Deep-field image from Hubble triggers competing reports.
Astronomers have caught a glimpse of galaxies that existed a mere 500 million years after the Big Bang, more than 13 billion years ago.
In August, NASA's Hubble Space Telescope made the deepest image ever of the Universe in near-infrared wavelengths using its new set of eyes, the Wide Field Camera 3 (WFC3) installed by astronauts in May. In the near infrared astronomers can detect galaxies that are so distant, and receding so quickly, that their light is stretched longer — or redder — than visible light. The more distant an object, the more its light is shifted red and the higher its 'redshift'.
For the past few months, researchers have been poring over this new data set — a sliver of sky about one-twelfth the diameter of the full Moon, viewed for 173,000 seconds over four days — searching for ancient galaxies that might deepen understanding of how the Universe evolved. The current record-holder for distance is a gamma ray burst, discovered in April, with a redshift of 8.2.
Now, astronomer Garth Illingworth at the University of California, Santa Cruz, and his colleagues have found tentative evidence of three galaxies with redshifts of around 10. These would have existed when the Universe was just 3-4% of its current age, and would be among the oldest objects ever seen. The findings have been posted on the preprint server arXiv.org1.
"Even though it's not really unexpected, finding galaxies at such early times is hugely exciting," says Illingworth, who also helped to create the publicly available Hubble image and data set. "There's no smoking gun, but we're confident that this is what we're really seeing."
Older than thou
The team's report of galaxies with a redshift of 10 isn't the first. For instance, Rogier Windhorst at Arizona State University in Tempe and his colleagues have reported finding 20 galaxies with a redshift near 10 in the same data set, although this 20 did not include Illingworth's three. Windhorst and his colleagues posted their findings on arXiv.org in October2.
The various teams disagree over how exactly to define these distant galaxies, and what constitutes a definitive detection. Illingworth argues that Windhorst's standards were not stringent enough, and that some of the galaxies the other team detected were close to large, bright galaxies that could have contaminated the results. "That could have confused their software and caused them to get a lot of objects that just were not real high-redshift galaxies," he says.
Illingworth also notes that 20 galaxies of redshift 10 in such a small region of space indicates a higher rate of star formation than has been predicted for that time. "It doesn't accord with what we would expect theoretically or logically," he says.
Windhorst acknowledges that his group "may have overshot", but adds that Illingworth's group was "overly conservative". Even so, Windhorst noted that two of the galaxies detected by Illingworth's team did not meet his team's criteria as high-redshift, and the third was not detected after the data were processed. "Being conservative is good, but I think they haven't reduced the data as good as they could have," Windhorst says.
Let there be light
But Illingworth's team "appears to have done a more careful job than Windhorst's", says astronomer Richard Ellis at the California Institute of Technology in Pasadena, who in 2007 reported candidate galaxies with redshifts between 8 and 10 using the Keck telescope in Hawaii. He added that Illingworth's team "goes to great lengths to justify their candidates", although he is cautious because the camera detects high-redshift objects only in one filter, and because many such claims, including his, have been revisited by other groups with mixed results.
"The history of claims for finding redshift-10 galaxies is a chequered one, but it's exciting," he says. "Ultimately, we will still have to verify these claims. It's going to be very hard, as it has turned out to be with [other] candidates."
The lack of many bright galaxies at redshift 10 offers clues to what kicked off the Universe's "reionization epoch" — a period between 500 million and 1 billion years after the Big Bang during which luminous objects such as galaxies and quasars ionized the intergalactic medium. The lack of bright galaxies at the start of this timespan suggests that they did not initiate the reionization process, Windhorst says.
"There are still a lot of questions to address," says Illingworth. "This is very tantalizing, but we need to understand the properties of those galaxies. That's where the real scientific interest is."
Bouwens, R. J. et al. http://arxiv.org/abs/0912.4263 (2009).
Yan, H. et al. http://arxiv.org/abs/0910.0077 (2009).