Distant Universe

Cosmic Evolution

Giant galaxies and quasars, with individual light outputs equivalent to many thousands of our Milky Way Galaxy, were dominant several billion years after the Big Bang when the universe was about a quarter its present age.  Quasars are powered by the release of gravitational energy as gas falls into “supermassive” black holes at the centers of the host galaxies.  These black holes are millions to billions of times the mass of our Sun.  Gas funneling into the black holes through giant, spinning “accretion” disks contributes to their ongoing growth.  Quasars that have exhausted their fuel supply and are no longer active have been recently detected in the local universe via the orbital motions of stars and gas near the galactic centers.  The relationship between these black holes and the star formation taking place in their host galaxies is a key area of exploration.

By the present time, these dinosaurs of the universe have almost all died out, leaving a local universe that is filled with more numerous smaller galaxies and “dead” quasars.  The reason for this “cosmic downsizing” is a major mystery,  one which can only be solved by observing the time-history of galaxy and supermassive black hole evolution.

Many active quasars in the distant universe are hidden from optical view by gas and dust.  However, after the launch in late 1999 of one of NASA’s Great Observatories, the Chandra X-ray Observatory, it became possible to detect high-energy X-rays emitted in the supermassive black hole accretion process, even when the black holes are highly obscured.

UW-Madison researchers made important optical studies of the X-ray sources that were discovered to determine their distances and nature and to map the accretion history of supermassive black holes.  A particularly profound implication of this work is that supermassive black holes are being assembled from the earliest times to the present, not just during the quasar era, as was previously thought.

Studies of local supermassive black holes find a remarkably tight relation between black hole mass and the random velocities of all the stars in the host galaxy, not just those that are directly influenced by the black hole’s gravitational field.  This indicates an intimate link between the assembly of a black hole and the formation of the stars in its host galaxy.

As with the energy output of quasars, galaxies in the distant past show the most vigorous star formation activity.  Since many star forming galaxies are also surrounded by dust and gas, they cannot be seen easily at optical wavelengths.  However, dust absorbs the radiation emitted by the stars and reradiates it at much longer wavelengths, in the far-infrared and submillimeter.

UW-Madison researchers helped make the first discovery in blank fields of giant, distant, dust-obscured galaxies at submillimeter wavelengths.  Although the number density of these galaxies is low compared to that of optically detected sources, they contribute large amounts of light and cumulatively dominate the star formation at early times.  However, just as with the quasars, these giant star formers have all but vanished at the present time.  The current star formation, which is still substantial, is produced by much larger numbers of smaller galaxies with modest star formation rates.

We now need to tie the galaxy and quasar populations together to establish their overall properties and to learn how the star formation episodes are related to supermassive black hole growth.  The resolution of single-dish submillimeter telescopes is poor, making it difficult to pinpoint the counterparts to the submillimeter sources at other wavelengths and learn about their properties.  However, advances have been made by exploiting the well-known correlation between radio continuum luminosities and far-infrared luminosities, which allows one to take advantage of the subarcsecond resolution of radio interferometry to identify counterparts and hence measure distances.

With distances it is possible to map the dusty star-formation history of the universe.  The overall star formation history is indeed found to parallel that of the accretion history.  Recent submillimeter interferometry has enabled UW-Madison researchers to identify sources at even greater distances, beyond what can be done with current radio sensitivities.  The future of this field lies with the remarkable Atacama Large Millimeter/submillimeter Array in Chile.

Run by Amy Barger.

Young Galaxies

High-redshift galaxy studies rely in large part on samples selected using color techniques or extreme emission-line properties.  For example, Lyman Break Galaxies (LBGs) are star-forming z>3 galaxies selected on the basis of their rest-frame UV colors, which are indicative of significant absorption at wavelengths below 912 A.  However, it is very difficult to know how such galaxy populations relate to one another or how they fit into the overall scheme of galaxy evolution.

A classic example of a highly valued population whose relationship to other samples is unclear is the Lyman alpha emitter (LAE) population.  Lyman alpha emission-line searches have been widely used to find the highest redshift galaxies (z>6).  This line is the only spectroscopic signature that can be used to confirm the redshift of a galaxy selected on the basis of its color properties.  However, Lyman alpha is a difficult line to interpret.  Because the line is resonantly scattered by neutral hydrogen, determining its escape path and hence its destruction by dust is an extremely complex problem.  Thus, our understanding of what determines the fraction of galaxies with Lyman alpha emission is weak.

Ideally, we would like to know whether the presence of Lyman alpha emission is related to other properties of the galaxy, such as its metallicity, extinction, morphology, or kinematics – i.e., what controls the escape of Lyman alpha photons, and how the properties of the highest redshift LAEs are modified by the intergalactic gas.

Due to the difficulty with observing in the UV, until recently we had much more information on the z~2-3 LAEs and how their properties related to those of other UV selected galaxies at these redshifts (LBGs) than we did on lower-redshift samples.  However, substantial samples of z~0.3 and z~1 LAEs have now been found with the Galaxy Evolution Explorer (GALEX) grism spectrographs.  These samples have many advantages.  The galaxies are bright and can be easily studied at other wavelengths.  Perhaps even more importantly, they can be integrated into comprehensive studies of galaxies at the same redshifts to understand some of the selection biases.  Moreover, by comparing the Lyman alpha properties of the low-redshift galaxies with their optical properties, including their H alpha line strengths, we can calibrate the conversion of Lyman alpha luminosity to star formation rate.

We are learning a lot about LAEs from these samples and from the comparison of these samples to those at high-redshifts.  For example, it appears that LAEs represent an early stage in a starburst when the star-forming gas is still relatively pristine and the primary star-forming region is small.  It also appears that there is a time sequence, with the Lyman alpha emission line dying away and the metallicity of the gas rising as the galaxy evolves.  We hope to make significant advances in placing the LAEs into context by studying these objects with the SALT spectrograph.

Another young galaxy population was discovered through imaging with narrowband filters built to find high-redshift LAEs.  These “late bloomers” are undergoing their first episodes of star formation quite recently (z~0.8) and were discovered by their unusually strong [OIII] emission.  They generally have very low metallicities, which means we may be able to obtain clues about the first stages of galaxy formation in the universe by studying in detail a large sample. We plan to conduct a wide-area search with the One Degree Imager on WIYN that will turn up thousands of these galaxies.  We will then be able to explore whether these newly forming galaxies contain some minimum amount of heavy elements, as might be expected if the intergalactic gas is relatively uniformly enriched with metals made in previous galaxy formation.  If extremely low metallicity galaxies are found, then we can conclude that portions of the intergalactic gas have remained relatively pristine.

Run by Amy Barger.

Galactic Winds

A central question in galaxy evolution is how the formation of stars is regulated. Supernovae and massive star winds are believed to provide a source of negative ‘feedback’ by imparting energy and momentum to the interstellar medium out of which stars are born.  In galaxies with a high rates of star formation, the net effect is to drive gas and dust out of the galaxy disk in a dramatic ‘galactic wind’.

Galactic winds contain a mixture of extremely hot metal-enriched supernova ejecta and cooler entrained gas and dust.  The amount of outflowing matter is substantial, comparable to the amount of matter that forms into stars.  Outflowing material has been observed at great distances from galaxies (10 to 100 kpc). In some cases outflows escape the galaxy potential well and pollute the intergalactic medium with heavy elements.

Galactic winds are widely recognized as important ingredients in galaxy evolution.  They cause galaxies to form stars more slowly over cosmic time, they influence the structure of galaxy disks, and they impact the chemical enrichment of galaxies and the intergalactic medium.  Winds are now routinely incorporated into theoretical models of galaxy evolution.  However, there is much about the underlying ephysics that is still poorly understood.  The key to improving our understanding is detailed observations.

Galactic winds are challenging to observe because the outflowing gas is diffuse and spans a wide range of temperatures (10 – 10^8 K).  Images of galaxies taken through special narrow-band filters that isolate the H-alpha and [NII] emission lines have revealed dramatic outflows in several local starbursts. One very famous example studied by UW astronomers is Messier 82 (see image at top).  A powerful new instrument on the SALT telescope called a Fabry-Perot will enable a host of new studies of this nature.

In more distant galaxies, the diffuse emission from the outflow becomes difficult to detect.  A better way to probe outflows in these objects is spectroscopy.  Cool gas in the wind absorbs photons produced by the galaxy’s stars.  By measuring the wavelength of the absorption lines and the detailed line profile shapes astronomers can infer the velocity of the gas and its density along the line of sight.  UW researchers are using this technique to examine how the velocity of galactic winds depends on galaxy physical properties such as mass and star formation rate.

Recently, astronomers have begun to consider the impact that an active black hole could have on its host galaxy’s interstellar medium.  AGN feedback [link to project under compact objects] is thought to play a role in shutting off star formation entirely in galaxies.  UW astronomers are using the Sloan Digital Sky Survey to identify and study galaxies whose star formation has been recently and abruptly quenched to gain insight into this phenomenon.

Run by Christy Tremonti.

Galaxy Mergers

Over the years evidence has mounted for a significant mode of galaxy evolution via mergers.  This process links gas-rich, disk galaxies; star-bursting galaxies; active galactic nuclei (AGN); post-starburst galaxies; and gas-poor, dynamically hot, elliptical galaxies, as objects representing different phases of major galaxy mergers.  Hydrodynamic simulations of galaxy mergers predict that as the galaxies coalesce, gravitational forces funnel gas toward the center, which provides a fuel reservoir to feed the central supermassive black hole and to form large numbers of stars in a nuclear starburst.  The AGN activity may be optically obscured for a time, until feedback due to either supernova winds or the AGN accretion blow out much of the surrounding gas in an outflow.  At this point the accreting black hole is exposed as an observable optical quasar.  After the blowout, remaining nuclear gas is soon used up, ending AGN activity and quickly quenching the nuclear starburst in the galaxy.  The galaxy then quiescently fades to a red elliptical.

The post-starburst (or E+A) phase is particularly interesting because nearly every galaxy that evolves from an actively star-forming phase to a quiescent, “red and dead” phase must pass through it.  In essence, the E+A phase is a sort of galaxy evolution “bottleneck” that indicates that a galaxy is actively evolving through several important physical transitions.  Careful study of E+A galaxies could in principle disentangle some of the many paths of galaxy evolution that lead to it – including major merger parameters, gas consumption, and AGN formation and duty cycles.

UW researchers are putting observational constraints on the timing of the phases of galaxy mergers by studying the transitional objects – those near the end of or after the starburst phase.  The study is based on a sample of galaxies identified from the SDSS that show a post-starburst signature in their spectra and have radio properties that indicate the presence of buried AGN.  Ongoing observational campaigns in multiple wavelength regimes provide the data from which timeline information is being extracted.

1. We use optical spectra that are spatially distributed across the galaxies from the WIYN and SALT telescopes to study the stellar populations.  How long ago did a burst of star formation happen?  How long did it last?  In which parts of the galaxy did it occur?  What stellar mass was formed during the burst?
2. We use high resolution near infrared images from the WHIRC camera on WIYN to determine from spatially distributed galaxy colors whether large amounts of dust could be obscuring ongoing star formation or AGN activity in the galaxies.
3. We use radio fluxes observed over a range of frequencies with the VLA and GMRT radio interferometers to estimate how long the radio emission from the AGN has been active.  Does the timing of the AGN match that of the starburst or not?

Run by Marsha Wolf and Eric Hooper.