A Deep View Down Broadway

Abell 2744 Parallel Deep Field from the Hubble Frontier Fields Project Credit: NASA, ESA, and J. Lotz, M. Mountain, A. Koekemoer, and the HFF Team (STScI)

Abell 2744 Parallel Deep Field from the Hubble Frontier Fields Project
Credit: NASA, ESA, and J. Lotz, M. Mountain, A. Koekemoer, and the HFF Team (STScI)

[Note: this blog post also appears on the Hubble’s Universe Unfiltered blog.]

One of the more philosophical concepts that astronomers have to deal with on an everyday basis is the commingling of space and time in astronomical images.

The underlying idea is straightforward. The speed of light is finite. Light from a star or nebula or galaxy takes a measurable amount of time to cross the space between it and us. Hence, the light we see now left that object at some previous time. We view astronomical objects as they were in the past. As I like to say, looking out in space is also looking back in time.

The implications of this maxim are considerable, especially in dealing with the deep field images from Hubble (see the accompanying image of the Abell 2744 Parallel deep field). Such images contain a wonderful assortment of galaxies, with a few stars here and there. Each object is at a different position in space, both in the two-dimensional sense of a different position within the image and in the three-dimensional sense of being at a different distance from Earth. Further, objects at different distances are seen at different times in the past. Hence, astronomers must examine these deep field images in four-dimensional space-time.

Tackling the expanse of space and time in these images can be mind-boggling. We’ll start with the stars, which are easier to understand. All the stars are local, within our Milky Way galaxy. These stars are generally hundreds to thousands of light-years away. The light we observe today might have left the star while the pyramids of Egypt were being built. Because stars don’t change appreciably on scales of thousands of years, stars in deep fields are just like stars in other astronomical images.

The galaxies, however, stretch much farther into space. The nearest are many millions of light-years away, while the most distant are around ten billion light-years away. Galaxies don’t change much on million-year timescales. For example, it takes over 200 million years for our Sun to orbit once within our galaxy. Even though the light may have left a galaxy when dinosaurs first started to dominate our planet, the same galaxy would look similar today. Thus, the nearby galaxies in these images are comparable to local galaxies.

Given billions of years, however, galaxies do change, and these deep field images provide compelling evidence. Distant galaxies do not have the standard spiral and elliptical shapes. They are often elongated, have bright spots of star formation, and are much smaller in size. We see galaxies as they were before the Sun, Earth, and the solar system formed. We study the development of galaxies over time to see how they form and grow. The perplexing point is that, for any given galaxy in the image, there is no distinct visual indicator of its distance in space or time. The layers of the universe are jumbled together across the image, and it is a grand puzzle of cosmology to sort them out.

The usual method to determine distances, and therefore times, is to measure the cosmological redshift of each galaxy. That concept has been discussed in a Frontier Fields blog post by Dr. Brandon Lawton: “Light Detectives: Using Color to Estimate Distance”. Thus, I’d like to take this essay in a different direction.

The Manhattan Deep Field

When discussing the cosmic mixture of space-time with an artist visiting from Spain, I happened upon a novel idea for a human-centric analogy.

Imagine you are in New York City, specifically Times Square in Manhattan. You look down Broadway to the southern end of the island about 4 miles away. If the speed of light were extremely slow, traveling only one mile per century, what would you see?

Each mile down Broadway would represent one hundred years of New York’s history. Each block would be 5 to 10 years earlier in the development of the metropolis.

A quarter of a mile away, the southern end of the theater district would appear as it did in the early 1990s when “Miss Saigon” came to Broadway. Only a few blocks farther would be the disco era and the civil unrest of the 1960s, then the World War II years and the Great Depression.

The Empire State Building, about a mile away, would vanish, as it was not built until 1931. At a similar distance, Madison Square Garden would be seen hosting heavyweight boxing matches in its original building, before the demolition and re-construction in the late 1920s.

Progressing another mile down Broadway to Union Square would travel back past the Civil War, Tammany Hall politics, economic growth fostered by the Erie Canal, and Alexander Hamilton’s original run on the New York stage.

The mile beyond to the SoHo district progresses through the times of New York as the capital of the United States, the Revolutionary War, the founding of Columbia University, and the importation of slaves by the Dutch West Indies Company.

The final mile to Battery Park leads through the colonial era alternately dominated by Dutch or English foreign powers, past the garrison of Fort Amsterdam, to the island’s Native American roots and the initial explorations by Henry Hudson.

A “slow speed of light” view from Times Square would lay out the entire history of the city of New York in a single view. The commingling of space and time would make it the historian’s exceptional equivalent of the astronomer’s standard observation: a deep view down Broadway.

This idea of a time-warped view of New York provides an analogy to what Hubble uncovers: the history of galaxies compressed and jumbled within each deep field. Perhaps it can help you to look at these images from that requisite four-dimensional perspective. These deep field images are truly a trip down memory lane.

Galaxy Shapes in the Frontier Fields Observations

We can learn a lot about galaxies by analyzing their light, through computer modeling, and using other complex techniques. But at the most basic level, we can learn about galaxies by studying their shapes.

Galaxy appearance immediately reveals certain characteristics. Elliptical galaxies contain a wealth of old stars. Spiral galaxies are full of gas and dust. Distorted galaxies have likely experienced a gravitational interaction with another galaxy that warped their structure.

The Mice, as these colliding galaxies are called, are a pair of spiral galaxies seen about 160 million years after their closest encounter. Gravity has drawn stars and gas out of the galaxies into long tails.  Credit: NASA, H. Ford (JHU), G. Illingworth (UCSC/LO), M.Clampin (STScI), G. Hartig (STScI), the ACS Science Team, and ESA

The Mice, as these distorted colliding galaxies are called, are a pair of spiral galaxies seen about 160 million years after their closest encounter. Gravity has drawn stars and gas out of the galaxies into long tails. Credit: NASA, H. Ford (JHU), G. Illingworth (UCSC/LO), M.Clampin (STScI), G. Hartig (STScI), the ACS Science Team, and ESA

The Frontier Fields project adds another dimension to this simple analysis. When we look at extremely distant galaxies with the magnification of gravitational lensing, we see new detail that was previously obscured by distance. Their shapes are clues to what occurred within those galaxies when they were very young.

Galaxies viewed through the gravitational lenses of the Frontier Fields clusters can be seen at a resolution 10 times greater than non-lensed galaxies. That means those tiny red dots that so thrill astronomers in normal Hubble images actually have some structure in Frontier Fields imagery.

Previous studies, such as the Hubble Ultra Deep Field, The Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey, or even adaptive optics-enhanced studies by ground telescopes have shown that young, star-forming galaxies at about a redshift of 2 (existing when the universe was about 3.3 billion years old) appear to have a certain lumpiness. But without gravitational lensing, we lack the resolution to say for sure whether those lumps were massive clusters of newly forming stars, or whether some other factor was causing those galaxies to have a clumpy appearance.

Frontier Fields has revealed that yes, many of those galaxies have star-forming knots that really are quite large, implying that star formation occurred in a very different way in the early universe, perhaps involving greater quantities of gas in those young galaxies than previously expected.

Frontier Fields has also given us a better grasp of the physical size of gravitationally lensed young galaxies even farther away, at a redshift of 9 (when the universe was around 500 million years old). Observations show that these galaxies are actually quite small – perhaps 200 parsecs across, while a typical galaxy you see today is closer to 10,000 parsecs across. These observations help plan future observations with the Webb Space Telescope, picking out what will hopefully be the best targets for study.

This composite image shows examples of galaxies similar to our Milky Way at various stages of construction over a time span of 11 billion years. The galaxies are arranged according to time. Those on the left reside nearby; those at far right existed when the cosmos was about 2 billion years old. The Frontier Fields project is collecting galaxies from the earliest epochs of the universe to add to such comparisons. Credit: NASA, ESA, P. van Dokkum (Yale University), S. Patel (Leiden University), and the 3D-HST Team

This composite image shows examples of galaxies similar to our Milky Way at various stages of construction over a time span of 11 billion years. The galaxies are arranged according to time. Those on the left reside nearby; those at far right existed when the cosmos was about 2 billion years old. The Frontier Fields project is collecting galaxies from the earliest epochs of the universe to add to such comparisons. Credit: NASA, ESA, P. van Dokkum (Yale University), S. Patel (Leiden University), and the 3D-HST Team

Galaxy shape also plays a role in discoveries in the Frontier Fields’ six parallel fields, which are unaffected by gravitational lensing but provide a view into space almost as deep as Hubble’s famous Ultra Deep Field, with three times the area.

It’s well known that galaxies collide and interact, drawn to one another by gravity. Most galaxies in the universe are thought to have gone through the merger process in the early universe, but the importance of this process is an open question. The transitional period during which galaxies are interacting and merging is relatively short, making it difficult to capture. A distant galaxy may appear clumpy and distorted, but is its appearance due to a merger – or is it just a clumpy galaxy?

Collision-related features — such as tails of stars and gas drawn out into space by gravity, or shells around elliptical galaxies that occur when stars get locked into certain orbits – are excellent indicators of merging galaxies but are hard to detect in distant galaxies with ordinary observations. Frontier Fields’ parallel fields are providing astronomers with a collection of faraway galaxies with these collision-related features, allowing astronomers to learn more about how these mergers affected the galaxies we see today.

As time goes on and the cluster and parallel Frontier Fields are explored in depth by astronomers, we expect to to learn much more about how galaxy evolution and galaxy shapes intertwine. New results are on the way.

MACS J0416 Data is Complete

Observations of another Frontier Fields galaxy cluster and parallel field are complete. This time, we have new images for you of MACS J0416.1-2403. Here’s the galaxy cluster:


And here is the parallel field:


Beautiful, aren’t they? This is the second Frontier Fields cluster and parallel field to be fully imaged. You can see the first here.

Remember that to maximize scientific discovery, Hubble is using two of its instruments simultaneously to examine both the cluster and the parallel field, then observing the same areas again with the instruments switched.

Hubble takes two sets of observations, called epochs, in order to thoroughly examine the two areas. During the first, Hubble spent 80 orbits with the Advanced Camera for Surveys (ACS) pointing at the main galaxy cluster, and Wide Field Camera 3 (WFC3) looking at the parallel field. ACS provides a visible-light view, and WFC3 adds near-infrared light.

During the second epoch, Hubble spent 70 orbits targeting WFC3 on the main cluster and ACS on the parallel field.

Scientists are poring over the new data, and one result is already in. Expect to hear more about these observations in the near future.

Frontier Fields Q&A: Redshift and Looking Back in Time

Q: What do you mean when you say you’re “seeing some of the earliest galaxies in the universe?” How does looking into deep space allow you to look back in time?

The simple answer is that light travels and the universe is huge. Light travels very fast – 186,000 miles (300,000 km) per second, but it still has to move across the vast distances of space. Remember that for us to see anything – from the flash of a camera to the glow of a really distant galaxy, we have to wait for its light to strike our eyes.

That camera flash shows in our vision instantaneously because it doesn’t have far to go. But distances in the cosmos are so vast that it takes light a long time to reach us. The light from our closest companion, the Moon, takes about 1.3 seconds to cross the 239,000 miles (390,000 km) between us. So when you look up at the sky, you don’t see the Moon as it currently is. You see it as it appeared 1.3 seconds ago.

This is so 1.3 seconds ago. Credit: Luc Viatour, Wikimedia Commons

This is so 1.3 seconds ago.
Credit: Luc Viatour, Wikimedia Commons

The greater the distances, the greater the time difference. Light from the Sun needs about 500 seconds, or about eight minutes, to reach us from 93,200 miles (150 million km) away. Light from Neptune needs about four hours to cross the solar system.

We refer to these distances by the time it takes light to cross them. So Neptune is four light-hours away, and the Sun is 500 light-seconds away. Light from the next nearest star, however, needs four years to reach us across space. We say that star is four light-years away. The light we see from that star in today’s sky is also four years old. For galaxies, we’re talking millions to billions of light years. So we see the farthest galaxies as they appeared in the early universe, because the light that left them way back then is finally reaching us just now.

Q: What does it mean when you talk about a galaxy’s redshift?

When we’re discussing the Frontier Fields project, we’re talking about something more precisely called “cosmological redshift.” The space light is traveling through is expanding. That means that the light wave gets stretched as it travels, like a spring being pulled into a different shape. This stretching shifts light into longer wavelengths.

Since red light has a longer wavelength than blue light, the light is said to be "red-shifted." Credit: NASA

Since red light has a longer wavelength than blue light, the light is said to be “redshifted.” Credit: NASA

The farthest galaxies in the universe would have originally emitted visible and ultraviolet light, but since that light has been stretched as it travels, those galaxies appear to us instead in the form of infrared light. Cosmological redshift refers to that change and the measure of that change.

Q: Why do we hear the Frontier Fields galaxies described in terms of redshift and light-years? Which is right?

They tell us different things. Light-years are a measurement of distance defined by the time it takes light to travel in a year. But distance is notoriously difficult to measure in astronomy.

Cosmological redshift is a direct measurement of the expansion of space. Astronomers describe galaxies in terms of their redshift because unlike distance, it’s a clear and definite value that’s relatively easy to measure without many errors.

Astronomers have different models of how the universe works, and they can plug the redshift into those models to get the distance to a galaxy – but the distance will differ depending on which model of the universe they use. The variations in those models include things like the shape of the universe, the rate at which it’s expanding, the amount of normal matter it contains, etc.

Astronomy is about figuring out how the universe works and narrowing down all those models to the best one, and we still have a long way to go. Projects like Frontier Fields will help us rule out those models that don’t fit the incoming data.

Q: Everywhere we look with the Frontier Fields project, galaxies appear to be moving away from us. Does this mean we’re in the center of the universe?

No. It’s evidence that space is expanding. The easiest way to visualize this is to imagine a balloon. If you cover the balloon with dots, and then inflate it, no matter which dot you pick to represent your position, all the other dots will appear to be moving away from it as the balloon expands. Imagine this happening in three dimensions instead of on a flat surface, and you can understand why it looks like other galaxies are rushing away.

Q: So space is expanding and the light from the earliest galaxies has traveled over 13 billion years to reach us. If space is expanding, are those galaxies even farther away now?

Yes. For nearby galaxies, the expansion doesn’t make much of a difference. But for galaxies extremely far away, the distance is significant. That’s because the farther away an object is, the more space there is between us and the object. That in turn means there’s more space to undergo expansion, so the objects appear to be moving away from us much faster. Light from the earliest galaxies may have traveled 13 billion years to reach us, but those galaxies could be around 45 billion light-years distant by now.

Q: Does this mean the galaxies are moving faster than the speed of light?

No. No object can travel through space faster than the speed of light. But the expansion of space itself is not so constrained – in fact, theories of the beginning of the universe visualize the initial expansion of the Big Bang happening with unthinkable speed. But because the speed of light is only so fast, there are galaxies in the distance whose light we cannot yet see. We call this the edge of the visible universe.

Q: What’s out there, past the edge?

Space dragons! Ok, probably not. Credit: Uranometria

DRAGONS! SPACE DRAGONS! GIANT, COSMIC FIRE-BREATHING SPACE DRA– Ok, fine, probably not. Credit: Uranometria, Wikimedia Commons

We expect more of the same, though this is still an open question that astronomers are researching and theorizing about. We’ve found we tend to see the same distribution of galaxies no matter which direction we look in the universe. If we were somehow transported to a galaxy on what, for Earth, is the edge of the visible universe, the border of the visible universe would move, but the universe would neither change nor look very different to us.

Q: Do you have a question about the Frontier Fields project?

Leave it in comments, and we’ll see if we can answer it.

Cosmic Archeology

Today’s guest post is by Dr. Mario Livio, Hubble astrophysicist and author of the blog “A Curious Mind.” A version of this post appeared previously on Dr. Livio’s blog.

During the Christmas season of 1995, the Hubble Space Telescope was pointed for 10 consecutive days at an area in the sky not larger than the one you would see through a drinking straw. The region of sky, in the Ursa Major constellation, was selected so as to be as “boring” as possible — empty of stars in both our own Milky Way galaxy and in relatively nearby galaxies. The idea was for Hubble to take as deep an image of the distant universe as possible. The resulting image was astounding.  With very few exceptions, every point of light in this image is an entire galaxy, with something like 100 billion stars like the Sun.


The original Hubble Deep Field image.

Detailed analysis revealed that the very remote galaxies were physically smaller in size than today’s galaxies, and that their morphologies were more disturbed. Unlike the grand-design spirals or smooth elliptical shapes that we see in relatively close galaxies, the distant objects look like train wrecks. Both of these observations fit nicely into the idea that galaxies evolve largely by “mergers and acquisitions.” Small building blocks merge together to form larger ones, or cold flows of dark matter along dense filaments fuel the growth. What we see in the distant past are those interacting—and hence smaller and less regular in shape—building blocks.

Since then, Hubble observed even deeper, producing the “Hubble Ultra Deep Field” in 2004, and then in 2009 an image that included infrared observations taken with the new Wide Field Camera 3. This observation allowed astronomers to glimpse the universe at its infancy, when it was less than 500 million years old (the universe today is 13.8 billion years old). The Deep Field observations have also enabled researchers to reconstruct the history of global cosmic star formation. We now know that about 8 billion years ago the universe reached its peak in terms of the new star-birth rate, and that rate has been declining ever since — our universe is past its prime.


This tiny object in the Hubble Ultra Deep Field is a compact galaxy of blue stars that existed 480 million years after the Big Bang. Its light traveled 13.2 billion years to reach Hubble.

The Chandra X-ray Observatory has created its own Deep Field observations, discovering hundreds of low-luminosity active galactic nuclei, where disks feed mass onto central black holes, and emit copious x-ray radiation.

Infrared observations with the Spitzer Space Telescope completed the picture of the deepest images of the cosmos to date. Together, Hubble, Chandra, and Spitzer have created a detailed tapestry of a dynamic, evolving universe in which some two hundred billion galaxies are within the reach of our present telescopes.

Currently, Hubble is engaged in observing six new deep fields, each one centering on a galaxy cluster whose gravity can deflect, multiply, and magnify the light from more distant objects (the effect is known as “gravitational lensing”). In parallel, Hubble will also observe six deep “blank” fields. The goal is to use those so-called “Frontier Fields” to reveal populations of fainter galaxies, and to characterize the morphologies of distant star-forming galaxies.

The first of these super-deep views of the universe has already revealed almost 3000 of previously unseen, distant galaxies.

To see the very first galaxies that formed in our universe, we will have to await the James Webb Space Telescope, on schedule for launch in 2018. From a cosmic perspective, new discoveries are just around the corner!