Gravity Spy Project Makes Waves with Citizen Science
Scotty Coughlin and Mike Zevin, project leaders and doctoral students.
Image Credit: Becca Sanchez
LIGO, the Laser Interferometer Gravitational-wave Observatory, first detected gravitational waves in September 2015. The discovery of these waves brought about an influx of valuable data. However, the extremely high sensitivity of LIGO detectors, which measure changes in distance down to 1/10,000th the diameter of a proton, results in data particularly prone to instrumental and environmental sources of noise called “glitches.”
To help address this problem, Northwestern University’s astrophysics center, CIERA (the Center for Interdisciplinary Research and Exploration in Astrophysics), in collaboration with the Zooniverse team at Chicago’s Adler Planetarium,the Image and Video Processing Laboratory at Northwestern, Syracuse University, and Cal State Fullerton, launched a new project called Gravity Spy. Gravity Spy aims to identify the different types of noise fluctuations that afflict the detectors to more easily identify actual gravitational waves in LIGO data.
Gravity Spy works on an online crowdsourcing platform called Zooniverse, engaging thousands of volunteers, dubbed ‘citizen scientists’ in classifying glitches from LIGO data. While crowdsourcing is useful, LIGO produces millions of glitch signals. This huge volume of data necessitated an innovation in the project’s design. The researchers leverage the work of citizen scientists to help train machine-learning algorithms, which can classify many more glitches. The better the computer is at classifying glitches, the more it can recognize and feed new glitches to human volunteers.
Gravity Spy is funded by the National Science Foundation and led by Northwestern Department of Physics & Astronomy doctoral students Mike Zevin and Scotty Coughlin. “As good as our computer algorithms are, people still have the edge as far as slight morphological differences. Humans are really good at spotting discrepancies so it fits naturally into this crowdsourcing idea,” says Zevin.
Gravity Spy does not upload a static set of data for volunteers to work with. Since the LIGO detectors run constantly, for weeks at a time, the data feed is continually updated. A goal of the project, and one of its biggest challenges, is to create a “live” system with a two-day lag that provides a constant feedback loop: when a user classifies an image, it updates the infrastructure of Gravity Spy.
Coughlin notes that challenges like this, and the interconnectedness of gravitational-wave physics, machine learning, and social science, one some the most exciting parts of the project. “It’s new. The fact that this was a new project with a brand-new take on how to do this certain set of three science disciplines in a new way, has allowed us to not have to work within any boxes or pre-existing infrastructure.”
Because this project involves not only astronomy, but advanced computer science and online learning techniques, Gravity Spy involves an interdisciplinary group of researchers, beyond LIGO scientists. The relationship between the machine-learning algorithms and citizen scientists opens a fertile area for social science and learning experiments. The project includes experts in LIGO detectors from Cal State Fullerton in addition to Northwestern, a team of social scientists from Syracuse University, collaborators at Northwestern in the Image and Video Processing Laboratory who specialize in machine learning, and support from Zooniverse web developers at the Adler Planetarium.
So far, Gravity Spy has been very successful and the future looks bright. Zevin notes that this project is “one of the most successful launches of a Zooniverse project created with a new lab setting. We have over a million classifications already.” More than that, the team has “set the stage for bringing future projects into this realm of dealing with big data and pairing computer algorithms with human classifications.”
The initial concern of the project was whether citizen scientists would be interested if they were looking through the debris of the data. But the passion citizen scientists have shown has been hugely encouraging.
“The huge response that we’ve had and the motivated volunteers that we’ve attracted is really a testament that people from the lay public care about scientific progress and it’s not just about being the person that makes the click to make the discovery, they actually care about what we’re doing and the things that really need to be done to make our job and to make science better in the long run,” says Zevin.
- Written by Becca Sanchez
The project reported in this story is funded by the National Science Foundation under grant 1547880. Any opinions, findings and conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
Gravity Spy is a collaboration led by Northwestern University (Michael Zevin and Scott Coughlin, with PI Vicky Kalogera and co-PI Aggelos Katsaggelos). Northwestern’s team is comprised of a LIGO group in CIERA and a machine learning group in the Image and Video Processing Laboratory. The collaboration also includes The Adler Planetarium Zooniverse team (led by co-PI Laura Trouille), Syracuse University, (led by co-PI’s Kevin Crowston and Carsten Østerlund), California State University at Fullerton (led by co-PI Josh Smith), and The LIGO Scientific Collaboration.
Explosive New Thinking on Stellar Evolution
No X-rays from SN 2014J
Image Credit: NASA / CXC / SAO / R. Margutti et al.
Until a few years ago, the accepted leading theory of stellar explosions stated that until a star explodes, its lifetime path to explosion is relatively uneventful. But the events behind Supernova 2009ip shook that theory when the star suffered a number of eruptions in 2010 and 2011—proving it survived the eruption episodes of 2009—and then experienced an unusual double brightening in 2012. These events led observer Raffaella Margutti to reconsider the final stages of evolution in massive stars.
Dr. Margutti is an Assistant Professor of Physics & Astronomy and a faculty member within CIERA at Northwestern. She studies and models transient astrophysical phenomena including stellar explosions and stellar tidal disruptions. By chance, a team of astronomers that included Margutti was observing what was thought to be a normal supernova explosion when SN 2009ip exploded again.
“This changes all our understanding of stellar evolution,” explains Margutti. The new information produced by SN 2009ip led her to research stellar explosions in terms of understanding how stars explode, why they explode at that particular time of their history, and which kinds of stars produce which kinds of explosions.
Margutti’s research is unique in that her group works with “broad-band observations” from all over the spectrum, from x-rays to gamma rays, “all the way down to radio”—an inclusive approach that researchers have not been using, as they typically specialize in one key field, like radio or optical astronomy.
The supernova community is still segregated into these wavelength domains and “up until now did not really speak to one another” or even use the same units, rendering a cohesive understanding more difficult to achieve. Margutti is bringing a wide scope of understanding to the supernova field.
To do this, she uses a variety of observatories to collect photonic data across the spectrum. They include an entire suite of NASA satellites and ASASSN (All-Sky Automated Survey for Supernovae) telescopes, as well as an orchestra of optical telescopes for spectroscopy and radio antennae. Considering that physical phenomena contribute to different parts of the spectrum, this range of data allows Margutti to better constrain the properties of the explosions, such as their environments and their progenitor stars. Ultimately, this variety of information can provide a more holistic understanding of the relationship between mass and environment to understand stellar evolution.
The next step to include in the investigation is non-photonic mass messengers like neutrinos and gravitational waves with the collaboration of CIERA Director, Vicky Kalogera. Dr. Kalogera’s research concentrates on information that can be extracted from gravitational wave emissions produced from events that are tightly connected with stellar explosions and mergers of compact objects, like neutron stars and black holes. Specifically, electromagnetic signals expected to accompany these emissions can potentially unveil the precise location of the event and thus a number of important properties, like astrophysical environment. Margutti notes that this collaboration puts Northwestern in a unique position to maximize the scientific reward from the “new era of gravitational wave astronomy.”
There are two theories Margutti proposes. One is the existence of processes in the interior of the star that create instabilities. The second is that these massive stars that explode are not alone but have companions whose presence creates instability.
“The bottom line is our understanding of stellar evolution is not exactly right. These are very different predictions in terms of how this behavior should show up so I am trying to test that…if one of the two theories is right.”
- Written by Becca Sanchez
This research has been supported by the National Aeronautics and Space Administration (NASA), the Smithsonian Astrophysical Observatory (SAO), and the Research Corporation for Science Advancement (RCSA).
Black Holes May be Born in Dense Star Clusters, Explain LIGO Observations
Black hole mosh pit.
In this simulation, 60 black holes and 500 stars interact with each other at the
chaotic core of
a globular cluster until two black holes combine to form a black hole binary.
Credit: Carl Rodriguez/Northwestern Visualization (Justin Muir, Matt McCrory, Michael Lannum)
In early 2016, the National Science Foundation and the LIGO Scientific Collaboration announced that their scientists had successfully, for the first time, directly detected gravitational waves - or ripples in the fabric of spacetime - using aLIGO (Advanced Laser Interferometer Gravitational-wave Observatory). These waves were emitted by a pair of black holes merging a billion years ago, and because mergers like these emit no light of any kind, finding them had previously been impossible. Thanks to aLIGO, scientists have an exciting and important new way to observe the universe.
Merging black holes, like those observed in aLIGO, require two black holes to be orbiting each other, in a binary black hole system. Researchers, like those here at CIERA, at Northwestern University, are asking big questions about the formation of these systems and the potential number of future findings by aLIGO.
Professor Fred Rasio and graduate student Carl Rodriguez, along with their collaborators, had already been working on exactly these questions, before the historic gravitational-wave announcement. They are studying huge, spherical collections of a million densely-packed stars called globular clusters. They’ve developed 52 computer models to demonstrate how a globular cluster acts as a dominant source of binary black holes, producing hundreds of black hole mergers over a cluster’s 12-billion-year lifetime. Their most massive simulation required 30,000 hours of computing power.
Even before the aLIGO detection, Rasio’s group had seen that binary black holes can be born in this chaotic ‘mosh pit’ of stars and black holes. This process is called dynamical formation theory. It’s one of two recognized main channels for forming the binary black holes detected by aLIGO.
Looking at all of their models, the team estimated that aLIGO could detect perhaps 100 merging binary black holes per year, from the cores of globular clusters. This is five times the rate that was previously predicted. This process of forming binary black holes within globular clusters could be a major source of merging black holes in aLIGO’s future, and a growing population of black holes will help astrophysicists learn more about the universe. “By the end of the decade, we expect LIGO to detect hundreds to thousands of binary black holes,” Rodriguez says.
This prediction is consistent with aLIGO’s observations of black holes merging, over the past year. “Thanks to LIGO, we’re not just theorists speculating anymore -- now we have data,” says Rasio. “A relatively simple and well understood process seems to work. Simple freshman physics -- Newton’s first law of motion -- explains the gravitational dynamics of the first black holes detected by LIGO.”
This research has been supported by the National Science Foundation and NASA. The Northwestern Visualization team, which works with CIERA to produce movies like the one shown in this article, is part of Northwestern University Information Technology (NUIT). Star cluster models were developed through a CIERA-supported interdisciplinary collaboration between Northwestern’s Physics and Astronomy department and the Electrical Engineering and Computer Science department at McCormick.
New Mirror Technology for the Next Generation of Space Telescopes
One of the Holy Grails of Space Astronomy is to build membrane mirrors of the quality of the Hubble Space Telescope mirrors that deploy in space. The ability to do so will mean that space telescopes with mirrors 10 times the diameter of Hubble or even larger would be within the grasp of the astronomical community. However, current technology doesn’t work well enough to provide Hubble-quality images. Thus in response to this obvious need, CIERA Professor Mel Ulmer lead a team that in June 2015 received a NASA Innovative Advanced Concepts award to serve as the Principal Investigator on the APERTURE project. APERTURE stands for “A Precise Extremely-large Reflective Telescope Using Re-configurable Elements.”
The concept is based on the idea that to make the deployed membrane design work, the membrane will be coated with a “magnetic smart material,” or MSM. These kinds of materials are specially designed to contract or expand very strongly when immersed in a magnetic field.
All ferromagnetic materials either contract or expand in magnetic fields, which is why the standing joke in the MSM community is “you have all heard of this effect… which makes transformers hum.” MSMs respond approximately 10-1,000 times more strongly than ordinary ferromagnetic materials.
When the telescope is in the rocket, the mirror membrane will be folded like an umbrella. This allows a much larger mirror to fit within the rocket, although of course the mirror then has to be unfolded, once in space. After the mirror unfolds, a curved arm (or arms) with a strong magnet (producing an approximately 1,000 times stronger magnetic field than the Earth’s surface magnetic field) on it rotates around the back of the membrane, moving to all of the locations where the mirror surface needs to be corrected. By precisely locating the magnetic field strength and direction of the magnetic field, the magnet (or magnets) will correct the shape of the mirror to produce high quality images. Though the concept of using a membrane is not new, the idea of applying the MSM plus magnetics to a deployable space mirror has never been explored before.
If this project is successful, this mirror design may be used for the next generation space telescope, such as “ATLAST”, an optical space telescope of 16 meters mirror in diameter to image exo-planets as well as the most distant galaxies in the Universe. This technology would also be a boon for Earth space observations so as to allow the image quality from geo-stationary orbit some 20,000 miles above the Earth vs. current non-stationary 400 miles orbiting space telescopes.
A project like this requires significant engineering expertise, which is why APERTURE is a multidisciplinary project, employing astrophysicists, material scientists, and engineers. Professor Ulmer is no stranger to combining these fields. He has worked on developing similar new mirrors for future X-ray telescopes, as well. That work has also attracted NASA funding related to building the next generation X-ray space telescope. The next generation X-ray telescope would ideally have 30 times the area of the current Chandra X-ray observatory. Chandra is the X-ray astronomer’s equivalent to Hubble. Ulmer and his research group at Northwestern are investigating using MSM-based-light-weight X-ray mirrors for the next generation X-ray telescope, since even if the Chandra mirrors were “for free”, launching 30 Chandra is not feasible.
This research has been supported by the National Aeronautics and Space Administration (NASA), through NASA's Innovative Advanced Concept (NIAC) Program. Previous work on magnetic smart materials was sponsored by NASA's Astrophysics Research and Analysis (APRA) Program as well as the Weinberg College of Arts and Sciences at Northwestern University.
At The Center of the Galaxy, A Dark Mystery Unfolds
This view shows several of the ALMA antennas and the central regions of the Milky Way above.
The image was taken during the ESO Ultra HD (UHD) Expedition.
Credit: ESO/B. Tafreshi (twanight.org)
At the center of the Galaxy lies a supermassive black hole - a condensed mass so dense that no light can escape, and so gravitationally powerful that nearby objects are at risk of being pulled in. Despite these extreme conditions, many stars and their neighbors manage to stay intact; some are even born in these treacherous waters.
Star formation is already one of the greatest mysteries of astronomy, made only more difficult to understand by the presence of the black hole. How do these stars form, and how do they manage to survive?
Professor Farhad Yusef-Zadeh is one of the CIERA researchers beginning to answer these questions. Yusef-Zadeh tries to understand the conditions that allow star formation at the nucleus of the Milky Way galaxy.
Star formation is poorly understood due to observational limitations, but it is thought that they are formed when stray clouds of helium and hydrogen collapse in on themselves due to gravity and other factors.
Since the supermassive black hole has such a strong gravitational force, it creates hostile conditions to star formation: the stray gas is attracted to the black hole rather than other gas particles, thereby stretching the cloud and preventing collapse.
However, Yusef-Zadeh has discovered that there are some stars that manage to form despite these conditions. But the question remains: how is that possible?
A recent study indicates that despite the strong tidal force by the massive black hole at the Galactic center, the external pressure in this dense region contribute substantially in the collapse of a cloud, thus forming stars. In addition, the formation of low-mass stars appears to be easier than high mass stars near the strong gravitational potential of the black hole. This is because of the criteria for the collapse of a cloud, under Roch and Jeans limits, depends on the distance from the massive black hole. Thus, the collapse of low-mass cloud is favored.
Yusef-Zadeh also studies the lives of older stars at the center of the galaxy, and how their orbits are affected by the supermassive black hole. The key to this lies with studying radio signals emitted by the black hole and its celestial neighbors.
“[Radio astronomy] is very accurate because radio astronomers have details about Earth’s rotation and what the climate is doing on a weekly and monthly basis,” he said. “All of these dozens of changes go into finding the absolute position of objects in the sky.”
Tracking these orbits by studying the radio signals they admit allow researchers to understand how supermassive black holes alter the fabric of spacetime. Spacetime is a mathematical model that joins the three dimensions of space with the fourth dimension of time as one continuous “fabric” that can be warped by the mass of the objects in it. Dense objects in space cause curvatures in this fabric, like marbles on a taut sheet. The curvature caused by an object as massive as our supermassive black hole dramatically alters the paths that less massive objects take in their orbit around it. Variations in these orbits are often attributed to changes in the spacetime curvature. Therefore, irregular orbits can signal the presence of black holes in the galaxy.
Yusef-Zadeh’s research is just scratching the surface of what’s really going on at the center of the galaxy. However, it still sheds new light on star formation by exploring the role that environment plays in the process.
“The effect of environment on star formation is a big topic, and over the years hasn’t really been established,” he said. “It’s a process we’re continuing working on it and continuing the same research on a different level.”
This research has been supported by the National Science Foundation, and utilized data from the National Radio Astronomy Observatory (NRAO), which is a facility of the National Science Foundation, operated by Associated Universities, Inc (AUI). This work also utilized data from ALMA (the Atacama Large Millimeter Array), which is operated by the European Southern Observatory, AUI/NRAO, and the National Astronomical Observatory of Japan.
Unraveling the Forces at Play in Early Galaxy Formation
Click image to view simulation video.
In its infancy, just after the Big Bang, the universe looked radically different than it does today. It housed merely a homogenous mixture of gases floating in space. Today, planets, stars, and galaxies have formed from these building blocks. However, there is a huge knowledge gap in understanding exactly how the homogenous post-Big Bang universe evolved into the exciting ecosystem of celestial bodies that we are familiar with today.
In order to understand how planets and stars formed, it is crucial to understand the original formation of the galaxies that contain them. CIERA-affiliated researcher Claude-André Faucher-Giguère is doing just that, via the study of a field called cosmological hydrodynamics. Faucher-Giguère uses simulations of the early universe to investigate a variety of forces that may have been at play.
One of the most important forces is gravity, which is responsible for much of the activity that takes place in the universe. Solutions to hydrodynamic equations - theoretical models that explain how gases interact - are also used. Together, they paint a picture of the early universe. More importantly, those solutions show that there are small fluctuations in the density of gas after the Big Bang. These same fluctuations came together to form the first galaxies.
While gravity played an important role in the formation of these dense proto-galaxies, feedback loops are also important. As massive stars are born and die over time, they stir the gas in the galaxy, and ionize the gas around them. This ionized gas is released back into the interstellar medium, free to begin the process anew. Its presence in different concentrations may explain the different properties and compositions of galaxies. Some of that ionized gas may escape the galaxy. In that way, the death of stars may help limit the number of stars that are born in the galaxy, in the future. In other words, the gas flows that result from the life cycles of these stars regulate future star formation in these galaxies.
Not all galaxies are created equal. The largest galaxies are often the most luminous, and appear red to the naked eye, while the smaller ones are blue and less luminous. In the future, Faucher-Giguère hopes that these modeling techniques will begin to account for these discrepancies, as well as understand the original structures in the universe.
“There have to be underlying physical explanations,” he said.
Learn more about how Professor Faucher-Giguère and his group use high-performance computers at Northwestern to study the evolution of galaxies.
This research has been supported by the Miller Institute for Basic Research in Science, by NASA through an Astrophysical Theory Grant and an Einstein Postdoctoral Fellowship Award, by the National Science Foundation, and by Northwestern University. Computer simulations used the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by the National Science Foundation.
Shapes of interstellar clouds yield clues to violent stellar explosions
Source: NASA, ESA, and the Hubble Heritage (STScI/AURA)-ESA/Hubble Collaboration; J. Hester (Arizona State University)
In a star field, the eye is often drawn to the superstars of the night sky: stars, planets, maybe galaxies. But what about the space between the stars? That seemingly empty space turns out to be just as busy -- and just as interesting -- as the bright stars that surround it. It contains the material that stars and planets are made from; and, in a case of cosmic recycling, that material comes from gas that stars throw off as they die. Astronomers call this material ‘The Interstellar Medium.’
Most of the interstellar medium consists of diffuse clouds and dust, making it easy to view the stars that they screen. Star formation occurs in the densest clouds, known as molecular clouds, that are dusty enough to block the light from background stars. The structure of these dense
clouds has been well studied, but the structure of the colder, diffuse interstellar medium clouds is not as well understood.
Obtaining these findings was no easy task: these clouds are difficult to see, as they’re made of cold gas that doesn’t emit much light, and therefore cannot be seen in the infrared spectrum. Meyer was able to obtain this information by performing optical spectroscopy of the light filtered through these clouds. UV light is high-energy light emitted by stars. When filtered through a spectrograph, this light is broken down into different wavelengths, allowing for in-depth study.
By comparing the strength of the interstellar sodium absorption toward the same stars in multiple spectra taken years apart, Meyer was able to determine that some of these cold diffuse gas clouds exhibit sheet-like structure as thin as the size of the solar system. If the clouds were thicker in size, the sodium absorption wouldn't change on timescales of years as the intervening clouds moved with respect to the targeted stars across the observed sightlines.
Such thin structures in the diffuse ISM had previously been considered rare by the scientific community, making Meyer’s findings particularly surprising. One explanation is that they are produced by colliding high-velocity gas flows driven by stellar explosions and/or stellar winds. Now that these diffuse ISM structures have been discovered, further study may provide new insight into the formation of molecular cloud structure and the process of star formation. In the future, Meyer hopes to better examine the physical character of the thin diffuse ISM structure through ultraviolet Hubble Space Telescope spectroscopy of other atoms.
This work used data taken over 10 years with the Kitt Peak National Observatory (KPNO) Coude Feed telescope. KPNO is part of the National Optical Astronomy Observatory (NOAO), which is operated by the Association of Universities for Research in Astronomy (AURA), under cooperative agreement with the National Science Foundation.
Gravitational Waves: Discovering a Revolutionary View of Black Holes and the Universe
(Updated March 2016)
Electromagnetic radiation, a name for “light” that spans radio waves, heat, optical light, and Xrays, is one of the tools that astronomers use to learn about the universe. A completely different type of radiation, called gravitational waves, is revolutionizing our view of the universe. Gravitational waves are created when objects, such as black holes, orbit each other. This creates ripples in the fabric of spacetime, a concept that joins the three dimensions of space with the fourth dimension of time as one continuous “fabric” that can be warped by the mass of the objects in it.
According to some theorists, gravitational waves can tell scientists about the existence and origin of compact objects. They may even reveal clues about the origin of the universe. Until recently, gravitational waves have been virtually impossible to detect with current technology.
CIERA Director Vicky Kalogera is part of the team that closed this technological gap. She is one of the researchers on an international project to develop LIGO, or Laser Interferometer Gravitational Wave Observatory, a new kind of “telescope” that can observe and record gravitational waves. On February 11, 2016 the National Science Foundation and the LIGO Scientific Collaboration announced that their scientists successfully, for the first time, directly detected gravitational waves - or ripples in the fabric of spacetime - using LIGO.
Kalogera works on modeling for the NSF-funded LIGO project. Specifically, she models pairs of compact objects: as mentioned above, one of the ways that compact objects can move (and thus create gravitational waves) is if they are part of a binary system, where the two compact objects orbit each other.
Binary systems with two compact objects are difficult to observe, so models become important for predicting how they can produce gravitational waves. Kalogera’s models of these binary systems allow the scientists developing LIGO to determine how sensitive the instrument will need to be, to see those systems.
The key to observing the gravitational waves produced by a compact object system is detector sensitivity. The LIGO detectors that came online in 2015 had a sensing range that reached 600 million light-years into the galaxy.
“If the [wave] source is common and there’s billions of the sources, then the chance of having something close is high,” she said. “If sources are rare because nature has a hard time forming them, the instrument has to be very sensitive because it has to see distant sources to catch a few of the rare emissions.”
Now that gravitational waves have been detected using this new telescope, astronomers are able to test Einstein’s Theory of Relativity, the theory that predicts the existence of spacetime. But, more importantly, astronomers can study the universe in a way that was previously impossible. Gravitational waves will allow researchers to “see” black holes for the first time, and accurately constrain their location and movement.
“This would be a new way of studying astrophysics in the universe and make measurements you would never be able to do before,” Kalogera said.
The Laser Interferometer Gravitational Wave Observatory (LIGO) project is funded by the National Science Foundation (NSF); the research above has also been funded by the NSF. The Northwestern LIGO group contributes to, and is part of, the LIGO Scientific Collaboration (LSC).
Magnetic Fields May Hold Key To Star Formation (December 2014)
Star formation is one of the greatest mysteries in astrophysics. Due to the limits of modern technology, most of the understanding of the mechanisms is purely theoretical, and decisive observations are few and far between.
One of the difficulties when examining star formation is understanding the rate at which it occurs. Stars form when gas clouds collapse in on each other, a process driven by gravity. However, models show that if gravity was the only force at play, star formation would produce many more newborn stars than are actually seen.
CIERA-affiliated professor Giles Novak is pushing the limits of what technology can do to get observations that can fill the gaps left by theoretical astronomy. Novak’s research explores the key role magnetic fields may play in the rate of star formation.
Professor Giles Novak and collaborators install the primary mirror onto the Balloon-borne Large Aperture
Telescope (BLAST) as one of the early steps in BLAST's pre-launch assembly in 2012.
Every star, every planet, and even the Galaxy itself has a magnetic field, similar to the one that exists on Earth. Models show that the magnetic field of the surrounding Galaxy interacts with gas clouds that collapse to create stars, creating tensional forces that may play a role in star formation.
Initial observations from Novak’s experiments show that the magnetic fields align themselves in opposition to the collapse. The tensional forces actually slow the collapse of the clouds, allowing for far fewer stars formed than would otherwise. These magnetic fields are partially responsible for the structure and makeup of our galaxy today.
The cutting-edge technology used in Novak’s projects are responsible for these observations. An international collaboration called BLAST uses a balloon-borne telescope to get above the atmosphere and record data with minimal interference. The telescope uses hundreds of pixels to record thousands of vectors, which indicate the direction of the magnetic field around star nurseries. Previously, only a few dozen vectors could be recorded; now, larger numbers of vectors show a coherent pattern around the star nursery.
The telescope determines the position of these vectors by recording the polarization of submillimeter light emitted by dust particles in space, which can then be used to infer magnetic fields.
Submillimeter light is low-energy infrared radiation that is emitted by many low-temperature objects, including the tiny particles of heavy-element dust “grains” that float and spin in space. The rotation axis of these spinning grains is aligned with the magnetic fields, like tiny compass needles in space. The angle of this rotation can be observed in the polarization, or the “tilt”, of submillimeter light reaching the telescopes.
Polarization is a property of waves that oscillate in more than one direction - as do light waves. The “tilt” of these light waves matches that of the rotation of the dust. As the magnetic fields are warped by star formation, the polarization of submillimeter light changes in accordance with the rotation axis of the dust, allowing researchers to get a clear picture of what is happening as the star forms.
A polarization map of a nearby gaseous molecular cloud (“OMC 3”): the blue and red colors in the background show where the
is concentrated, and the black lines show the polarization of light from throughout the cloud (Chapman et al, in preparation).
As technology continues to improve, Novak will be able to expand these discoveries: the next generation of BLAST will use thousands of pixels, allowing the telescope to record half a million vectors.
The BLAST project is supported by NASA, the Leverhulme Trust, and Canada's Natural Sciences and Engineering Research Council (NSERC), as well as the National Science Foundation's Office of Polar Programs.
Foreign Worlds Come into Focus, Thanks to New Findings (July 2014)
Three planets in the Kepler-11 system as they simultaneously transit their star as imagined by by a NASA artist (Image credit: NASA)
Until recently, the only planets astronomers knew of were those in our own solar system. But in the past twenty years, thousands of “exoplanets” have been discovered, with NASA’s Kepler telescope team announcing the discovery of over 800 of these planets just this past January.
Astronomers have been finding exoplanets in solar systems that are very different from our own. Though many such planets have been found, much remains unknown about them, including their weight and chemical composition, both of which are key to understanding how they formed.
Yoram Lithwick is an Assistant Professor of Physics and Astronomy at Northwestern whose research focuses on exoplanets. His research has taken him from Toronto to California and finally to CIERA, with the goal of answering what he calls the “fundamental” questions of planet formation.
“[Exoplanets] is an explosive field. There’s suddenly a lot of data and many new puzzles,” he said. “It’s expanding our knowledge of the universe.”
Lithwick’s research has uncovered some important information about what these exoplanets actually look like, which may hold the key to understanding their creation. A study published in January by Professor Lithwick and graduate student Sam Hadden determined the densities, and therefore compositions, of 60 exoplanets larger than Earth and smaller than Neptune.
The chemical compositions of these planets allow researchers to better understand what they look like - and many are radically different than those of our solar system. Exoplanets that are roughly comparable to Earth in size are dense and rocky, but those twice as large are much more gaseous - leading Lithwick to conclude that they have a large hydrogen atmosphere.
Lithwick uses observational data called transit time variations to infer the properties of these exoplanets.
Transit time is the time at which an exoplanet passes in front of its star during its orbit. The light from the star dims during a transit event, allowing researchers to record it. The gravitational pull between two planets orbiting the same star can create slight variations in the transit times usually observed by each planet.
“If there’s a single planet going around a star, it would transit every, say, 10 days, assuming it takes 10 days to go around the star," Lithwick explained. “But if there are two planets, one of them ‘tugs’ on the other, and the transit time could be 10 days and 2 minutes or 10 days minus 10 minutes.”
These variations in transit time allow scientists like Lithwick to learn more about their mass and density. For instance, a more massive, denser planet will “tug” more strongly, leading to a larger variation in transit time.
Transit time variations also allow for greater understanding of the planets' orbits, which can lead to learning more about the origin and history of these distant systems.
The understanding of the chemical composition of exoplanets allows for a coherent understanding of what these distant worlds look like, and helps unravel the mystery of their origin. However, with the discovery of so many planets, the new data has created a wealth of new information and challenges that Lithwick and his group have only just begun to tap into with their new findings.
"There is a hope that in the near future we will finally understand how planets form, " he said. “All this new data has overturned most of what we thought we understood. I think this is a big opportunity.”
Two 16 Year-Old Scientists Challenge Recent NAS Article Findings. (December 2013)
An October PNAS article on exoplanets by astronomers at Berkeley and Hawaii has received a large amount of popular coverage, popping up in such unlikely places as CNN and Fox news. The study projects a “22% occurrence rate of Earth-size planets in habitable zones of Sun-like stars.” Last June, however, high school students of CIERA’s Dr. Laura Trouille performed a similar study with differing results.
Rebekah and Jennifer Kahn, homeschooled twin sisters in the Computational Astrophysics course of Northwestern’s Center for Talent Development, estimated 10% in answering the same question. The girls explain the difference saying, “Petigura et al. used different defining parameters. We think ours are more realistic.” In particular they point to differences in Habitable Zone (HZ) calculations. “The PNAS article uses a non-standard HZ of .5 to 2.0 AU, while we use Kopparapu’s 2013 study of .99 to 1.7 AU; although they do note in a table that, if they had used Kopparapu, they would have gotten only 8.6% as their estimate – which is a lot closer to ours.” Additional differences include the PNAS article analyzing only G and K type stars, while the Kahn twins included F types as well. “It seems that if you are not going to just use Sol-like G types, then you should probably use both similar cooler and hotter (F and K type), not just similar cooler (K type) like they did.”
Despite their different results, the 16-year-old astronomers note that they used similar methods. “We used the same current NASA data, primarily from Kepler, and developed some very similar scatter diagrams.” The students generated their results using their own Python based programs, a language they learned in the course. “We are really thrilled not only to have been able to do some ‘real’ computational astrophysics, but also to find that ‘real’ astrophysicists are working on the same problem as well.”
The student study was presented in a 35 minute video, with the first half providing background information for high school or beginning college students, while the second half presents their computational analysis.
-Written by Rebekah Kahn, Jennifer Kahn, and Laura Trouille
Looking for Oceans on Exoplanets --- Finding Snow. (July 2012)
Astronomers are keen to figure out how to detect oceans on exoplanets, because biologists assure us that liquid water is necessary for life as we know it. The challenge is that exoplanets are very far away, so at best they appear as Carl Sagan's proverbial "pale blue dot." (At this point it's important to note that the blueness of Earth, when seen from far away, has nothing to do with its oceans and everything to do with Rayleigh scattering in our atmosphere. In other words, Earth looks blue from far away for the same reason that the sky looks blue to us.)
One of the proposed methods to tell whether a given pale blue dot harbors oceans is to look for glint. Glint is that really bright shiny spot on the water when the Sun is low in the sky. Water can do this because it is much smoother than land, trees or snow. Even though the glint spot is small, it makes a planet look abnormally bright at crescent phase. Therefore, if you could keep track of a planet's brightness as it orbits its host star, you might be able to infer the presence of a glinting ocean.
That's where our study comes in. We discovered something else that makes a planet look bright at crescent phase. The crux is that the light we see from a planet at crescent phase is hitting the planet at a glancing angle. What kinds of places receive glancing sunlight? You might be catching a region right at dusk or dawn, but more likely you are seeing a cold place, since glancing sunlight is precisely what makes a place cold. Low temperatures mean the surface is more likely to be covered in snow and ice. Since snow and ice are very reflective, the net effect is that the planet looks abnormally bright at crescent phases, regardless of whether it has oceans! Bummer.
But there may still be hope! The saving grace is that there is a way to roughly map out which parts of a planet are more reflective and which regions we are seeing at any given point in time. This sort of "exo-cartography" is possible because as a planet rotates, different surface features come in and out of view, slightly changing the brightness and color of the pale blue dot. And as the planet orbits its star, different latitudes are illuminated. This is what causes our seasons, and for exoplanets it may allow us make a coarse 2D map of the planet. With such a map in hand it might be possible to properly interpret the brightening of the planet at crescent phases. In other words, we might be able to tell whether it really has liquid water oceans, or simply has snow and ice in its cold regions.
A Mass Transfer Origin for Blue Stragglers in NGC 188 as Revealed by Half Solar Mass Companions (October 2011)
Blue stragglers are stars observed to be brighter and bluer than normal main sequence stars of a similar mass and age, and therefore long ago should have evolved to become giant stars and stellar remnants. Since the discovery of blue stragglers, some 60 years ago in the globular cluster M3 by Allan Sandage, blue stragglers have been observed in essentially every stellar population where astronomers have looked, from the very dense cores of globular clusters, to the moderately dense open star clusters, to the sparse regions of the Galactic field. Astronomers have been debating the origins of blue stragglers within these various environments since their discovery. The currently accepted theories include formation through collisions, mergers and mass transfer.
Stellar collisions can occur in the dense regions of star clusters when two stars physically run into each other, and combine to form one more massive star, that would then be observed as a blue straggler. It is most likely that such collisions would occur as a result of dynamical encounters involving binary stars, given their larger physical cross sections for interactions. Therefore a blue straggler created in a collision may retain a companion that participated in the encounter.
Mergers are similar to collisions in that two stars come together to form one blue straggler. However in this scenario the stars begin bound in a close binary system. In fact, the two stars are so close that, given an angular momentum loss mechanism (like magnetic braking), the two stars eventually touch and spiral in to merge. This particular scenario would produce a single blue straggler. Recently this mechanism has been discussed in the context of a triple star system where Kozai cycles drive the inner binary to high eccentricity, and tidal dissipation during close pericenter passages shrinks the orbit to induce a merger. In this new scenario, the original tertiary star would remain bound to the blue straggler as a binary companion.
Mass transfer also occurs within a binary system, but in this scenario the stars do not merge. Instead one star evolves to become a giant and then overfills its Roche Lobe. The giant transfers mass from its envelope to the main sequence star, which accretes the material and eventually becomes a blue straggler. Once the giant has donated its entire envelope, all that remains as a companion to the blue straggler is a white dwarf (the remnant core of the giant star donor), which is predicted to be about 0.5 solar masses.
Lindheimer Postdoctoral Fellow Aaron Geller, and collaborator Robert Mathieu from the University of Wisconsin - Madison, tested these formation hypotheses against their observations of the blue stragglers in the old (7 Gyr) open cluster NGC 188 in a recent letter in the journal Nature. The NGC 188 blue stragglers have a remarkably high binary frequency, and nearly all of the companions to the blue stragglers orbit with periods near 1000 days (as presented in their previous Nature letter). Importantly, Geller & Mathieu also find that these long-period blue straggler binaries in NGC 188 all have companions of about half a solar mass.
Through comparisons of these detailed observations to blue stragglers created within a sophisticated N-body model of NGC 188, Geller & Mathieu conclusively rule out an origin in collisions for these long-period blue straggler binaries. Blue stragglers formed by collisions, which remain in binaries, have significantly higher-mass companions and significantly higher eccentricities than are observed. Mergers in hierarchical triples are marginally permitted by the observations, but the data do not favor this hypothesis. Instead the data are closely consistent with a mass transfer origin for the long-period blue straggler binaries in NGC 188, in which the companions would be half-solar-mass white dwarfs.
This is the first time that the origins of nearly every blue straggler in a star cluster have been determined, and the mass transfer mechanism has emerged as a compelling candidate. Geller and collaborators aim to confirm this result through forthcoming Hubble Space Telescope observations aimed at directly detecting the ultraviolet flux from the white dwarf companions predicted by the mass transfer mechanism.
X-ray astronomy, like many other fields in astronomy, is all about good mirrors. Modern technology has enabled the production of some really powerful mirrors, making enormous strides in the capacities of X-ray observation. Sounds good, right? Well, the problem isn't in the quality of the mirrors anymore—it's in the cost. The super mirror system atop NASA's Chandra X-ray Observatory, for example, would cost almost one billion dollars to reproduce. So the major goal in X-ray astronomy now is to significantly reduce costs while still achieving the exquisite angular resolution of systems like Chandra. Professor Melville Ulmer and his team have set out to do exactly that.
Professor Ulmer, along with Mechanical Engineering Professor Jian Cao, Material Science faculty Michael E. Grahm and Semyon Vaynman, graduate students Xiaoli Wang and Rui Zhou, and undergraduate Bridget Bellavia, is investigating the potential use of a magnetic smart material (MSM) applied as a coating around traditional mirrors in order to enhance their angular resolution. The idea is that the patented material, called "KelvinAll®," is applied in an incredibly thin coating around a mirror—a coating only one-two microns thick (the width of an average human hair is approximately fifty microns). This process is extremely slow and delicate; the application of a two micron coating takes over six hours to complete because the KelvinAll® is laid out atom by atom.
In the presence of the magnetic field (which is created in combination with the mirror), KelvinAll® expands and contracts approximately 1,000 times more strongly than ordinary magnetic materials would—an effect we all know well from the low humming noise often produced by transformers, such as those in TVs, when the iron inside expands and contracts. Programmed expansion and contraction produces stresses, which when strategically implanted into the mirror with a magnetic write-head, will cause the shape of the mirror to be favorably altered—making it more powerful.
So far this process has only been tried in very small areas. The plan is to slowly increase the size of the trials as the precise correlation of stress and coating is better understood and the desired effect more readily predictable. This process has been studied before, but never applied to x-ray mirrors, placing this project at the cutting edge of research in the field. Professor Ulmer and his team could be on track to make momentous progress not only in x-ray astronomy observation, but also in medicine, as this technology could prove very useful in harnessing next-generation x-ray beams that are capable of exposing the atoms in proteins and opening the doors to all kinds of revolutionary drugs and medicines.
The Local Leo Cloud and New Limits on a Local Hot Bubble (June 2011)
At optical wavelengths, cold interstellar gas is typically invisible in the depths of Galactic space far from any star, but Professor Dave Meyer has become an expert at locating it. He uses a technique called spectroscopy, which is a basic astronomical approach that allows one to determine the brightness of objects as a function of wavelength. The characteristic spectra of different types of stars are readily known—we know that giant, super-hot O-type stars should emit light that looks one way, for example, and small, cool M-type stars should look emit light that looks another. These characteristic spectra become very useful in confirming the existence and location of gas clouds. Cold gas clouds do not emit optical light on their own, but when interstellar gas comes between a star and our line of sight, the gas will block some of the light and reveal an absorption line on the spectrum of the star (corresponding to the atomic makeup of the cloud). When compared to the known characteristic spectrum of the star—which has certain absorption lines associated with the atomic makeup of its own atmosphere—this altered spectrum will confirm the presence and composition of the gas.
It is extremely difficult to determine the distance to interstellar gas clouds. One cannot use a technique called parallax, which is a reliable method of measuring distances to stars within 200 parsecs, since interstellar clouds are too big and amorphous (even if they are very close to us). One can, however, use spectroscopic analysis on many different stars of known distances that all pass through the same cloud of gas, making it possible to indirectly measure the distance to the gas itself. It is this method that Professor Dave Meyer employs to make some of the trickiest distance measurements in modern observational astronomy.
Based on observations of a diffuse sky background of X-ray radiation, for the past 40 years many astronomers have believed that our sun is surrounded by a bubble of very hot (one million degrees Kelvin), tenuous gas, out to an average distance of 100 parsecs. This interstellar region has been named the "Local Bubble."
About six years ago noted radio astronomer Carl Heiles (UC-Berkeley) came to Northwestern and gave a talk on an interesting interstellar gas cloud he had been studying. It was an unusual cloud because it was very cold, large, and diffuse—only about 20 degrees Kelvin, and covering approximately 30 full-moon widths across the sky. Such very cold gas temperatures had only previously been found in the dark cores of dense molecular clouds through radio observations, and for this reason Heiles' diffuse cloud was somewhat of a curiosity. At the time, no one had completed absorption line studies on the cloud, since it is in the direction of the Galactic halo, and the young, hot stars that have the smoothest spectra for interstellar absorption- line studies are rare in that region. The distance to this cloud was completely unknown, but Professor Meyer realized that the absorption-line technique might work if applied to the many cooler stars in the direction of the cloud.
Through very high-resolution optical spectroscopy, Professor Meyer could use any star in the sky—even the coolest M-type stars—to pull out any intervening interstellar absorption lines. Utilizing such instrumentation on a telescope at Kitt Peak National Observatory in Arizona, he began by observing stars over 200 parsecs away toward the cold cloud and went down to 40 parsecs still finding absorption lines in the spectra—meaning the cloud was within 40 parsecs! These findings were published in a 2006 Astrophysical Journal (Letters) article entitled "A Cold Nearby Cloud Inside the Local Bubble" in collaboration with Jim Lauroesch, Carl Heiles, Josh Peek, and Kyle Engelhorn.
The discovery left an interesting puzzle, however, due to this hot "Local Bubble" believed to surround the Sun out to 100 parsecs. How could such a cold cloud of gas exist within this extremely hot bubble? And so the 2011 paper, "The Local Leo Cold Cloud and New Limits on a Local Hot Bubble," set out to test the hypothesis: is there really a hot Local Bubble? In this new effort, led by Meyer's collaborator Josh Peek (Columbia University), new limits were placed on the distance of the cloud (between 11.3 and 24.3 parsecs) and, using the cloud's X-ray "shadow", it was determined that the bulk of the local X-ray background originates in front of the cloud.
The most likely source of these X-rays is in the vicinity of the solar system. Studies of comets have shown that X-rays are emitted when the solar wind interacts with the neutral gas in comets. A much larger reservoir of neutral gas resides on the outskirts of the solar system, which interacts with the solar wind in an area called the heliopause. It is this region that is the most likely origin of the local X-ray background. This conclusion is supported by other recent pieces of evidence and is opening up a new view of the local interstellar medium.
Accretion Activity in Supermassive Black Holes (December 2010)
This two-panel graphic contains two composite images of galaxies used in a recent study of supermassive black holes. In each of the galaxies, data from NASA's Chandra X-ray Observatory are blue, and optical data from the Sloan Digital Sky survey are shown in red, yellow and white. The galaxy on the left, Abell 644, is in the center of a galaxy cluster that lies about 920 million light years from Earth. On the right is an isolated, or "field," galaxy named SDSS J1021+1312, which is located about 1.1 billion light years away. At the center of both of these galaxies is a growing supermassive black hole, called an active galactic nucleus (AGN) by astronomers, which is pulling in large quantities of gas. (Credit: X-ray: NASA/CXC/Northwestern Univ/D.Haggard et al, Optical: SDSS)
CIERA fellow Daryl Haggard's recent study utilizes the Chandra X-ray Observatory and the Sloan Digital Sky Survey and tells scientists how often the biggest black holes in field galaxies like SDSS J1021+1312 have been active over the last few billion years. This has important implications for how environment affects black hole growth. Haggard's study found that only about one percent of field galaxies with masses similar to the Milky Way contain supermassive black holes in their most active phase. It also reports that the most massive galaxies are the most likely to host these AGN, and that there is a gradual decline in the AGN fraction with cosmic time. Finally, the AGN fraction for field galaxies was found to be indistinguishable from that for galaxies in dense clusters, like Abell 644.
X-ray Binary Formation via Mass Exchange (October 2010)
The massive black hole in M33 X-7 is hidden in the center of the X-ray-bright, pancake-shaped accretion disk of matter (orange). In a tight binary orbit, the black hole’s hot and massive stellar companion (blue) is losing mass via a stellar wind to the black hole, with the wind matter settling into a disk around the black hole. (Credit: Matthew McCrory, Francesca Valsecchi and Vicky Kalogera, Northwestern University)
Francesca Valsecchi, a graduate student working with Vicky Kalogera, recently was part of a collaboration that published on Nature the following article:
“Formation of the Black-Hole X-Ray Binary M33 X-7 via Mass Exchange in a Tight Massive System”, by F. Valsecchi, E. Glebbeek, W. Farr, T. Fragos, B. Willems, J. Orosz, J. Liu, V. Kalogera, Nature,(accepted for publication 2010). M33 X-7 is a recently discovered (2007) X-Ray luminous binary system hosting one of the most massive black holes among all X-Ray binaries known at present, a 15.65 solar mass black hole orbiting a 70 solar mass stellar companion in a Keplerian orbit of 3.45 days. The massive components (the star is the most massive star ever discovered in this class of systems), and the tight orbit, challenge our understanding of typically invoked black-hole X-Ray binaries formation channels. For the first time, we present a solution to the evolutionary history of M33 X-7 that is consistent with the complete set of observational constraints.
Triple Black Hole Interaction The evolution of a triple black hole system, showing its chaotic dynamics, and the eventual ejection of a single and a binary black hole. Credit: Stefan Umbreit (Northwestern)
Merger Tree Figure A cartoon of the merger-tree history of a Milky-Way sized galaxy and
its central black hole in cold dark matter cosmology. Time increases
from top (“small branches”) to bottom (“trunk”, at the current time). A
galaxy is formed by the sequential merger of smaller systems (white
circles), increasing in mass in a hierarchical fashion. Black holes
(black dots) evolve in a similar way - merging when their hosts do. If a
black hole binary has not coalesced yet when the host galaxy
experiences a subsequent merger, a triple black hole interaction
occurs. Credit: Marta Volonteri (University of Michigan)
Gravitational Waves as a New Probe of Dense Nuclear Matter (April 2004)
Neutron stars are the remnants of stars ten to twenty times more massive
than the Sun that end their lives with the catastrophic gravitational
collapse of their cores. The collapse is stopped when atoms are crushed to
such a high density that all the atomic nuclei are packed together into one
giant ball of pure nuclear matter. Astrophysicists estimate that a typical
neutron star contains about one and a half times the mass of the Sun
compacted into a sphere with a radius between 5 and 10 miles. The large
uncertainty in the radius of a neutron star comes from the fundamental
uncertainties in the physics of dense nuclear matter. Therefore measurements
of neutron star masses and radii could lead to significant improvements in
our basic understanding of nuclear matter and the associated fundamental
forces of Nature.
This animation shows a relativistic calculation of the inspiral and final merger of two
neutron stars driven by energy losses from gravitational radiation. A stable, rapidly rotating remnant is
Such measurements are extremely difficult to make using ordinary
electromagnetic radiation because this radiation is emitted near the surface
of the star (typically in X-rays from gas falling onto the neutron star
surface), and therefore contains very little information about the behavior
of the matter deep inside. In contrast, gravitational waves are emitted by
the bulk of the matter in the interior of a star and can therefore provide a
direct probe into the dense nuclear matter inside a neutron star. However,
to produce detectable amounts of gravitational radiation, the stars must be
moving at extremely relativistic speeds (approaching the speed of light).
This is thought to happen when two neutron stars in orbit around each other
spiral in as they emit gravitational waves, eventually colliding and merging
together into a single object. (These coalescing pairs of neutron stars have
been observed directly by radio astronomers in the form of binary pulsars.)
Northwestern University astrophysicists Joshua Faber,
and Fred Rasio (in
collaboration with K. Taniguchi of the University of Tokyo) have shown how
the detection of gravitational wave signals from binary neutron stars could lead
to a measurement of the masses and radii of the stars. While the masses of
the stars can be determined accurately during the last few minutes of the
inspiral, the radii can be measured directly by observing particular
features of the gravitational wave signal during the final merger of the two
stars. These measurements could become possible within the next few years as
the sensitivity of detectors such as LIGO (the Laser
Interferometer Gravitational-Wave Observatory) keeps improving.
Dynamics of Extrasolar Planetary Systems (March 2004)
A new era in astronomy began in the early 1990's with the first clear
detections of several Jupiter-type planets around nearby
solar-like stars. The field of extrasolar planetary astronomy went
through a real explosion in the following decade and we now
know more than 100 planets around other stars, including several
systems of multiple planets orbiting the same star (For the most recent
update, see The Extrasolar Planets
Encyclopaedia). These discoveries will no doubt lead to significant
improvements in our understanding of many processes related to planet formation, structure and evolution,
as well as deeper questions such as the existence of extraterrestrial life in
the Universe. NASA has identified the search for and
characterization of planetary systems around other stars as one of
its highest scientific priorities for the next decade.
Northwestern astrophysicists have been exploring the theoretical implications of these recent
discoveries. In 1996
Fred Rasio and
Eric Ford (at the time a Physics sophomore at MIT) wrote the first
proposing that the development of dynamical instabilities
during the early stages of planet formation could provide a natural
explanation for the very surprising orbital characteristics of the observed
systems, which include giant planets in highly eccentric
orbits, or in very tight circular orbits with periods as short as a few days.
This idea is by now well accepted and has received further
observational support with the later detections of systems containing several
giant planets of comparable masses in marginally stable
configurations (for example, in the Upsilon Andromedae system). The long-term
stability of the Solar System, in spite of its chaotic
nature, may have been necessary for the development of intelligent life. However,
it may also be very atypical, and may in fact
require very special conditions during the early stages of planet formation.
One of these special conditions may be the formation of a
single, dominant giant planet (Jupiter in the Solar System). Our work, based
on numerical integrations for a wide variety of plausible
initial conditions, examines the dynamical consequences of the presence of
several giant planets of comparable masses in the same system.
Instabilities tend naturally to develop in these chaotic systems and lead to
strong interactions between planets, often leaving them on
highly eccentric orbits, as observed. For a recent update on this work,
see the 2002 article by Ford, Rasio, & Yu (
“Dynamical Instabilities in Extrasolar Planetary Systems”).
A related area of research in the Theoretical Astrophysics Group at
Northwestern is the study of planets that have been detected
around radio pulsars (rapidly rotating neutron stars detected
from their regular pulses of radio waves). To this day, the only
example we know of a system of earth-mass planets orbiting any
star other than our own Sun is the system of planets around
the radio pulsar PSR 1257+12, discovered by Alex Wolszczan in 1992.
Clear confirmation for at least two earth-mass planets in this system came in 1994 when
predictions made by Rasio and collaborators for a near-resonant
gravitational perturbation effect in the system were verified. More
recent theoretical modeling of the pulsar timing data has revealed
the possible presence of a more distant, giant planet in the system. A
giant planet has also been detected in orbit around PSR 1620-26, a
binary pulsar located near the center of the globular cluster M4. Among the many
of this extraordinary system is that planets must also be
common in older stellar populations such as those found in globular clusters and galactic halos.
Recently, Northwestern astrophysicists Joshua Faber, and Fred
Rasio have studied the fate of gas giant planets scattered into
highly elliptical orbits which take them extremely close to the star,
somewhere between 1.5-10 stellar radii away at closest approach.
With the support of the Northwestern University
Visualization Lab, former Northwestern undergraduate
Jones converted these results into animated
Neutron Stars and Black Holes as X-Ray Sources (February 2004)
One of the great challenges of modern astrophysics is
understanding the physics of compact objects: white
dwarfs, neutron stars, and black holes. These objects
are the endpoints of ordinary stars when they run out
of nuclear fuel in their cores. Once a core is dead,
gravity can either be balanced by pressure forces of
high-density, degenerate matter (and thus form a white
dwarf or a neutron star), or win altogether and lead
to complete gravitational collapse and the formation
of a black hole. These extreme physical states are
believed to lie at the heart of some of the most
violent phenomena in the universe including various types of X-ray and
gamma-ray bursts. Compact objects are also expected to be detectable sources of gravitational waves.
Studies of compact objects in the theoretical astrophysics group at
Northwestern focus on
the evolution of their progenitors,
how their formation is determined by and affects their stellar environments, and
how they manifest their presence when found in 'binary systems' orbiting a normal star or another compact object.
Binary systems are especially interesting because they can revive the
mostly unobservable compact objects as sources of
electromagnetic and gravitational radiation. Therefore
binary systems provide a unique physical laboratory
for the study of compact objects. Some of our work has a strong
computational component and involves the construction
of theoretical models for both specific observed
systems and whole populations of compact object
systems. Progress in our understanding of these
objects is achieved by comparisons of these
theoretical models with current observations.
The 1999 launch of Chandra (one of
NASA's space observatories) has initiated a new era in X-ray
astrophysics, much in the same way that the Hubble
Space Telescope has revolutionized optical astronomy.
Chandra's sensitivity combined with its unprecedented
capability to produce sharp X-ray images makes
possible studies of compact objects that were
unthinkable before its launch.
Black holes are discovered in binary systems
detected as X-ray sources (known as X-ray binaries)
because matter from a close stellar companion is
transferred to and accreted by the black hole. The
current thinking concerning the formation of such
systems is strongly influenced by our understanding of
X-ray binaries with neutron stars (known for more than
30 years). Results from previous work lead us to
suspect that the formation of black-hole systems may
be surprisingly different than what the standard
paradigm built on their neutron-star analogs implies
at present. There are key differences between the two
classes related to the main physical driver of mass
transfer to the compact object and the type of X-ray
emission produced. We are developing models of
binary evolution and mass transfer to investigate
these differences and assess the fraction of
black-hole systems that may be hidden in the Milky
Among Chandra's great successes has been the opening of
a previously unexplored research area: X-ray binary
populations (containing both black holes and neutron stars) in galaxies
other than our own. The time
is ripe for developing sophisticated models of X-ray
binaries formed in widely different environments: from
very old galaxies to those with recent bursts of star
formation. The observed distributions of X-ray
luminosities have revealed systematic similarities and
differences among various types of galaxies, but their
physical causes are not at all understood.
The theoretical models that we are developing at Northwestern
directly take into account our knowledge of the
galactic environments. We use these models to examine
the dependence of observed population properties on
star-formation history and other environmental
factors. This work has benefited greatly from
ongoing collaborations between theorists at Northwestern
(especially V. Kalogera) and Chandra observers at the
Harvard-Smithsonian Center for
Astrophysics (G. Fabbiano's group).
Einstein's theory of General Relativity predicts that moving masses should
emit gravitational waves. These waves are perturbations of space-time itself
that propagate at the speed of light. They can in principle be detected
directly through the distortion they introduce in our local space-time on
Earth, effectively modulating distances measured between objects.
Unfortunately, the amplitudes of the waves from all known sources in the
Universe are so tiny (changing distances by at most one part in
10 21 = 1,000,000,000,000,000,000,000 !) that they are extraordinarily difficult to
detect. However, considerable efforts by physicists worldwide are currently
being devoted to their direct detection with the construction of various
detectors both on Earth (see, e.g., the
Laser Interferometer Gravitational Wave Observatory) and in space (see the
Laser Interferometer Space Antenna).
To extract the faint cosmic signals from detector noise, these
instruments will rely heavily on theoretical predictions for what the
signals should look like. This is the main reason why many theoretical
astrophysicists are devoting a lot of time to the modeling of potential
gravitational wave sources, which is also one of our major areas of research
Perhaps the most promising source of gravitational waves from the point of
view of direct detection is a system containing two black holes orbiting
each other closely. Such a system has never been detected by astronomers
(almost by definition, since it does not emit any electromagnetic
radiation), but we have a lot of indirect evidence for the existence of at
least two types of binary black holes: “ordinary” stellar-mass black hole
pairs, produced by the evolution of massive binary stars, and “supermassive”
black hole binaries produced by mergers of whole galaxies.
Curvature of space-time around a binary black hole. This particular plot of
the “lapse function” shows how time is stretched by the strong gravitational
field around each black hole.
Binary black holes lose energy as they emit gravitational waves and, as a
result, the two black holes slowly spiral inward until they finally undergo
complete coalescence. The gravitational field around the black holes is so
strong that it must be calculated using the full machinery of General
Relativity. This is a very technically challenging problem and, to this day,
no exact solution has been obtained. Two main approximate techniques exist,
one based on semi-analytic perturbation expansions (the “post-Newtonian”
approach), the other using computational algorithms to solve the equations
numerically (the “numerical relativity” approach). Of great concern is that,
until recently, the two approaches would give very different results.
Northwestern astrophysicist Philippe Grandclément,
an expert in numerical relativity, has now developed a new method to compute the structure of
space-time generated by two black holes in a close circular orbit. This was
done as part of a larger project in numerical relativity with collaborators
in Meudon, France . A new
formalism has been derived for the Einstein field equations, which are then solved numerically using fast and highly accurate
algorithms. For the first time ever, the numerical results are found to be
in good agreement with those provided by post-Newtonian techniques. This is
a very important and encouraging result, showing that this new approach
should be further pursued. In particular, we plan to apply it to a new study
of hybrid binary systems containing a black hole with a neutron star
companion. The gravitational waves emitted by these systems should also
contain important information about the dense nuclear matter inside neutron
The Common Envelope Phase of Binary Evolution (September 2000)
Binary star systems containing compact objects (white dwarfs, neutron stars,
or black holes) have come to the forefront of contemporary astrophysics as
their energetic and rapid time variability observed from space provide
important diagnostic clues of their underlying nature. Since the
interactions between the stellar components are fundamental for providing an
understanding of the observed phenomenon, they must be in close proximity to
each other. However, the formation of compact objects requires that their
progenitors must have evolved through a giant phase in the past with radii
much larger than the present day orbital separation of the binary system.
The challenge is to understand the manner in which the system was
transformed. In the Theoretical Astrophysics group at Northwestern, we are
investigating the process of common envelope evolution as a means
for facilitating this transformation. In this picture, the giant progenitor
of the compact star and its stellar companion are immersed within a
differentially rotating common gaseous envelope. The friction associated
with the gravitational interaction of these two components within such an
envelope leads to the shrinkage of the orbit and to the ejection of the
common envelope. Such a process can lead to the formation of a tight binary
containing the core of the evolved star (the immediate progenitor of the
compact object) and its stellar companion, or to a complete merger of the
system resulting in a single rapidly rotating stellar remnant. This
particular evolutionary phase is central for providing an understanding of
the formation of systems undergoing a variety of outburst phenomena (such as
X-ray binaries and cataclysmic variables) as well as double compact objects
(such as double neutron stars or black hole binaries), which are important
sources of gravitational waves.
The images below illustrate the evolution of such
a common envelope system containing a red giant star and its main-sequence
companion. These images are based on three-dimensional hydrodynamic
calculations carried out by
in collaboration with
(San Diego State University) and A. Burkert.
Common Envelope Simulation: A solar-mass red giant with a 0.35 solar mass companion.
The figures illustrate the density distribution of the binary system
entering and evolving into the common envelope phase. For clarity, the
main sequence star and the core of the red giant (with
a mass of 0.45 msun) are represented as solid
black dots in the above figures and the density logarithmically increases
from red to yellow in color. During the early phases of the evolution a
small amount of matter is ejected from the outer layers of the envelope
in the form of a spiral pattern as seen in the equatorial plane. As the
orbit shrinks the evolution becomes more symmetric as the period of the
binary orbit becomes less than the time for the orbit to decay. As viewed
perpendicular to the equatorial plane the ejection of matter is seen to be
confined toward this plane. That is, the spin up of the envelope leads to
the development of a gaseous outflow in the equatorial plane with a
corresponding inflow along the polar direction. The successful ejection
of the entire envelope is likely to lead to the formation of a system
consisting of a main sequence-like star with a white dwarf companion in a
short period orbit (i.e., a cataclysmic variable).
Given the geometry of the outflowing matter, the interaction
of the stellar wind from the recently formed white dwarf will lead to
significant deviations from spherical symmetry in the optical appearance of
the post common envelope nebulae.
Our work is supported by several grants from the National Science
Foundation and from the NASA Astrophysics Theory Program. Any opinions,
findings, and conclusions or recommendations expressed in this material are
those of the authors and do not necessarily reflect the views of the
National Science Foundation or of the National Aeronautics and Space Administration.