Picture the universe in the hundreds of millions of years after the Big Bang – a pitch-black veil of gases spread evenly through the cosmos. The Dark Ages of space-time. Over the course of thousands of millennia, minute density fluctuations, magnified by time and gravity, caused clumps of helium and hydrogen — the only elements in the pre-stellar universe —to cool and congeal inside a cosmic web of intersecting filaments. These primordial objects coalesced into massive, short-lived stars a hundred times the size of the sun that exploded as supernovae or imploded to form massive black holes, ejecting the materials that ultimately formed stars like the sun and planets like Earth.

These events happened so long ago that not the faintest trace is visible with today’s telescopes. At the final frontier of observational cosmology, there is little direct proof to support his theory of structure formation, admits Dr. Volker Bromm, assistant professor of Astronomy at The University of Texas at Austin. However, Bromm asserts that this scenario is most likely how the early universe evolved, and much of the scientific community agrees. With the launch of the successor to NASA’s Hubble Space Telescope in 2013, Bromm will have a chance to see his predictions proven right…or wrong.

In the absence of powerful enough telescopes, Bromm and other astronomers rely on supercomputers to peer deep into the past. “Astronomy at this point wouldn’t exist without computation,” Bromm said. “The advance of supercomputers means you can ask more ambitious questions and put more physics into your models. It’s all driven by supercomputers.”

Using the supercomputing resources at the Texas Advanced Computing Center (TACC), Bromm and his team, including graduate students Thomas Greif and Jarrett Johnson, have theoretically modeled aspects of the first billion years of the 14 billion-year-old universe, computing the interactions of molecules on the largest and smallest scales over millions of years and changing atmospheric conditions. Bromm’s conclusions closely match those of other cosmologists (scientists who study the origin and development of the universe), whose computational simulations also predict giant, bright, unstable stars that played a pivotal role in the enrichment of the universe.

“When I started working on this problem almost ten years ago, few people in the world were studying the early universe because there was no data to constrain the theory,” Bromm recalls. “But there was excitement that there was this important question that was completely unanswered, what we called the ‘End of the Cosmic Dark Ages’.

“We knew there was this time in cosmic history when something dramatic happened, when for the first time gravity had accumulated the cosmic matter to make the first gravitationally bound objects that could become stars,” Bromm continued. “And we also knew that these stars would be vitally important because they would completely change conditions in the universe for subsequent history. It was a crucially important question and we had no idea what the properties of those first stars were.”

Bromm combined the physical characteristics of the original gases with the initial conditions of the universe in a box of space large enough to generalize the universe, and implemented a numerical method to simulate the evolution of the first stars. The data that emerged from his simulations at TACC has led to a picture of the early universe evolving from darkness and simplicity to light and chaos. “Massive supercomputers such as Lonestar enable researchers like Volker to simulate the world and universe around us, including acting as time machines to understand even the distant past and to model the future,” said TACC Director, Jay Boisseau.

According to Bromm’s simulations, the cooling of the universe caused gases to contract, separating them from the dark matter and enabling them form high-density clumps inside the cosmic network of intersecting dark matter filaments. At the nodes of these networks, haloes formed, and within the haloes the denser regions of gas collapsed to form the first stars. The first stars lived only a few million years but emitted the ultraviolet radiation that ionized the universe. Their demise gave rise to the materials and the evolving characteristics that set the universe down the road toward complexity. Though Bromm’s model is primarily theoretical, it has been validated indirectly through stellar archeology and helps explain observable phenomenon like gamma ray bursts and cosmic microwave background radiation.

Bromm’s initial discoveries laid the groundwork for the history of the first billion years of the universe. Now, he is studying further aspects of cosmological evolution, such as how the massive first stars assembled into the first galaxies, and how new elements were produced from primordial hydrogen and helium?

“That first discovery was really abstract physics,” Bromm said. “The next step in this great game of understanding the first structures is going from the first stars to the first galaxies, which occurred a little further on in time. The first galaxies are more complex because the stars engage in cross-talk to each other – one star forms, dies and explodes and it influences the formation of other stars. There’s a non-linear feedback loop that is impossible to understand without visualization, so we have found it useful to elicit the message of how the first stars formed by doing visualization work with TACC.”

Working with Paul Navratil and Karla Vega of the TACC scientific visualization group, Greif, Johnson and Bromm created dynamic images and animations that trace the transition from an initially homogeneous state to one of increasing complexity and capture the interactions of the first stars with the surrounding matter.

"Datasets like these are the most exciting to visualize,” said Navratil. “We’re not only contributing to scientific understanding, we’re visualizing phenomena that have never been seen before."

With the planned launch of NASA’s James Webb Space Telescope (JWST) set for 2013, Bromm and his colleagues will learn more about the nature of the early universe. They may even obtain an observational answer to what has been, to date, a theoretical problem — JWST’s primary science goal is to find light from the first objects that coalesced after the Big Bang. However, preliminary studies suggest the space telescope still won’t have the sensitivity to see isolated early stars, so Bromm is working on a series of studies to help guide the JWST’s detection systems in their search for small, galaxy-sized aggregates. If the process works, Bromm will be able to compare observational data with his computational models to refine his theory of the first stars. “It’s exciting to make these theoretical predictions, but it’s nice to know that at some point in the not-too-distant future, your work will be tested,” Bromm said.

With each advance in the interpretive and processing power of telescopes and supercomputers, cosmologists’ ability to explore how light, life, and complexity evolved in the universe improves.

“For the first time in human history, we have the chance to answer the big questions. How old is the universe? Where did we come from? Where are we going? What are the basic constituents of matter?” Bromm concluded. “And now we see that things are falling into place. This is a golden age of discovery with a lot of opportunities for exciting breakthroughs.”

To read more about Dr. Bromm’s simulations of the first stars and to see visualizations of stellar evolution, visit: http://www.tacc.utexas.edu/~pnav/firststars/.

by Aaron Dubrow
Science and Technology Writer
Texas Advanced Computing Center