Friday, November 30, 2012

Progress in the Quest for the Most Distant Galaxies

One of the goals of CANDELS is to study galaxies during Cosmic Dawn, when galaxies were just beginning to form. Other teams are also pursuing this kind of research, using the CANDELS data as well as deeper observations from the Hubble space telescope.

There has been some interesting progress in the past few weeks.

The Most Distant Galaxy in the Hubble Ultradeep Field


Redshift z>9 candidate galaxy from the
Hubble Ultra-deep field. From Rychard
Bouwens' preprint.
On November 13, Rychard Bouwens posted a preprint that solidified the evidence that a previously-identified galaxy is really very distant. This galaxy is located in the Hubble Ultra-Deep Field, which is located within the CANDELS survey region. But this galaxy is too faint too see with the exposure times that we use for the rest of the CANDELS survey. It took four days of exposure time with the WFC3 infrared camera on Hubble to detect the galaxy. For comparison, the deepest CANDELS exposures are only about six hours. The best guess is that this galaxy is at a redshift of z>9. At redshift 9, the universe was 550 million years old. The light from that galaxy has taken at least 13.1 billion years to reach us.  This particular candidate had been found previously by the same team using about three days of exposure time, so it's nice to see it confirmed with deeper data.

This galaxy, along with the other ones mentioned below, was found using the Lyman Break technique, which fellow blogger Russell Ryan describes in his blog post.  Interestingly, the galaxy is invisible through all but one filter. The very fact that it is visible in the reddest filter, and invisible in all the bluer filters, is the strongest evidence that it is a very high-redshift galaxy.  The interpretation is bolstered a bit by the fact that it is not detected in very deep observations with the Spitzer infrared observatory. It seems pretty likely that this object is not an old or dusty red galaxy at lower redshift.  The source is just a little bit fuzzy in the Hubble images, so it doesn't appear to be a very faint red star in the Milky Way either.

So, it seems like a pretty good candidate. The new observations have made the detection through the one filter and the non-detection through the others more solid. But the evidence that it is truly distant is still quite tenuous, and based largely on ruling out other possibilities.

Finding a Distant Galaxy with a Little Help from a Gravitational Lens

On November 15, Dan Coe and the CLASH team posted a preprint identifying a very solid Lyman-break candidate with a redshift z~10.7. This galaxy turned up in infrared observations that used only four hours of exposure time compared to four days. Hubble got some help for this galaxy from the gravitational lensing effect of a giant cluster of galaxies located between us and the distant object. This boosted the light from the distant galaxy by about a factor of fifteen. It also resulted in three separate images of the galaxy. Each of the images is detected through the two reddest filters, but not detected at other wavelengths. So this is a really solid detection. There are no other objects in the entire CLASH survey (so far) that have similar colors.

Three separate images of the gravitationally-lensed galaxy at redshift z~10.7, from the CLASH preprint. The gravitational lens split the light into these three separate images, all of which have the color expected for a very  distant galaxy. The galaxy is invisible at wavelength shorter than ~1.4 microns due to absorption by hydrogen clouds between us and the galaxy. This is known as the Lyman Break, and is the classic signature of a distant star-forming galaxy. The Lyman Break is particularly strong in this one.

The positions of the three images of the CLASH galaxy are marked as JD1, JD2 and JD3 on this image from their preprint. Many of the galaxies that you can see near the center of the image lie in the foreground cluster. This massive cluster acts as a gravitational lens, to magnify the images of the distant galaxy (and split it into several separate images). You can see this kind of lensing effect if you look at a candle through the bottom of a wine glass. As you move the glass around, you will see the image of the candle stretch and sometimes break into several separate images. The curves drawn in the image show the "critical lines" of maximum magnification for background sources at different redshifts. The fact that the JD images lie where they do (particularly JD1 and JD2) adds a lot of support to the interpretation that this is the most distant galaxy yet discovered.
Not only is it a solid detection --  the evidence that it is a high-redshift galaxy is about as solid as it can get without actually measuring the spectrum. The three images of the galaxy lie just about where they are expected to lie based on the models of the gravitational-lens that lies in the foreground. There is some uncertainty in those models, but those models will get better as the CLASH data analysis proceeds, so even without any further Hubble observations we might learn some more about this particular candidate.

Meanwhile, Back In the Ultra-Deep Field


Images of some of the new candidates identified in the Hubble 
Ultra-deep Field, from the UDF12-team preprint. The galaxy at 
the bottom is the same one that is shown at the top of this blog 
post, which illustrates how different image processing can produce 
a very different appearance. Nevertheless, both teams agree
that the galaxy is well detected at 1.6 microns and undetected 
through all the filters at shorter wavelengths.
Just today the UDF12 team, led by Richard Ellis, have posted a preprint on their website about new candidate galaxies in the Hubble Ultra-Deep Field. These are the same observations used by Rychard Bouwens mentioned above, but analyzed by a different group. The paper concurs on the interpretation of the Bouwens object, estimating its redshift to be z=11.9. The paper also identifies a half-dozen other candidates at z>8.5. These other candidates are all detected in more than one filter (barely), but are so faint that it would not be at all surprising if a few of them turn out to be either spurious or at a different redshift. For statistical purposes, that's probably fine, because it's possible to estimate with reasonable confidence how many will turn out to be wrong.

The latest CANDELS Sample: A Bit Closer and a bit Brighter


Images of some of the brighter, closer, candidates
found in CANDELS, from Haojing Yan's preprint.
Their colors suggest that these galaxies are at
redshifts z>8. 
In this latter vein, Haojing Yan's CANDELS-team study of brighter candidates at z~8 (submitted last December) has now been accepted for publication. The CANDELS observations survey a wider area, but with shorter exposure times. So we can find the rarer, brighter sources, but not the fainter ones.  There are seven decent candidates so far, all at z ~ 8 - so a bit closer than the galaxies discussed above. The evidence that these are distant galaxies is comparable to that presented in the Ellis and Bouwens papers, but not as strong as that in the Coe paper.

What Will it Take to Improve the Evidence that these are Really Distant Galaxies?


The quest to find and confirm the most distant galaxy will continue. The gold-standard for confirming that these are really distant galaxies will be to obtain a spectrum and measure the redshift precisely.  The James Webb Space Telescope, slated for launch in 2018, can obtain a low-resolution spectrum for any galaxy that Hubble can see, with just about the same exposure time as the Hubble observations. We might get lucky and confirm some of these distant-galaxy candidates before then, but large-scale studies of galaxies at Cosmic Dawn will really take off when Webb flies.

Wednesday, November 28, 2012

Measuring the Universe

In a lot of previous posts you have read about redshift and the distance between the Milky Way and other galaxies. In this post, we step back a little bit and explore the size scales in the Universe and how distances can be measured.

First off, let's start in our Solar System, on planet Earth. Assume the size of the Earth is represented by a peppercorn (a size of about 0.08 inch). Using the same size scaling, the Sun can be represented by blowing up a balloon until it is 8 inches in diameter. In reality the Earth's diameter is about 8000 miles wide, the Sun's diameter is 800 thousand miles. This means that in the peppercorn model we assumed a single inch stands for 100 thousand miles. A yard (or 36 inches) then represent 3.6 million miles. If we rank all other planets in the solar system accordingly, Venus, our sister planet, is also a peppercorn. Both Mars and Mercury are smaller than Earth and Venus. They can be represented by the head of a pin (about 0.03 inches wide). For the size of Jupiter, the first gas planet we would encounter when travelling out of the solar system, we can use a walnut (about 0.9 inches wide). Saturn is a little smaller, about the size of an acorn (0.7 inches). Uranus and Neptune are even smaller still, so a peanut for each is adequate (0.3 inches). If you still count Pluto as a planet, than you would want to represent it by another pin head. 

Size comparisons between Solar System planets and other stars. Image source here and here.

So now, that we have established the sizes of the planets using peppercorns and nuts and so on, what about the distance between them? Well, if you take a huge step that is about 1 yard wide, you travel those 3.6 million miles described above. If you take 10 steps away from the Sun, you reached your pinhead Mercury (it's about 36 million miles away from the Sun). After another 9 steps you'll find Venus, one of the peppercorns. Take another 7 steps and you finally reached your Earth peppercorn. So the distance between the Sun and the Earth is 93 million miles (also called 1 Astronomical Unit) means taking 26 steps from the balloon Sun and your Earth peppercorn. From Earth you have to take another 14 steps to reach Mars. From there now, it is a much larger distance to reach Jupiter. You have to walk 95 steps. Remember, each step has to be the size of about 1 yard! In comparison to this distance, look at the size of Jupiter, which we represented by a walnut. From now on you have to take more and more steps between the planets to reach the next one. Saturn takes another 112 steps. From there Uranus is another 249 steps away. And to get to Neptune from Uranus, walk another 281 steps. That is nearly 3 football fields! And if you still care about Pluto, then walk another 242 steps to reach it.

Now you have walked more than 1000 yards, or across about 10 football fields and the planets have merely the size of nuts or smaller! There is a lot of space between them! 

To reach the next star, Proxima Centauri, from the Sun, a travel of 4.21 light years is required (or 1.20 parsecs). A light year is the distance light travels in a year, which is 5,878,625 million miles or 63,241 times the distance between the Earth and the Sun. In the scale we assumed for the solar system above this is about 16443 football fields or the distance between Tucson and Houston. So to reach the next star in that scale you have to travel this distance more than 4 times. You could reach Oslo in Norway from Washington D.C. and that would still be not quite far enough to go.

Our Solar System is located in the Milky Way, our home galaxy. But the Sun is only one star among 100 billion in it. And there are hundreds of billions of galaxies in the Universe. 

The cosmological distance ladder

So how do astronomers measure distances to so many far away objects, may it be other stars or other galaxies? Well, we use what is called the cosmological distance ladder. In order to reach the next step on the ladder you have to be sure about the step you are standing on. The principle of the ladder is based on the fact that each method to measure distances overlaps with another method, so that the next can be calibrated with the previous. 
Measuring distances to other stars via the Parallax and the distance R between the Earth and the Sun.

The distance ladder starts in our Solar System (or even on Earth if you want). In this previous post about the Venus transit we explained how some 100 years ago astronomers measured the distance between the Earth and the Sun and the size of the planets. Nowadays distances in our Solar System can be determined using radar, the same technique with which ships try to find the location of other ships on the ocean. Once we know the distance between the Earth and the Sun, we can determine the distance to other nearby stars using the Parallax method (see figure above). The Parallax method works out to distances of about 100 light years. The Milky Way has a diameter of about 100,000 light years. So with the Parallax we can measure only our more local neighbourhood.

For the next step on the ladder, the so-called Main Sequence Fitting, is used. With this method the distance to star clusters can be determined by exploiting the relation between brightness and colour of the cluster stars, i.e. their position in the Hertzsprung-Russell Diagram. Stars like our Sun line up on the Main Sequence in this diagram and by measuring the properties of stars, their absolute brightness can be estimated and used as distance indicator. Main Sequence Fitting can be applied across the Milky Way.

Beyond the Milky Way out to other close-by galaxies (within 10 million light years) a relation between the period of variability of Cepheid stars and their luminosity serves to measure the distance to these stars. Cepheid stars are so-called standard candles. This means that their brightness is very well known. From the difference between the observed and the known brightness, the distance can then be measured. The calibration for this method can be achieved using stars in the Small Magellanic Cloud for which main sequence fitting is still possible.

Similarly to the Cepheids, certain types of Supernovae, can also be used as standard candles. Supernovae are exploding dying stars and we already told you a lot about them here. For the calibration of this method one looks for a Supernovae in a galaxy for which the distance could be determined with Cepheid stars. Again the comparison between the absolute and the observed brightness of the Supernova allows us to calculate the distance. With Supernovae distances out to about 10 billion light years (or a redshift of about 1) can be measured.

Beyond Supernovae, Hubble's law and redshifts are used for distance measurements. Hubble's law relates the distance of an object to the speed with which it moves away from us due to the expansion of the Universe. In modern astronomy most distances beyond our local galaxy group system are given in terms of redshift. 

Having reached for out in the Universe at the end of this blog post, we have long outgrown the simple scaling that we used at the beginning to describe the distances and sizes of planets in the Solar System. And even my astronomer mind is regularly boggled by the truly astronomical scales I am confronted with every day.

Monday, November 19, 2012

Astronomer of the Month: Steven Boada

Each month we will highlight a member of the CANDELS team by presenting an interview introducing them and what it's like to be an astronomer. This month's Astronomer is Steven Boada.

Tell us a little about yourself!

My name is Steven Boada; I'm a third year graduate student at Texas A&M University in College Station, Texas. I'm originally from Pegram, Tennessee, which is a small town just outside of Nashville. Before coming to Texas, I received an undergraduate and a Master's degree in physics from the University of Tennessee. While at Tennessee, I had the wonderful opportunity to work at the National Center for Computational Science at Oak Ridge National Lab. I've done a few other interesting things along the way. I worked for a newspaper as a photographer for three years. I made thousands of ravioli for a pasta company (after getting my Master's degree). I can boil an egg in a paper cup.
What is your specific area of research? What is your role within the CANDELS team? 

I am interested in galaxy assembly and evolution. I investigate how galaxies put themselves together while evolving into the galaxies we see today. Recently, I have started looking at clusters of galaxies, which I am hoping to use to help us better understand our cosmological models of the Universe. As a part of CANDELS, I am a junior scientist, working with Dr. Casey Papovich, and an active member of the galaxy morphology team. 
What made you want to become an astronomer? At what age did you know you were interested in astronomy? 

I have been interested in astronomy for as long as I can remember. I grew up in the country just outside of Nashville, which provided slightly darker than usual skies. I recall looking at the sky and being fascinated about what I saw. Books from the library, and programs on public television only fueled that interest. When I went to university, I went in wanting to major in physics (only because there was no astronomy degree). Odd as it may sound, I have only recently learned what professional astronomers actual do. I didn't know any growing up, and my college education was almost exclusively from physicists. It has just been something that I have always been interested in, and so I decided from an early age that it was for me.
What obstacles have you encountered on your path to becoming an astronomer and how did you overcome them? 

My sophomore year I took my first class on modern physics. It was taught by a professor who was a great researcher, but a terrible teacher. I struggled, a lot. This was the first time that I really thought about changing careers. Looking back on it, I have never been that great of a physicist. Equations and math was never something that came easy for me. I have never made straight A's. I was told by a math teacher in high school that I would never amount to anything. After that modern physics course, I looked through the course catalog for something else that I would rather do with my life. I couldn't find anything that piqued my interest as much as physics and astronomy. So I decided to just work hard and make it through. So I hired a graduate student to tutor me, and six years later, I am living in Texas on my way to a PhD. 

Who has been your biggest scientific role model and why? 

There isn't a single person. Like most scientists, I have been inspired by many of the greats over the years. I would also have to give credit to the good teachers that I have had, and to my family for encouraging me a long the way. I most certainly couldn't have done it without their love and support. 

What is it like to be an astronomer? What is your favorite aspect? 

It's one of those things that is both extremely challenging and deeply rewarding. It's hard, because we don't always know what we don't know. Doing research isn't always easy. Often times it is frustrating and confusing. It can be all consuming when you are constantly thinking about a difficult problem that you are working on. But when you learn something knew or inspire some one to learn about science, it makes it all worth it.

What motivates you in your research? 

First and foremost you have to like doing research. You have to like trying to do things that people have never done before. Not being able to Google the answer to your question can, at times, be awfully frustrating. I like learning new things. I like working on hard problems and trying to think about things differently from before. Then, of course, I just plain like astronomy.

What is your favorite astronomical facility? (This could include telescopes or super computers, for example) 

There are too many to choose a favorite. Because I am still relatively new, I haven't had the opportunity to visit too many of the facilities in person. I have done some work at McDonald Observatory out in West Texas. That was a lot of fun. There are no telescope operators or support staff that stay up with you all night, so it is just you and the scope. If you break something, it is your own fault, and you  have to do a lot of the setup yourself. That makes you feel like you are doing real astronomy. With my computing background, I always have a special place in my heart for a good super-computer. ORNL, just got the fastest one in the world, Titan. 

Where do you see yourself in the future? What are your career aspirations? 

Goal number one: finish my PhD. Doing as separate Master's degree has helped in a lot of ways, but it also stretches my time in graduate school. So I am fully focused on my research and the few remaining classes. Luckily for me, I'm in a great place. In the future? Who knows. The job market in astronomy (and it seems in most sciences) is more and more in flux these days. I'd like to stay in academia, be a professional astronomer somewhere. We'll see.
If you could have any astronomy related wish, what would it be? 

More money for astronomy, always. But I'd like to go to the moon, or even just space in general. Sign me up.

What is your favorite, most mind-boggling astronomy fact? 

I love the 'Pale blue dot' photograph from Voyager. The distances in astronomy are so crazy and we are so small. It seems to me that astronomy is the most humbling science. I am just constantly reminded of how big the universe is. And not even the universe as a whole. There are galaxies one hundred times bigger than our galaxy. There are stars extremely more massive than our Sun. 

One could say the numbers are mind-boggling. They're astronomical. 

Is there anything else you would like for the public to know about you or astronomy in general? 

Nope. Remember to support your local astronomer. I play some mean ping-pong.

Friday, November 16, 2012

Unveiling Obscured Supermassive Black Hole Growth

In a previous post, we discussed the class of objects known as Active Galactic Nuclei (AGN). AGN are actively growing black holes with masses as large as 100 billion times that of our Sun. These supermassive black holes were once thought to be rare, but are now known to live at the centers of nearly all galaxies. While the majority of supermassive black holes are inactive (at least in the present day), their extreme masses exert a strong gravitational pull on their surroundings capable of ripping a star apart if it wanders too close. While black holes occasionally grow by this violent process, they are more commonly fed by interstellar gas that first settles into a rapidly rotating disk around the black hole before being accreted (eaten). 

Artist's illustration of the accretion disk and jets surrounding
a black hole. Credit: Cosmovision, Dr. Wolfgang Steffen.
For an animated version, click here.
It is this accretion disk and other features in its vicinity that make actively accreting supermassive black holes visible and that make AGN among the most luminous objects in the Universe across much of the electromagnetic spectrum. The accretion disk itself gives off bright ultraviolet and optical/visible emission capable of outshining the light from the AGN’s host galaxy. A corona surrounding the accretion disk is thought to be responsible for the bright X-rays emitted by AGN, and in approximately 15% of cases, jets launched near the supermassive black hole give rise to bright radio emission. Incidentally, it was this combination of bright radio emission and bright, yet extremely compact, optical emission that gave AGN their other well-known name: quasars, or “quasi-stellar radio sources”.

Obscured AGN
The combination of bright ultraviolet, optical, X-ray, and occasionally radio emission should give astronomers plenty of ways to identify AGN. However, actively accreting supermassive black holes appear to be surrounded not only by an accretion disk of hot gas, but also by a donut-shaped torus of colder gas and dust (astronomer lingo for very small solid particles in space). This torus lies further from the black hole than the accretion disk, and while its origin and properties are still an active field of research, one thing is clear. If we are lucky enough to view an AGN from above or below the torus (e.g., though the hole in the donut), we get a clear view of the accretion disk and its surroundings. If, however, the torus is positioned so that we must look through it to see the accreting black hole, the picture changes. 

Artist's illustration of the dusty torus that surrounds an
AGN's accretion disk.  Credit: NASA/CXC/M.Weiss
Dust, it turns out, is very good at blocking UV and optical light, and gas is very good at blocking X-ray light. When we view an AGN through the torus, much of the light we would normally see is therefore missing, or significantly weakened. We call such unfortunately aligned sources ‘obscured AGN’. If these sources were rare, this would not be a big problem for our study of supermassive black hole growth. However, in the local Universe, obscured AGN are four times more common than unobscured AGN, and some studies point to an even higher fraction of obscured AGN at earlier times. In order to study how and why supermassive black holes grow, we must first find ways to identify these obscured AGN.

Finding Obscured AGN
Thankfully, not all obscured AGN remain entirely hidden. Radio emission is relatively insensitive to dust and gas, leading to fairly complete samples for the 15% of AGN with radio jets. Furthermore, while X-rays can be blocked by gas, it takes a lot to fully block the X-ray emission observed by the current generation of X-ray satellites. Deep X-ray observations by Chandra and XMM-Newton can therefore detect many sources that would be missed in the UV and optical, and recent X-ray missions like Swift, Integral, Suzaku and NuSTAR are opening our window on the heavily obscured Universe by probing more energetic X-rays that are even harder to obscure. Certain lines in the optical spectra of AGN are also emitted in a region beyond the torus, and can be used to identify obscured AGN missed by other techniques. 

And then there is the torus itself. When the dust in the torus absorbs the UV and optical light, it heats up to temperatures as high as 1500K (~2200 degrees Fahrenheit). While this may seem quite hot, it is several times cooler than the surface of the Sun. As such, the torus does not emit in the optical like a star might, but at lower energies in the infrared. (Note: at approximately 100 degrees Fahrenheit, we also emit our own infrared or ‘thermal’ emission). While warm dusty objects in the Universe are therefore faint in the UV and optical, they can be bright in the infrared, and the same is true for AGN. The torus that obscures our view of the AGN in the UV, optical, and X-ray in fact provides us with a way of identifying luminous AGN using the infrared satellites Spitzer and WISE

Visible (left) and infrared (right) images of a person with their hand in a bag.  While the visible light is blocked by the bag, the man's infrared emission passes through the bag.  Similarly, the ultraviolet and optical light absorbed by an AGN's torus warms the dust in the torus itself, which then produces infrared radiation. Credit: NASA/IPAC

Why Obscured AGN Matter
Obscured AGN make up the biggest fraction of the total AGN population, so it is crucial that we be able to detect and study these elusive sources if we hope to understand when, where, and why supermassive black holes formed. However, obscured AGN provide astronomers with another advantage. Unlike bright unobscured AGN, whose UV and optical light outshines the light from their host galaxies, the host galaxies of obscured AGN can often be seen quite clearly with minimal interference from the AGN itself. This is crucial, as one of the major open questions in astronomy concerns the relationship between AGN and their hosts. 

The Antennae Galaxies, the nearest example of a major
galaxy merger.  Credit: NASA/ESA and the Hubble Heritage Team 
As recently as 12 years ago, astronomers believed that AGN and their host galaxies grew and evolved independently of one another. However, the surprising discovery of a tight correlation between the mass of a supermassive black hole and the mass of its host galaxy’s bulge (the so-called 'M-sigma relation') indicates that the evolution of galaxies and their supermassive black holes are tightly coupled. One possible explanation for this connection is the merger of two or more near-equal-mass galaxies, a common outcome, particularly in the early Universe. These major galaxy mergers are thought to drive gas and dust into the central regions of a galaxy, fueling both star-formation and black hole growth. If emission from the supermassive black hole then shuts off both of these processes when the AGN reaches a certain luminosity (a process astronomers call ‘feedback’), the bulge and supermassive black hole cease to grow and end up with masses that are related to one another. However, there is increasing evidence that this scenario may only be important at high luminosities and/or at high redshifts (e.g., in the early Universe). This is where CANDELS come in. 

Not only were AGN most active when the Universe was only ~3 billion years old, but the major merger scenario proposed to explain the correlation between bulge and black hole mass may also be most relevant at this time. If light traveled infinitely fast, we would have no way of knowing what AGN or their hosts looked like ~11 billion years ago. Thankfully, however, light has a finite speed, so the more distant an object is, the longer its light has been traveling to us, and the younger it was when that light was emitted. Along the way, light also gets stretched by the expanding Universe, so that light that left a distant galaxy in the optical is shifted into the near-infrared by the time it reaches us. In practice, this means that the CANDELS deep near-infrared data gives us a snapshot of the optical light that was emitted by distant AGN and their host galaxies when the Universe was still quite young. For bright unobscured AGN, CANDELS therefore allows us to study accreting supermassive black holes in the early Universe. For fainter AGN and luminous obscured AGN, however, CANDELS is providing us with the first optical images of AGN hosts in the early Universe. What will we see?  Are AGN hosts undergoing major galaxy mergers, or are other processes driving the growth of supermassive black holes and their host galaxies? Our first study of the host galaxy properties of X-ray selected AGN in CANDELS is discussed here. Stay tuned for more results! 

Wednesday, November 14, 2012

The Clearing of the Cosmic Fog and the End of the Dark Ages

If one looks at the night sky, during new moon, it appears completely dark. Using powerful telescopes, astronomers discovered that it is not entirely dark, but the faint diffuse component is barely detectable. The major contributors to this emission, spanning the wavelengths from very high energy (gamma, X-ray) to very low frequencies (far-IR, sub-mm, radio), are: particles within the Solar system; gas and dust in the Milky Way (our Galaxy); stellar plus dust radiation by galaxies beyond our own, and Active Galactic Nuclei (AGN). The latter two are also known as Extragalactic Background Light (EBL).

The Universe today is around 13.7 billion years old. During this long period, stars and AGNs had time to produce the EBL we see today, but at present we know that the Universe was very different when it was young. Before the first stars and galaxies formed, the Universe was filled with electrically neutral hydrogen gas (HI), which absorbs ultraviolet light. Thus, there was an epoch, when the Universe had an age less than approximately 400-780 million years (or redshift grater than 7-12), when it was completely dark. As the ultraviolet radiation from the first galaxies and AGN excited the gas, making it electrically charged (ionized), it gradually became transparent to ultraviolet light. This process is technically known as reionization, as there is thought to have been a brief period within the first 100,000 years after the Big Bang in which the hydrogen was also ionized. This transition from neutral to ionized hydrogen is also known as the End of the Dark Ages.

Cosmic Time and Reionization epoch. From cosmic dark ages to
light. Credit: isciencetimes
But scientists do not know exactly when the first sources of light formed and when this reionization process started to occur. An international team of astronomers used the VLT as a time machine, to look back into the early Universe and observe several of the most distant galaxies ever detected. They have been able to measure their distances accurately and find that we are seeing them as they were between 780 million and a billion years after the Big Bang (or at redshift greater than 7). These new observations have allowed astronomers to establish a time-line for the epoch of reionization for the first time. During this phase the fog of hydrogen gas in the early Universe was clearing, allowing ultraviolet light to pass unhindered for the first time.

Another pressing question for astronomers working on the high redshift Universe is the real nature of the first sources that ended the Dark Ages. Plausible candidates for the reionization processes are stars in galaxies and AGN. At high redshifts, however, the number density of luminous AGN starts to decrease, and it is rare to find super massive Black Holes actively accreting matter (AGN) at redshift greater than 6-7. Although other exotic explanations can be found, the simplest explanation for Reionization is the ubiquitous presence of galaxies in the high redshift Universe. With HST and its new instrument WFC3 working on the near-IR wavelengths, astronomers have started to find candidate galaxies at redshift greater than 7, routinely observing galaxies at redshift 8-9, and possibly a few candidates at redshift greater than 10. However, these sources have not yet been confirmed through spectroscopic observations (as has been done at redshift 7 by the VLT): at the moment they are only candidates for being the most distant sources observed in the Universe.

The Hubble Ultra Deep Field (HUDF). Credit: wikipedia
The search for candidate galaxies at high redshift is routinely carried out with different techniques (for an example see this post by K. Caputi), but a simple and widely used technique is the so-called drop-out selection. The neutral hydrogen in each galaxy, part of the Interstellar Medium (ISM), imprints a characteristic shape on the spectral energy distribution of galaxies at high redshift, absorbing almost totally the light at rest frame wavelengths less than 912 Angstroms, and also partially the light at between 912 and 1216 Angstroms at redshifts greater than 3.

Because the Universe is expanding, the wavelength of light from galaxies gets stretched as it passes through space. The further light has to travel, the more its wavelength is stretched. As red is the longest wavelength visible to our eyes, the characteristic red colour this gives to extremely distant objects has become known as redshift.’ Although it is technically a measure of how the colour of an object’s light has been affected, it is also by extension a measure both of the object’s distance, and of how long after the Big Bang we see it.

Thus, the combination of the Universe's expansion and ISM absorption turns into a typical colour combination for galaxies at high redshifts, which are typically "red at short wavelengths and blue at long wavelengths", or more simply "drop-out" galaxies, since they tend to disappear in the blue bands.

A galaxy candidate at redshift 3.
Galaxies at redshift 3 are "drop-out" in the
U band (around 3600 Angstrom observed,

corresponding to 900 Angstrom rest frame).
Credit: R. Ellis
The so called drop-out technique has allowed astronomers to find thousands of galaxies at redshift greater than 3, most of which have been successfully confirmed through spectroscopic observations. Astronomical spectroscopy is a technique which involves splitting and spreading out the light from the galaxy into its component colours, much like a prism splits sunlight into a rainbow.

Thanks to the drop-out technique and to the availability of powerful near-IR instruments like WFC3 on-board HST, we were able to find more than 150 candidate galaxies around z=7, reaching very faint luminosities. The faintest galaxies we found have a luminosity which is 1.6 billion fainter than the faintest stars we can see with the naked eye on a dark night.

To derive the contributions of these galaxies to the Reionization processes we need to know three quantities: their number density, their efficiency at emit ionizing radiation (called escape fraction, see this post by H. Teplitz), and a measurement of the non homogeneity of the Universe (called clumpiness). If galaxies at redshift around 7 are ubiquitous and numerous in number, if they emit a lot of UV ionizing radiation and if the material between one galaxy and another (called Intergalactic Medium or IGM) is distributed in a homogeneous way, than it is easy for galaxies at redshift 7 to reionize the Universe.

High surface brightness galaxy (upper left) compared with
three low surface brightness galaxies. Low surface brightness
galaxies, which have low contrast compared to the brightness
of the sky, are hard to find, even at low redshift!

Credit: University of Arizona
Using HST data, we have measured the number density of galaxies at redshift around 7. This is a very delicate measurement, since it involves a number of corrections for incompleteness and systematics that have been derived through long and complex simulations. In particular, one of the most important factors in these simulations is the surface brightness bias effect. If two galaxies have the same luminosity, but different sizes, it is easier to find the compact galaxy (with high surface brightness) than the larger one (characterized by low surface brightness).

At redshift 7 there is a relation between the size and luminosity of galaxies, with fainter galaxies being smaller, and hence with higher surface brightness. Using this relation, we were able to derive with great accuracy the number density of galaxies at redshift 7, also known as the Luminosity Function.

The Luminosity Function of galaxies at redshift 7. Blue and magenta
points and lines have been derived using CANDELS+HUDF data.
An interesting result of this work is that the number of ionizing photons emitted from galaxies at redshift around 7 cannot keep the Universe reionized if the IGM is clumpy and the ionizing escape fraction of high-z galaxies is relatively low (less than 30%). We are currently waiting for deeper and wider data from HST to confirm this result and put strict limits on the role of galaxies in the reionization of the Universe.

This research was presented in a paper called “The size-luminosity relation at z=7 in CANDELS and its implication on reionization”, to appear in the Astronomy & Astrophysics journal (A&A).

Monday, November 12, 2012

Where and When do Galaxies Form their Stars?

One of the biggest questions in galaxy formation is how, where, and why galaxies form their stars. The CANDELS project is working towards a complete census of galaxies in five deep regions of the sky, probing significant numbers of galaxies from the first billion years of the Universe, all the way to the very dim galaxies in the nearby Universe. We are fundamentally interested not just in completing the census, but in developing theoretical models that explain how this diversity of galaxy properties arises.

Within the last 15-20 years, we have developed a very compelling cosmological framework, called "Lambda-CDM," which explains the contents of the Universe and its evolution on large scales. This model has been tested by a wide range of data including data from the cosmic microwave background, which provides a detailed picture of the tiny fluctuations in the early Universe that gave rise to all of the structure we see today; the evolution of the most massive objects in the Universe, galaxy clusters, which provide a way to follow how this structure grows with time; and stellar explosions called supernovae, which provide a way to probe the expansion history of the Universe. Within this framework, most of the mass of the Universe is not in stars at all, or even in the gas that fuels galaxies. Most of the mass is "dark matter", made of a particle which does not absorb or emit light.

Although we can't see dark matter directly, we can probe its effects on the light emitted by galaxies, and in fact can use galaxies to trace where and how dark matter gets put together over time.  In our modern picture of galaxy formation, every galaxy in the Universe forms at the center of a gravitationally bound clump of dark matter, called a dark matter halo. We can use large computer simulations to determine how dark matter halos get built up over time, providing a scaffolding on which galaxy formation, galaxy growth, and galaxy evolution take place.

The movie above shows this process, just for the dark matter in a region of the Universe that becomes a massive galaxy cluster today. The bright spots here are just dense clumps of dark matter, but in our Universe this is where we expect to form galaxies.  (If you'd like to learn more about our movies, check out this recent Science Bytes special "Dark Matters" on!)

My group has been developing models of how galaxies build up on this dark matter scaffolding, and using observational data, from CANDELS and elsewhere, to infer the process of how galaxies grow and form stars within their dark matter halos. We have compiled a large amount of data from several observational surveys, and used this to say something about the average star formation histories of galaxies over time, as related to the growth of the dark matter clumps that they live in.  On average, the star formation rate in all galaxies peaked nearly 10 billion years ago, and has been declining since then. However, the rate at which star formation peaks in galaxies depends on their mass.  In a recent study, we found something very interesting. Although galaxies come with a range of shapes, sizes, masses, and star formation rates, we found that the efficiency of turning gas into stars in galaxies is nearly constant with time, and is peaked at a specific mass.  At all times over the last 10 billion years, stars form most efficiently in galaxies roughly the mass of our own galaxy the Milky Way.  This efficiency is also nearly constant over this whole time period. This means that nearly 2/3 of all stars ever formed in the Universe were formed in systems roughly the size of our own galaxy!

Halo mass as a function of time since the Big Bang. Colored bands
represent different star formation efficiencies; the yellow region indicates

the mass at which halos turn the largest fraction of their gas into stars.
Credit: Behroozi, Wechsler, and Conroy 2012
This basic picture helps us understand a number of galaxy properties. In the picture on the left, the yellow band is the region in halo mass where galaxies are able to form stars most efficiently. Small things (bottom white line), though very numerous, are not very efficient at forming stars. Today, they are generally blue, because they are still forming stars today (but slowly!). Big things, like the giant elliptical galaxies at the centers of clusters, formed a lot of stars early on (when they were roughly the mass of our own galaxy), but their halos continued to grow (top white line) and they stopped forming stars efficiently once they got more massive than the Milky Way. They are red because they haven't formed stars recently. Galaxies similar to our own actually spend a lot of their history in the region that is most efficient, and thus they form more stars relative to their total mass.

So far, this study compiled a range of data from various studies in the literature, and did not use data directly from CANDELS. The exciting thing about CANDELS is that we will be able to study galaxy star formation rates and galaxy masses for both very small and very large galaxies in the same way across more than 10 billion years of cosmic time. For the same galaxies, we will also have information about their environments, which are related to the mass of the dark matter halos that they live in, and their morphologies, which are likely related to the merger histories of the galaxies in those halos. This will allow us to build up a full picture of the diversity of galaxy growth, from the first billion years of the Universe until today.  Theorists will have a lot more work to do to understand these data, but we are looking forward to it!