s circling them.

This was an intriguing but difficult to confirm finding, since the traditional ways of measuring obliquity required a tricky procedure and lots of telescope time. Researchers came up with an alternative, faster way to approximate the measurement, but progress was still slow and it was difficult to assemble an unbiased sample of stars. Since it was easier to measure the obliquities of cool stars with large planets, astronomers simply hadn’t attempted to analyze many hot stars with small companions.

New Observations

That’s where Emma Louden (Yale University) and collaborators came in. Over the course of several months, Louden and the team periodically commandeered the High Resolution Echelle Spectrometer (HIRES) instrument on the Keck I telescope to collect spectra of nearly 500 stars. Some of their targets had planets while others did not, though the barren systems were otherwise indistinguishable from their planet-hosting counterparts and formed a crucial control sample for the analysis. Since the team was aiming to fill in gaps of previous catalogs, they went after numerous hot stars and stars with planets smaller than Jupiter. After vetting their candidates and cleaning up their data, they were left with a sample that had 500% more hot stars than any previous attempt, and 50% more cool stars for good measure too.

A summary plot of the findings. Histograms on the right show the projected inclinations of different subsets of the sample, while the scatterplot on the left shows the full sample. Circles outlined in black denote planets around hot stars. Click to enlarge. [Louden et al. 2024]

The team found that the trend noticed years ago does indeed seem to hold up to this stronger statistical pressure test: hot stars, regardless of the size of their planets, tend to be more misaligned than cool stars. Going further, they also took advantage of their larger sample and looked for trends with other properties, finding weak evidence that planet period might also correlate with stellar misalignment. However, even though their sample was larger than any previously assembled, it isn’t large enough to do more than hint at this additional find.

Still, knowing that hot stars seem to misbehave in almost all circumstances does put some pressure on our theories of planetary formation and evolution. While there is still no clear-cut winner, and as the authors note, likely more than one mechanism is at play, we’re beginning to understand what’s more and what’s less likely to have happened to these tipsy, boiling stars and their planets.

Citation

“A Larger Sample Confirms Small Planets around Hot Stars Are Misaligned,” Emma M. Louden et al 2024 ApJL 968 L2. doi:10.3847/2041-8213/ad4b1b

1 … 11 12 13 14 15 … 111