WFC3 Early Release Science (ERS) Observations

The GOODS (North and South) fields are two major fields observed by the CANDELS program. The GOODS-N field will be fully covered by CANDELS near-IR observations, while only ~70% of the GOODS-S field area will be observed by CANDELS near-IR observations. The main reason is that the remaining ~30% of the area in the GOODS-S field was already observed by the Wide Field Camera 3 (WFC3) Early Release Science (ERS) observations. Today, I will summarize these ERS observations.

The multiwavelength 10-band color image of the entire ERS mosaic in the GOODS-South field combines a broad range of colors, from the ultraviolet, through visible light, and into the near infrared. Such a detailed multi-color view of the universe has never before been assembled in such a combination of color, clarity, accuracy, and depth. Credit: NASA, ESA, R. Windhorst, S. Cohen, M. Mechtley, and M. Rutkowski, R. O'Connell, P. McCarthy, N. Hathi, R. Ryan, H. Yan, and A. Koekemoer.

The Scientific Oversight Committee (SOC) was constituted in 1998 to oversee the science definition and the technical development of the WFC3. In recognition of their effort during 11 years of WFC3 development, the STScI director allocated ~210 orbits (~300 hours) of the HST time to the SOC to do challenging science programs designed to test key capabilities of the WFC3. To this effect, the SOC distributed these orbits equally between two major programs, to investigate (1) nearby galaxies and (2) distant galaxies. The combination of these two programs is referred to as the 'WFC3 Early Release Science (ERS) Program." I will only talk about the 'distant galaxies' program, as it is most relevant to the CANDELS program. The 'distant galaxies' program was allocated 104 orbits (~150 hours).

I got involved in this program through my PhD advisor, Rogier Windhorst, who is a member of the SOC. Myself, Seth Cohen, Russell Ryan, and few SOC members were involved in planning this 'distant galaxies' program. The GOODS-S field was chosen because of the availability of extensive ancillary data in this field. The primary goal of the ERS program was to showcase capabilities of the WFC3, so we planned 3 sets of observations: ultraviolet imaging (total 40 orbits), near-infrared imaging (total 60 orbits), and near-infrared grism (low resolution spectroscopy, total 4 orbits). The ultraviolet and near infrared imaging covers the full ERS field (~45 arcmin²), while the grism observations cover only ~1/10th of the ERS field.

Multiwavelength postage stamps of objects with interesting morphological structure
in the 10-band ERS color images of the GOODS-South field: from left
to right, high signal-to-noise detections of ERS galaxies resembling the main
cosmological parameters w, λ and ρ respectively. These images illustrate the rich and
unique morphological information available in the 10-band panchromatic ERS data set.
Credit: Rogier Windhorst and WFC3 SOC

These observations are being used for various science goals, including identifying and understanding passive galaxies, star-forming galaxies at intermediate redshifts, high redshift galaxies, emission-line galaxies, and galaxy morphology/mergers at various redshifts. These data are also useful to identify faint red stars in our own galaxy. Details of ERS observations are in Windhorst et al. 2011. The intermediate depth ERS observations are excellent complementary data to the CANDELS wide shallow and the CANDELS deep observations. CANDELS team is generating a large near-infrared mosaic in the GOODS-S field, which will include ERS as well as CANDELS datasets in a single image. Keep coming back for more results from these datasets.

Meet Christina Williams

My name is Christina Williams, and I am a grad student in Astronomy at the University of Massachusetts in Amherst. I study galaxy evolution with CANDELS data as part of my PhD thesis. I'm working with Mauro Giavalisco as my thesis adviser, studying compact, massive, elliptical galaxies and their evolution. How and why I arrived at an astronomy department for work is in some ways similar to other astronomers, and in some ways different. Like many, I had an early fascination with the world around me, how nature works, and in particular the night sky, which is what led me here! I grew up in Washington, DC, a city full of culture and people from all over the world, but not so much in the way of nature and dark skies. I was lucky to have extremely interesting and inspiring science teachers in elementary and middle school, who showed us all sorts of fascinating gadgets in the lab and taught us about things like volcanoes and tornadoes, which you don't find in Washington. But I was especially lucky to have science teachers and advisers throughout my life, who took a special interest in my ambition to learn science, and made sure to foster it. By high school, I was convinced Astronomy was my route in life, giving in to my hunger to learn more and more fundamental aspects of science. The universe, it seemed to me, was about as fundamental as it gets!

I went to college to study Physics, not too far from home at Johns Hopkins University (JHU) in Baltimore, MD, which had a very research oriented department of Physics and Astronomy. With its location across the street from Space Telescope Science Institute, it seemed like the perfect place to be introduced to the world of astronomical research. I immediately joined a research group focused on low-mass stars and brown dwarfs, with whom I worked for all my four years there. Since JHU is a part of the Sloan Digital Sky Survey, we had loads of telescope time at its host observatory, Apache Point, in New Mexico. I traveled there several times for observing runs and also observed remotely (through a computer) from Baltimore.

Aurora Borealis over my cabin in Fairbanks, AK.
Photo credit: Christina Williams

The summer before my senior year, I decided to try something new for one summer, and received an internship studying the polar ice caps of Mars at the Geophysical Institute at the University of Alaska in Fairbanks (UAF), as part of the Research Experience for Undergraduates program (REU; see this recent blog post by a CANDELS REU student). This was an important scientific, and life changing, experience. It was there that I first learned how to make physical models with a computer, which is what many astronomers spend much of their time doing. And it was then that I fell in love with Alaska, with its big mountains, glaciers, and vast wilderness to explore. Not ready to leave Alaska, but desiring still to continue with scientific research, I took a brief hiatus from astronomy and enrolled in a masters program in Geophysics at UAF. While there, I wrote a thesis on the formation of the Arctic Ocean and its tectonic history, (something totally new for me), which is a part of the Earth that is still very poorly understood because the sea ice in the Arctic makes studying the ocean floor difficult. Living year-round in Alaska offers many exotic experiences for an exploratory spirit. Like many other Fairbanksans, I lived in a cabin without running water. There were Northern Lights to photograph, hot springs to ski to in the winter, and remote and wild rivers to float in the summer.

Fieldwork on the sea ice near Barrow, AK

I finished my masters degree mid-(academic) year, so before returning to the world of Astronomy and starting my PhD the next fall, I accepted a job with the arctic sea ice research group at the Geophysical Institute at UAF. They hired me to work with the marine radars they had set up on the Arctic coast, as part of an ice observatory, which monitors the real-time motion and other changes in the coastal sea ice. I also went to Barrow, Alaska, for fieldwork out on the ice. To get out there, we rode snow-machines on landfast sea ice (sea ice which is grounded off the coast after winter), testing the electric conductivity and albedo of the ice, and taking core samples to learn about how the ice composition changes over time. One of many goals of this constant monitoring is to learn about how the ice patterns have been changing in recent decades. This is increasingly important in part because Inupiat Eskimo communities, who have relied on knowledge of seasonal landfast ice patterns for subsistence hunting for centuries, are now faced with the need to adapt to changes in the arctic.

ASTE site, in the Atacama Desert, Chile

The next fall I started my PhD in Astronomy at UMass. The astronomy program here is structured such that you have two different research projects before choosing a thesis topic, and in this way get exposed to a variety of sub-fields within astronomy. In my first few years here, I studied sub-millimeter galaxies (SMGs), and their clustering (see this recent blog post, and also this one). These are galaxies which are so dust-obscured, they are often only observable at long wavelengths. The dust blocks and absorbs the starlight which heats the dust, and the dust re-radiates in the far-infrared part of the spectrum. By the time the light reaches us, it is red-shifted to the sub-millimeter part of the spectrum. The wavelength we observed in is 1.1 mm, and is extremely sensitive to the amount of water vapor in the atmosphere. This means observations need to be taken from extremely dry regions of the Earth. We were using a Japanese telescope located in the Atacama Desert in Chile, called the Atacama Submillimeter Telescope Experiment (ASTE), which at the time was using a sub-millimeter detector called AzTEC (acronym translation available here) that was developed here at UMass. (This detector is now in Mexico being tested on the new Large Millimeter Telescope (LMT), which will be the biggest single dish sub-millimeter telescope on Earth!) To balance all the time I end up sitting in front of a computer, I went down to Chile for observing and general observatory maintenance at ASTE for a month!

Sub-millimeter astronomy is intriguing because very little is known about the galaxies that produce this kind of light because detectors and observatories in this wavelength regime are relatively new advances in technology. So there are many unanswered questions, which hopefully other new observatories such as the Atacama Large Millimeter Array (ALMA) in the Atacama desert and the LMT in Mexico will help us understand. Part of my goal in studying clustering of SMGs was to understand if they have an evolutionary connection to massive elliptical galaxies. But the unfortunate thing about current sub-millimeter observatories is that they have very low resolution imaging, which means you rarely get to see the shape and morphology of what you're looking at in detail. Perhaps that's one of the reasons that led me to Hubble Space Telescope (HST) and working with CANDELS data for the rest of my PhD thesis. I definitely love looking at the beautiful high-resolution images from HST that show the morphologies of galaxies!

Climbing Pigeon Spire, Bugaboo Provincial Park,
British Columbia

But, I might be a pretty boring scientist if science was all I was interested in. Getting the mind off work periodically is really important, not only because we are human, but also because it allows the brain some perspective for solving problems and can result in small epiphanies. What better way to gain a little different perspective than to climb hundreds of feet off the ground? Probably my biggest passion outside of science is climbing, which I started in college, and it has taken me to remote corners of the world. But even exploring the climbing here in New England has been a wonderful opportunity, because Amherst is centrally located between many world-class climbing areas (which are much more accessible than much of the climbing in Alaska!).

Where in the world I go from here is anyone's guess! I hope to graduate soon, and find a good postdoctoral position where i can continue exploring unanswered questions about high-redshift galaxies. Some serious hurdles currently face junior astronomers on the job market. The number of PhDs in astronomy exceeds the number of available permanent positions in astronomy, so competition for jobs is fierce and the prospects can be quite daunting. Astronomers typically do two postdoctoral positions before finding a permanent position. These are things I will face soon enough. But its important to keep in mind that often a big hurdle to success is lack of confidence in ones own abilities. Among all the science I've learned in graduate school, one of the most important lessons I've learned thus far is that believing in yourself is not just a cliche phrase, but has some serious truth in it. None of us arrived where we are by doubting ourselves. And this lesson I'll take with me no matter where I go.

Studying Galaxy Evolution with CANDELS Observations of High Redshift Clusters

Galaxy clusters and superclusters represent the largest bound structures in the universe. Because of this they are key probes of the large scale physics driving the expansion of the universe and the coalescence of matter which ultimately spawns the galaxies we observe. Studies of galaxy clusters also played an important role in the discovery of dark matter, the bullet cluster ultimately being a smoking gun. The topic of this blog post, however, is using galaxy clusters as laboratories for studying galaxy evolution.

By the early 1980's astronomers were finding strong evidence that the morphology of galaxies is correlated with their environment (the density of galaxies in the vicinity). Galaxies observed in high density environments are far more likely to be massive, elliptical, and passively evolving (little or no ongoing star formation), but the physical origins of this relationship are still not entirely known. It is likely related to interactions between the galaxies and the cluster and/or interactions between the cluster galaxies themselves. Galaxy-galaxy interactions include galaxy mergers which can significantly disrupt and randomize the orbits of their stars, growing and transforming the galaxy in the process.

Top: false colour image of the cluster combining
images from Subaru and the Spitzer Space telescope.
Bottom: Detection image for the cluster showing
the overdensity of galaxies at the redshift of the cluster.

Recently a technique has been developed for discovering galaxy clusters at high redshift which exploits the fact that as you observe galaxies at increasing distances certain spectral features redshift into redder photometric bands. Using data from large scale, deep photometric surveys it is possible to measure the density of galaxies which have similar colors measured in various bands. However, because these discoveries are based on photometry, the certainty you have that clusters you find are not just chance alignments of unassociated galaxies is relatively low (more on this unfortunate fact can be found in this previous post). Nevertheless, this technique was used by CANDELS team member Casey Papovich and collaborators to discover a galaxy cluster at a redshift of 1.62, meaning light we detect from these objects was emitted when the universe was less than a third of its current age!

Since its discovery, this cluster has been extensively studied. Spectroscopic observations were undertaken in order to confirm the narrow range in redshifts for potential cluster members. Despite difficulties in observing strong spectral features at this redshift (due to overlap with strong night sky emission), redshifts for 11 galaxies were obtained and found to lie close to the expected value for this cluster. X-ray observations were also taken using NASA's Chandra X-ray Observatory finding that the x-ray emission from this cluster is dominated by a single point source with weak extended emission. This, along with the spatial distribution of cluster members, supports a hypothesis in which this is a young cluster still in the process of forming (it has also been dubbed a "proto-cluster").

In 2011, I joined the team studying this cluster led by Casey Papovich, just as the first Hubble imaging from the CANDELS project was being released. The depth and resolution of these images finally made it possible to study the structures and morphologies of these galaxies. Because the large field of view of these images allows cluster and field populations to be examined in a single uniform data set, CANDELS imaging is ideal for studies of environmental trends. Using a code called GALFIT, I was able to measure sizes of these galaxies, as well as a parameter which measures how compact these galaxies are: the Sersic index. A general rule of thumb is disky galaxies have Sersic indices around 1 while spheroidal galaxies have Sersic indices between 2.5 and 4.

Hubble Space Telescope coverage from CANDELS. The cluster sits at the edge of a much larger field allowing us to construct a large comparison sample of non-cluster galaxies from a single uniform dataset.

In the first paper published using the results of these fits, we looked primarily at the sizes of passive galaxies. We first found that, at this redshift, in both the cluster and the field, passive galaxies have smaller sizes than similar galaxies in the local universe. This means that these galaxies must evolve in size over the subsequent ~10 billion years. Second, in the cluster there is a dearth of compact passive galaxies, possibly indicating that the galaxies within the cluster are experiencing a accelerated evolution compared with those in the field. This accelerated evolution could be related to an enhanced rate of mergers in the higher density environments. Evidence supporting this scenario was found in a study by Jennifer Lotz who shows that members of this cluster are far more likely to have very close neighboring galaxies than members of the field population (this study will be the topic of an upcoming blog post).

We extended the analysis of our first paper by folding in the structure of these galaxies using the Sersic index. If this cluster were to follow a trend similar to the morphology density relation we see locally, one would expect passive cluster galaxies to be more compact (and therefor have larger Sersic indices). When we looked at this, however, we found just the opposite! The passive galaxies in the cluster were found to have more disk like structures and larger sizes (when compared with other passive galaxies in our sample), attributes which are more characteristic of galaxies which are actively forming stars. The spatial distribution of these galaxies was also found to be very interesting: the structural differences we observe appear to be driven by passive galaxies in the outskirts of the cluster. Passive galaxies in the cluster core and the field appear to make up a more uniform sample of compact spheroids.

So what's going on here? One possibility is that these galaxies were actively star forming, disk galaxies which were recently accreted into the cluster environment. During the process of accretion, interactions with the cluster itself stopped their star formation. This could be caused by gas (which fuels star formation) being stripped from the galaxy or being heated and thus preventing it from collapsing into stars. This heating could possibly be caused by intergalactic shock waves that are propagated as a cluster is in the process of collapsing. These shock waves are formed when the galaxies being rapidly pulled inwards meet the ambient medium surrounding the cluster. While these shocks are largely theoretical at this point, there has been recent evidence in their favour in the nearby Coma cluster.

One caveat to this study is that there is a small number of cluster galaxies which are bright enough for us to observe and many of our distance measures have a large uncertainty. Because of this, our results are at the edge of statistical significance. However, similar results were found by Ruth Grützbauch and collaborators for a more highly evolved cluster at a slightly lower redshift. Their cluster, XMMU J2235.3-2557 at a redshift of 1.39, is found to have what they describe as a "quenching radius". Any galaxy within this distance from the cluster is extremely passive, but as you move further out the star formation rate increases. A picture is emerging in which galaxy clusters influence the star formation of galaxies at extremely large distances from the cluster core, much further out than previous estimates. To confirm this result will require more similar studies of high redshift clusters, so stay tuned!

What is an Observing Proposal?

Have you ever wondered what astronomers have to do to get to go on observing runs to telescopes? You might think that there are a lot of telescopes and thus astronomers can go observing whenever they want to or you might think that they observe every night. Actually, time on telescopes is in very high demand and astronomers have to compete with each other for every night of telescope time they get. In order to enter this process, astronomers must write a document called an observing proposal. This is the typical process for almost all telescopes, including big and small facilities, facilities run by a single University, those run by groups of Universities, and national facilities. Space-based telescopes (like Hubble!) also follow this procedure.

Image of Kitt Peak National Observatory, managed by NOAO. Image credit: Michael L. Weasner

In order to apply for telescope time, the first step is to come up with a good idea. Astronomers often have multiple projects going on at once and we are always thinking of new ideas and questions and ways to improve upon what we know. Once we have an idea we have to decide what telescope (and instrument) would be the most suited to accomplishing the science goals we have in mind. The instrument needed can often be more important than the telescope itself. Is the goal imaging or spectroscopy? Is there a particular wavelength range or filter needed? Do we need to target a single object or small patch of sky or are we surveying a large area? All of these factors go into selecting the best telescope-instrument pair. This selection can also depend on the University or country that the astronomer is at. Different Universities have access to different Observatories depending on funding, instrument development, and various other partnerships. There are also various facilities run on a national level -- for example NOAO in the US or ESO in Europe.

The next step is the bulk of the work: writing the actual proposal. A typical observing proposal has several components. The main one is called the Scientific Justification -- basically, describe why the science project you want to do is interesting. This is the place where an astronomer has to really sell their idea and convince others that answering this particular question is very important and must be done. Often, those reviewing proposals do not work in the particular specialty of the proposer so a good proposal is one that can be understood by any astronomer, not just experts on that particular topic. We must also clearly lay out the strategy of the science project here: how will the observations that we are proposing for answer this important question? What kind of data will be taken and how will this data be used to solve the problem presented? Often astronomers are limited to only a couple of pages of text so it can take a lot of work to say everything you want to say succinctly.

Another typical component of an observing proposal is a technical section. This is where an astronomer must go into detail about the instrument and telescope they are proposing to use and say why this particular combination is well suited. They must clearly demonstrate how much data they need and how much time this will take overall. Since observing time is a precious commodity, any time request must be clearly justified - if you say you will need two nights to accomplish your goals then you must show that two nights are really needed and one night would not be enough. Often, this portion is reviewed by people who are experts with the given instrument and who understand how well the instrument will perform.

It is important that a proposal be very well written! In fact, the ability to write well is a very important job skill for astronomers in general. An astronomer must be clear and concise in their proposal. If there is confusion about the goals or how they are going to address a particular problem, this could negatively impact the proposal's chance of success. It is also very useful to include informative graphics that illustrate the science goals and method presented. As they say, a picture is worth a thousand words, and a clear well-thought out figure can really strengthen a proposal. Finally, this might seem obvious but it is very important that all of the rules be followed! A proposal cannot be longer than the given limit and cannot be written in too small of a font or with tiny margins. This might seem picky, but when a person has to review a lot of proposals they all need to be easy to read. At last, the proposal is complete and can be submitted (and must be on time!). At this stage, the astronomer can sit back, relax, and start thinking about their next big idea.

But the process has only just started on the receiving end. Most observing proposals are then evaluated by a committee of peers (other astronomers, either from the specific institute that runs that particular telescope, or selected from all over). Every proposal is read by the entire committee and the committee gets together to discuss each one over the course of a few days. It's not always easy to pick out the best proposals to award time to. Often there are more excellent proposals than there are nights to be awarded. Intense discussions about the merits of each proposal results in a ranked list and time is given to those proposals at the very top. Every one that does not get their proposal accepted must try again next time.

This may sound like a lot of work, and it is, but whenever a proposal is accepted and we get the opportunity to observe and collect new data, it is all worth it. It is a great feeling to know that your peers have found your ideas worthy of supporting! This is a process that we go through once or twice a year for each telescope we would like to collect data with. One of the major observing seasons of the year just finished this past September. Luckily there is a little bit of a break before the next major deadline in February (for HST). This break is needed so that we can work on analyzing all of the data from the previous year!

The Bright Galaxies that HST Can Barely See

The large number of galaxies that are clearly detected in the deep HST/CANDELS images enable us to carry out very exciting studies that we regularly report in this blog. Today, instead we will focus on a special type of galaxies that are very faint in all the CANDELS images, but at least 40 times brighter at longer wavelengths, in the so-called mid-infrared regime. Until very recently, only a few isolated cases of these galaxies were known, but thanks to the depth of the CANDELS data, and making use of mid-infrared Spitzer Space Telescope images, we have discovered 25 such galaxies within a single CANDELS field.

Since its launch in 2003, the Spitzer Space Telescope has allowed us to study, in a systematic way, the infrared emission of galaxies at different cosmic times. With respect to previous infrared observatories, Spitzer represented a major step in infrared astronomy, which was possible thanks to the fast progress of infrared detector technology over the last three decades.

Four examples of sources that are bright in the Spitzer Space Telescope mid-infrared
images, but very faint in the HST/CANDELS images. The multi-wavelength analysis of these
sources indicates that they are very likely massive galaxies formed in the first two billion years
of cosmic time. Image credit: Caputi et al. (2012), ApJ, 750, L20.

At low redshifts, the Spitzer mid-infrared images trace the dust emission of star forming galaxies, which occurs after the dust is heated by the UV photons produced by the new stars. The UV photons that are the consequence of accretion of matter onto a galaxy's central black hole can have a similar effect, namely heating any surrounding dust and making it emit at mid-infrared wavelengths. But for high redshift galaxies, the mid-infrared emission seen in the Spitzer maps has a rather different origin: it directly traces the redshifted light of the galaxy oldest stars. The CANDELS images, in turn, show the redshifted light of a (high-redshift) galaxy's young stars.

By comparing the multi-wavelength emission of our 25 Spitzer-bright galaxies with theoretical galaxy spectral models, we determined that the vast majority of these sources are very likely at high redshifts (z>3), which means that we are observing the light that was emitted by these galaxies when the Universe was less than two billion years old. Actually, nowadays we know many z>3 galaxies, but the properties of our newly discovered 25 galaxies are very special: the fact that they are bright in the Spitzer images, but much fainter in the CANDELS maps, indicates that these objects should be among the oldest and most massive galaxies to be found at such early cosmic times.

Finding massive galaxies in the early Universe has important implications for galaxy formation theories, which need to explain how such objects could have formed so quickly and efficiently only a few billion years after the Big Bang. According to our most-accepted cosmological model, the Cold Dark Matter model, galaxies are embedded in dark matter halos, and grow with them hierarchically, through mergers, from small to larger units, through mergers. In such a scenario, one would expect that massive galaxies are the last to form. But different astronomical observations  conducted over the last decade indicate that the most massive galaxies that we see in the Universe today basically finished their growth when the Universe was less than a half of its present age (this is 8 billion years ago), while less massive galaxies continued forming a significant amount of stars later. This phenomenon is what extragalactic astronomers call 'galaxy downsizing.' So, searching for massive galaxies further back in time is very important for understanding when galaxies could assemble a large mass for the first time in the history of the Universe, and thus constraining galaxy formation models.

Another exciting aspect of our new, massive galaxy candidates at high redshifts is that they potentially constitute the 'tip of the iceberg' of a  larger galaxy population that still remains to be discovered. To fully understand the importance of such a galaxy population we will have to wait for the advent of the HST successor, the James Webb Space Telescope (JWST), which is due for launch in 2018.  The JWST will provide us with much deeper images than current telescopes, thanks to its large collecting area  -- seven times larger than the HST, and almost 60 times larger than Spitzer. With these deeper images,  we will be able to search for fainter analogues of our galaxies at higher redshifts. In the meanwhile, we are trying to follow up our galaxies at far infrared wavelengths with the Atacama Large Millimetre Array (ALMA), which is the only instrument that, currently, can independently confirm the nature of our sources.

Ultraviolet Observations in CANDELS

The CANDELS project is primarily focused on observations at optical and near-infrared wavelengths, that is light that is received at the wavelengths that the human eye can see and a little bit longer. However, a fortunate trick of the Hubble Space Telescope (HST) orbit also allows us to observe one of the five CANDELS fields at ultraviolet (UV) wavelengths, that is slightly shorter wavelengths than the blue end of the rainbow. UV observations are especially interesting, because very massive, very hot stars emit light strongly at those wavelengths.

How We Get UV Observations

Optical HST image of the Hubble Deep Field (in gray scale) with
Far-ultraviolet observations over layed (in purple). The Far-ultraviolet
are wavelengths even further in the blue direction than the UV
observations planned for CANDELS. Most galaxies are not detected
in the far-UV, despite the image being very deep, because their light is
shifted redward by the Doppler shift ("redshift"). Image Credit: Harry Teplitz

HST orbits the Earth, which means that most of the time when it wants to point at an interesting position on the sky, it can only do so during a fraction of the orbit during which the Earth isn't in the way. There are a few places on the sky, though, for which HST can point continuously during the entire orbit. The GOODS-North field in CANDELS is one of these special places. This means that we have about twice as much time available to observe that field than usual. CANDELS uses some of the extra time to observe GOODS-North in the UV.

As with all CANDELS fields, GOODS-North is observed many times, building up the signal through repeated observations. The schedule of the repetitions is designed for the supernova search. In practice, this means that even though we got the first UV observations in the spring of 2012, it will be spring of 2013 before we have enough data to see most of the UV objects in the images.

Major Science Goals

The most massive, young, and hot stars emit light strongly in the UV. This makes UV observations particularly effective for studying galaxies that are forming many stars.  These data allow us to find these galaxies and to study how they formed.  

In a little more detail, we have three major goals for the UV observations:

1.  Finding and studying strongly star-forming galaxies:  

The most popular way to find distant star-forming galaxies is to look for a strong feature in the distribution of their light across the spectrum.  This technique, referred to as looking for the "Lyman break" or "dropouts", will be familiar to people who have read about the rest of the goals for CANDELS, because it is also used to find galaxies in the very distant Universe. The advantage of adding UV data to CANDELS is that it allows us to use the same method to find galaxies when the Universe was about 25% of its current age. And, by using the same technique to find them, we can compare those galaxies directly to the much more distant ones.
 

In practice, the "dropout" technique means that we look for galaxies that are bright in most of the CANDELS bands, but are much fainter in the UV.  That is, they "drop out" of the UV. This is an indication that the UV light that they emit has been absorbed by neutral hydrogen before it reaches HST.

Once we find these galaxies, we can ask many intersting questions about them: Do they tend to be big or small? Do they tend to have a lot of dust? Do they tend to be in groups together or are they isolated?  

2.  The build up of galaxy structure from sub-galactic clumps

Galaxies grow and develop structure over time. Through gravitational effects, possibly including merging with other galaxies, they become like the galaxies we see today. We know that they undergo periods of intense star formation, but we are still learning about how they form their distinctive structures like spiral arms.  

There is evidence that as many galaxies grow, they form small clusters of hot stars, which are often called "clumps", which then migrate together into larger structures. Measuring the number, size, and brightness of these clumps can help us understand how galaxies form their structure. It can also help us distinguish which kinds of galaxies form through the mergers of smaller galaxies, and which kinds form primarily by themselves.  

Clumps have been studied extensively in galaxies when the Universe was about 25% of its current age. UV observations, which will see the hot stars that make up the clumps, will allow us to study them in galaxies later in the histroy of the Universe, when it was around 50% of its current age.

3.  How does ionizing radiation escape from galaxies?

There was a period of time in the relatively early Universe, known as the "dark ages", when most of the electrons and protons in the Universe were together in Hydrogen atoms. When galaxies began to form, they emitted a lot of energetic photons (referred to as "ionizing radiation") which broke apart the Hydrogen atoms, an event called "Reionization." Exactly how this happened is one of the great mysteries of cosmology. In particular, we don't know how the photons got out of the galaxies that contained the hot stars that emitted them.

 
In order to figure out how this ionization radiation escapes, we need to be able to observe it. The best way to do this is to look at galaxies that are similar to those which caused Reionization, but are a little closer so that we can study them. 

The CANDELS UV observations will allow us to make some of the best measurements ever taken of ionizing radiation escaping from galaxies.

The Next Steps

CANDELS is currently taking UV images, with a single pass of the GOODS-North field once every couple of months. It will take about a year to build up enough of these images to reach the sensitivity needed for the science goals described above.  We are eagerly waiting for next year, when we can begin to look at these exciting data and see what we can learn about strongly star-forming galaxies, the build up of galaxy structure, and how ionizing radiation escapes from galaxies. Stay tuned!

How Old Are Galaxies?

"How old are you?" A simple question, frequently asked of children, but rarely asked of an adult in polite company. When we ask the question, we are asking for the number of years that have elapsed since you were born. But how old are you, really? There were nine months of cellular development before birth, so for some purposes, maybe we should include that. Also, the average age of cells in your skin is less than two weeks, your stomach cells are typically less than 5 days old, and your blood cells are less than 4 months old. The average age of all the cells in your body is less than 10 years. So how old are you, really?

Okay, so for humans, if you want to get persnickety, maybe age isn't such a well-defined concept, but still, time since birth is generally a pretty useful definition.

When did galaxies form?

Two nearby galaxies. At the center is an elliptical galaxy,
which is basically a ball of old stars. A spiral galaxy, similar to
Milky Way,  appears in the upper right. Spiral galaxies are
still forming stars today. Do these galaxies have the same age?
This turns out to be a tricky question to answer.
Image from the Hubble Space Telescope.

When does the clock start for galaxies? When I was in graduate school, there was a concept of the epoch of galaxy formation, when big galaxies started forming. This idea was largely driven by the observation that nearby elliptical galaxies - and the central bulges of spiral galaxies - are filled with old stars that are not organized into thin, rotating disks.  It seemed likely that this was because galaxies formed their stars before the gas had a chance to settle into a thin disk.

If galaxies formed fast, then they should have been very bright when they were forming all those stars. So bright that once telescopes were equipped with modern CCD detectors in the late 1970's, it should have been possible to see them at large distances. No one could find them. Astronomers found lots of little faint blue galaxies, but these were smaller galaxies, that were much closer to us than expected.

At about the same time that observers weren't finding the epoch of galaxy formation, theorists were developing the idea that galaxies formed hierarchically, centered on the densest patches of dark matter in the early universe.  These early galaxies started forming their first stars when they had acquired only a tiny fraction of their present day material. They subsequently grew larger as more gas fell in at later times, and when they merged with other galaxies. This hierarchical theory simultaneously explains why we couldn't find the epoch of galaxy formation and why galaxies cluster together on large scales.

The implication is that galaxies started out with very few stars, and may have taken a long time to reach their peak star-formation rate. That's indeed what we infer today when we estimate the average number of stars formed per unit volume in the universe. This cosmic star-formation rate peaked when the universe was about 3.5 billion years old, even though we have now found galaxies that existed when the universe was less than a billion years old.

Inferring galaxy ages from colors

Even though the hierarchical models have been several decades and we have known for more than a decade that the globally-averaged cosmic star-formation rate started out much lower than the peak rate, it has taken a while for astronomers to realize that when they try to infer the age of an individual galaxy from its colors, the traditional assumptions are probabably incorrect. The traditional assumption has been that the star-formation starts off high and drops off exponentially. This was motivated by the idea that a galaxy starts with a fixed reservoir of gas and its star formation must keep dropping as that gas is used up. These are known as "tau models" in the jargon (not this kind of tau model!), because the Greek letter tau is usually used to designate the timescale for star-formation to reach roughly half of its initial peak.  Recent studies have shown that analyzing galaxy colors using tau models can give very misleading estimates of star-formation rates and ages, and that adopting different models can improve the estimates. See, for example, papers by my fellow bloggers Janine Pforr, Stijn Wuyts and my former student Joshua Lee.

Star-formation histories of galaxies from a hierarchical model (blue), compared to the best-fit tau models that were inferred from the galaxies' colors. You can see that the tau models are not at all representative of the true star-formation histories. But the problem is that the blue curves are models too. For real galaxies, we don't know what the true answer is. What we need is simple models that work well for a wider variety of possible star-forming histories than the tau models. From Lee et al. 2009.

 

A new definition of age?

 

Back when it seemed reasonable to use tau models, it seemed reasonable to designate the galaxy age as the elapsed time since it formed its first stars. This would be equivalent to galaxy birth. But now that we know that galaxies formed slowly, it's much less obvious what we should use as an age. On the one hand, the "chemical evolution" of a galaxy - the build-up of the heavy elements created in stars - is highly influenced by the first few stars. Once the gas in a galaxy contains heavy elements, it's ability to cool to form stars is dramatically altered. So we are interested in when galaxies formed their first stars. But on the other hand, most of the stars in a typical galaxy didn't form until billions of years later.

When using galaxy colors to estimate ages, the simplest approach is to estimate when the galaxy had formed half of its present stellar mass. This is probably the most reliable estimate of age that we can make if we have nothing but the galaxies' colors to work with, and no particular preconceptions about their star-formation histories.  But this is like starting the clock when the galaxy was a teenager. It's also a peculiar definition in the sense that a galaxy's age will not increase linearly with time. In fact, under this definition, if a galaxy doubles its stellar mass every 100 million years, then it will always be 100 million years old. If it then suddenly doubles its mass in only 10 million years, it will become only 10 million years old. This can be a bit confusing.

Another possibility is to adopt a different set of models that might be more representative of the true star-formation histories of galaxies than than the tau models. We can then define age as the time since t=0 in those models, as was done for the tau models. Unfortunately, there is currently no consensus on what to use as an alternative model. This is a topic of debate and discussion at galaxy evolution conferences. Maybe this will settle out and we will all agree on how to set our clocks. Or maybe the consensus will end up being that age is not a useful concept for galaxies. It's too soon to tell.

So the question "how old is this galaxy?" turns out to be more subtle than we might have thought. Scientific progress is a continual process of unveiling our ignorance. Each advance in our understanding leaves us still ignorant, but ignorant at a much deeper level.