CAPTION Top: Actual Hubble observations of gas density in the central portion of two galaxies. Bottom: Computer simulations of knots of star formation in the two galaxies show how gas falling into a galaxy's center is controlled by jets from the central black hole. CREDIT Credits: NASA/ESA/M. Donahue/Y. Li

Astronomers have uncovered a unique process for how the universe's largest elliptical galaxies continue making stars long after their peak years of star birth. NASA's Hubble Space Telescope's exquisite high resolution and ultraviolet-light sensitivity allowed the astronomers to see brilliant knots of hot, blue stars forming along the jets of active black holes found in the centers of giant elliptical galaxies.

Combining Hubble data with observations from a suite of ground-based and space telescopes, two independent teams found that that the black hole, jets, and newborn stars are all parts of a self-regulating cycle. High-energy jets shooting from the black hole heat a halo of surrounding gas, controlling the rate at which the gas cools and falls into the galaxy.

"Think of the gas surrounding a galaxy as an atmosphere," explained the lead of the first study, Megan Donahue of Michigan State University. "That atmosphere can contain material in different states, just like our own atmosphere has gas, clouds, and rain. What we are seeing is a process like a thunderstorm. As the jets propel gas outward from the center of the galaxy, some of that gas cools and precipitates into cold clumps that fall back toward the galaxy's center like raindrops."

"The 'raindrops' eventually cool enough to become star-forming clouds of cold molecular gas, and the unique far ultraviolet capabilities of Hubble allowed us to directly observe these 'showers' of star formation," explained the lead of the second study, Grant Tremblay of Yale University. "We know that these showers are linked to the jets because they're found in filaments and tendrils that wrap around the jets or hug the edges of giant bubbles that the jets have inflated," said Tremblay, "And they end up making a swirling 'puddle' of star-forming gas around the central black hole."

But what should be a monsoon of raining gas is reduced to a mere drizzle by the black hole. While some outwardly flowing gas will cool, the black hole heats the rest of the gas around a galaxy, which prevents the whole gaseous envelope from cooling more quickly. The entire cycle is a self-regulating feedback mechanism, like the thermostat on a house's heating and cooling system, because the "puddle" of gas around the black hole provides the fuel that powers the jets. If too much cooling happens, the jets become more powerful and add more heat. And if the jets add too much heat, they reduce their fuel supply and eventually weaken.

This discovery explains the mystery of why many elliptical galaxies in the present-day universe are not ablaze with a higher rate of star birth. For many years, the question has persisted of why galaxies awash in gas don't turn all of that gas into stars. Theoretical models of galaxy evolution predict that present-day galaxies more massive than the Milky Way should be bursting with star formation, but that is not the case.

Now scientists understand this case of arrested development, where a cycle of heating and cooling keeps star birth in check. A light drizzle of cooling gas provides enough fuel for the central black hole's jets to keep the rest of the galaxy's gas hot. The researchers show that galaxies don't need fantastic and catastrophic events such as galaxy collisions to explain the showers of star birth they see.

The study led by Donahue looked at far-ultraviolet light from a variety of massive elliptical galaxies found in the Cluster Lensing And Supernova Survey with Hubble (CLASH), which contains elliptical galaxies in the distant universe. These included galaxies that are raining and forming stars, and others that are not. By comparison, the study by Tremblay and his colleagues looked at only elliptical galaxies in the nearby universe with fireworks at their centers. In both cases, the filaments and knots of star-birth appear to be very similar phenomena. An earlier, independent study, led by Rupal Mittal of the Rochester Institute of Technology and the Max Planck Institute for Gravitational Physics, also analyzed the star-birth rates in the same galaxies as Tremblay's sample.

The researchers were aided by an exciting, new set of supercomputer simulations of the hydrodynamics of the gas flows developed by Yuan Li of the University of Michigan. "This is the first time we now have models in hand that predict how these things ought to look," explains Donahue. "And when we compare the models to the data, there's a stunning similarity between the star-forming showers we observe and ones that occur in simulations. We're getting a physical insight that we can then apply to models

Along with Hubble, which shows where the old and the new stars are, the researchers used the Galaxy Evolution Explorer (GALEX), the Herschel Space Observatory, the Spitzer Space Telescope, the Chandra X-ray Observatory, the X-ray Multi-Mirror Mission (XMM-Newton), the National Radio Astronomy Observatory (NRAO)'s Jansky Very Large Array (JVLA), the National Optical Astronomy Observatory (NOAO)'s Kitt Peak WIYN 3.5 meter telescope, and the Magellan Baade 6.5 meter telescope. Together these observatories paint the complete picture of where all of the gas is, from the hottest to the coldest. The suite of telescopes shows how galaxy ecosystems work, including the black hole and its influence on its host galaxy and the gas surrounding that galaxy.

CAPTION This image, captured by the Very Large Array, shows the atomic hydrogen distribution of the Whirlpool Galaxy. The "X" marks the dwarf companion satellite. Dynamical simulations can recover its location and mass. CREDIT VLA, Chakrabarti et al. 2011

RIT professor Sukanya Chakrabarti wins NSF grant to explore extended galactic disks 

A ripple in the outskirts of the Milky Way--and a hunch--led Rochester Institute of Technology astrophysicist Sukanya Chakrabarti to a previously undetected dwarf galaxy hidden under a veil of dark matter. Now Chakrabarti is refining her technique to uncover dwarf galaxies and understand dark matter by simulating the evolutionary histories of galactic disks, rich in atomic hydrogen, and their satellite populations.

Chakrabarti's study on these overlapping regions found in spiral galaxies, like the Milky Way, is funded by a three-year $325,053 grant from the National Science Foundation. Her research seeks to solve an astrophysical conundrum dubbed "the missing satellites problem," in which theoretical simulations that predict an abundance of satellite galaxies are unsupported by observational data.

Earlier this year, Chakrabarti, assistant professor of physics in RIT's School of Physics and Astronomy, validated her prediction of a previously unseen satellite galaxy located close to the plane of the Milky Way. In her new study, Chakrabarti and Andy Lipnicky, a Ph.D. student in RIT's astrophysical sciences and technology program, will create the first "mock" map and catalogue of satellite populations from analyzing extended atomic hydrogen disks.

"We will produce models that are consistent with both the atomic hydrogen and stellar data of our galaxy, which displays large ripples in the outskirts, a prominent warp and vertical waves in the galactic disk," Chakrabarti said.

Chakrabarti's goal of gaining an understanding of the distribution of dark matter combines her method with gravitational lensing. She will analyze the ripples in the atomic hydrogen map and results from gravitational lensing--a technique that uses the bending of light to weigh distant galaxies and reconstruct the dark-matter background.

"Comparing and contrasting results from both methods might improve the statistics of detecting dark-matter dominated dwarf galaxies," Chakrabarti said.

Hubble Space Telescope image of the cluster of galaxies MACS0416.1-2403, one of the Hubble ‘Frontier Fields’. Bright yellow ‘elliptical’ galaxies can be seen, surrounded by numerous blue spiral and amorphous (star-forming) galaxies. Gravitational arcs can also be seen. This image forms the test data that the machine learning algorithm is applied to, having not previously ‘seen’ the image. Credit: NASA / ESA / J. Geach / A. Hocking.

A team of astronomers and computer scientists at the University of Hertfordshire have taught a machine to 'see' astronomical images. The technique, which uses a form of artificial intelligence called unsupervised machine learning, allows galaxies to be automatically classified at high speed, something previously done by thousands of human volunteers in projects like Galaxy Zoo. Masters student Alex Hocking led the new work and presented it for the first time in a paper today (July 8) at the National Astronomy Meeting at Venue Cymru, Llandudno, Wales.

The team have demonstrated their algorithm using data from the Hubble Space Telescope ‘Frontier Fields’: exquisite images of distant clusters of galaxies that contain several different types of galaxy.

Mr Hocking, who led the new work, commented: “The important thing about our algorithm is that we have not told the machine what to look for in the images, but instead taught it how to 'see'."

His supervisor and fellow team member Dr James Geach added: “A human looking at these images can intuitively pick out and instinctively classify different types of object without being given any additional information. We have taught a machine to do the same thing."

‘Our aim is to deploy this tool on the next generation of giant imaging surveys where no human, or even group of humans, could closely inspect every piece of data. But this algorithm has a huge number of applications far beyond astronomy, and investigating these applications will be our next step," concludes Geach.

The scientists are now looking for collaborators, making good use of the technique in applications like medicine, where it could for example help doctors to spot tumours, and in security, to find suspicious items in airport scans.

Hocking network full mediumVisualisation of the neural network representing the ‘brain’ of the machine learning algorithm. The intersections of lines are called nodes, and these represent a map of the input data. Nodes that are closer to each other represent similar features within the data. Fainter lines show how the network has evolved over time as the algorithm processes more data. Credit: J. Geach / A. Hocking. 

Hocking network detail mediumA zoom-in of part of the network described above. Credit: J. Geach / A. Hocking.

Hocking galaxy classification 1 mediumImage showing the MACS0416.1-2403 cluster, highlighting parts of the image that the algorithm has identified as ‘star-forming’ galaxies. Credit: NASA / ESA / J. Geach / A. Hocking.

Hocking galaxy classification 2 mediumImage showing the MACS0416.1-2403 cluster, highlighting parts of the image that the algorithm has identified as ‘elliptical’ galaxies. Credit: NASA / ESA / J. Geach / A. Hocking. 

The Coma Galaxy Cluster is a massive cluster of galaxies in the constellation Coma Berenices. Each point of light in this image may look like a star but in fact they are, mostly galaxies. With over 650 galaxies in the cluster, Abell 1656 is one of the densest collections of galaxies in the entire sky. Credit: Greg Parker, New Forest Observatory.

Galaxies in a cluster roughly 300 million light years from Earth could contain as much as 100 times more dark matter than visible matter, according to an Australian study.

The research, published today, used powerful supercomputer simulations to study galaxies that have fallen into the Coma Cluster, one of the largest structures in the Universe in which thousands of galaxies are bound together by gravity.

"It found the galaxies could have fallen into the cluster as early as seven billion years ago, which, if our current theories of galaxies evolution are correct, suggests they must have lots of dark matter protecting the visible matter from being ripped apart by the cluster."

Dark matter cannot be seen directly but the mysterious substance is thought to make up about 84 per cent of the matter in the Universe.

International Centre for Radio Astronomy Research PhD student Cameron Yozin, who led the study, says the paper demonstrates for the first time that some galaxies that have fallen into the cluster could plausibly have as much as 100 times more dark matter than visible matter.

Yozin, who is based at The University of Western Australia, says the galaxies he studied in the Coma Cluster are about the same size as our own Milky Way but contain only one per cent of the stars.

He says the galaxies appear to have stopped making new stars when they first fell into the cluster between seven and ten billion years ago and have been dead ever since, leading astrophysicists to label them "failed" galaxies.

This end to star formation is known as "quenching".

"Galaxies originally form when large clouds of hydrogen gas collapse and are converted to stars--if you remove that gas, the galaxy cannot grow further," Yozin says.

"Falling into a cluster is one way in which this can happen. The immense gravitational force of the cluster pulls in the galaxy, but its gas is pushed out and essentially stolen by hot gas in the cluster itself.

"For the first time, my simulations have demonstrated that these galaxies could have been quenched by the cluster as early as seven billion years ago.

"They have however avoided being ripped apart completely in this environment because they fell in with enough dark matter to protect their visible matter."

This research was motivated by the recent observational discovery of these galaxies by an American and Canadian team led Professor Pieter van Dokkum of Yale University. 

Using the data the North American team published last year, Yozin was able to create supercomputer simulations to model how the galaxies evolved into what we can see today.

Experiments at the Z machine at Sandia National Laboratories have provided data that may explain why Saturn is 2 billion years younger than Jupiter in some simulations.

How does Saturn hide its age?

Planets tend to cool as they get older, but Saturn is hotter than astrophysicists say it should be without some additional energy source. 

The unexplained heat has caused a two-billion-year discrepancy for supercomputer models estimating Saturn's age. "Models that correctly predict Jupiter to be 4.5 billion years old find Saturn to be only 2.5 billion years old," says Thomas Mattsson, manager of Sandia's high-energy-density physics theory group.

Experiments at Sandia's Z machine may have helped solve that problem when they verified an 80-year-old proposition that molecular hydrogen, normally an insulator, becomes metallic if squeezed by enough pressure. Physicists Eugene Wigner and Hilliard Huntington predicted in 1935 that a pressured lattice of hydrogen molecules would break up into individual hydrogen atoms, releasing free-floating electrons that could carry a current.

"That long-ago prediction would explain Saturn's temperature because, when hydrogen metallizes and mixes with helium in a dense liquid, it can release helium rain," said Sandia researcher Mike Desjarlais. Helium rain is an energy source that can alter the evolution of a planet.

"Essentially, helium rain would keep Saturn warmer than calculations of planetary age alone would predict," said Marcus Knudson. Knudson and Desjarlais are the lead authors of a June 26 Science article, "Direct observation of an abrupt insulator-to-metal transition in dense liquid deuterium."

This proposed density-driven hydrogen transition had never been observed experimentally until Sandia's recent experiments.

The tests ran on Sandia's Z machine, the world's most powerful pulsed-power machine, which sends a huge but precisely tuned sub-microsecond pulse of electricity at a target. The correspondingly strong magnetic field surrounding the pulse was used to shocklessly squeeze deuterium -- a heavier variant of hydrogen -- at relatively low temperatures. Previous experiments elsewhere used gas guns to shock the gas. This increased its pressure but at the same time raised its temperature beyond the range of interest for the density-driven phase transition.

"We started at 20 degrees Kelvin, where hydrogen is a liquid, and sent a few hundred kilobar shock -- a tiny flyer plate pushed by Z's magnetic field into the hydrogen -- to warm the liquid," said Knudson. "Then we used Z's magnetic field to further compress the hydrogen shocklessly, which kept it right above the liquid-solid line at about 1,000 degrees K."

Said Desjarlais, "When the liquid was compressed to over 12 times its starting density, we saw the signs that it became atomic rather than molecular. The transition, at three megabars of pressure, gives theorists a solid figure to use in their calculations and helps identify the best theoretical framework for modeling these extreme conditions."

The results need to be plugged into astrophysical models to see whether the now-confirmed transition to atomic hydrogen significantly decreases the age gap between the two huge planets.

"The Sandia work shows that dense hydrogen can be metallic, which in turn changes the coexistence of hydrogen and helium in the planet," says Mattsson. "The mechanism of helium rain that has been proposed is therefore very plausible, given our results, but the scientific discussion will continue over the next few years in establishing a new consensus."

Interestingly, the determination that a metallic phase was reached was made optically. "There's too much electrical noise in Z to make an electrical test, though we plan to directly measure current down the road," Knudson said.

Optical tests rely on the transition from zero reflectivity (insulators) to the reflectivity achieved by metals.

"The only way you get reflectivity is when a material is metallic," Knudson said. Reflectivity was tested across the visible spectrum -- from 450 to 750 nanometers. "The experiment itself produced light," he said. "We collected it, put it through a spectrometer to disperse it and passed it into a camera to observe it."

When the hydrogen insulator reached enough pressure to become metallized, the researchers observed 45 percent reflectivity, an excellent agreement with theoretical calculations, said Desjarlais.

"This is a very nice merging of theory and experiment," he said. "We threw all our computational tools -- which are significant -- at providing verification and interpretation of the complex experimental observations at Z."

The work was done in collaboration with professor Ronald Redmer's research group at University of Rostock in Germany and is a part of the Z Fundamental Science Program at Sandia. The multidisciplinary team included researchers with expertise in innovative experimental design, diagnostics and pulse-shaping capabilities, matched with theoretical analysis using methods based on quantum mechanics.

Other authors besides Knudson, Desjarlais and Mattsson include Redmer and Andreas Becker at University of Rostock and Ray Lemke, Kyle Cochrane, Mark Savage, and Dave Bliss at Sandia.

The Z machine is a National High Energy Density Science Facility supported by the National Nuclear Security Administration. 

Page 4 of 13